CAMxUsersGuide_v6.30x CAMx Users Guide V6 30

User Manual:

Open the PDF directly: View PDF .
Page Count: 286

Download
Open PDF In Browser	View PDF

March 201
16

CAMx User’s Guide Versio
on 6.3

Useer’s Guide
COMPREHHENSIVE AIR QUALITY MO
ODEL
WITH EXTENSIO
ONS
Version
n 6.3

Raamboll Envviron
7733 San Marin Drive, Suite 2115
Novato, California, 9
94998
www.ramboll‐environ
n.com
www.camxx.com
P‐415‐899‐‐0700
F‐415‐899‐‐0707

March 2016

CAMx User’s Guide Version 6.3

Copyright: Ramboll Environ
1997 – 2016

This publication may be reproduced for
non‐commercial purposes with appropriate attribution.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

www.camx.com

March 2016

CAMx User’s Guide Version 6.3

ACKNOWLEDGMENTS
Ramboll Environ acknowledges the following groups for their contributions to the development
of CAMx:


The Texas Commission on Environmental Quality (TCEQ), for sponsoring the development,
testing, and review of numerous components of the model;



The Lake Michigan Air Directors Consortium (LADCo), for sponsoring the development,
testing and review of numerous components of the model;



The U.S. Environmental Protection Agency (EPA), for sponsoring the development, testing,
and review of numerous components of the model, and for co‐sponsoring the
development and testing of the MPI parallel processing capability. Special thanks to Dr.
Jon Pleim for assistance with implementation of ACM2;



The Coordinating Research Council (CRC), for sponsoring the development, testing, and
review of numerous components of the model;



The Texas Air Quality Research Program (AQRP), for sponsoring the development, testing,
and review of numerous components of the model;



Dr. Sasha Madronich (NCAR) for development of the TUV radiative transfer model and
assistance with incorporating the in‐line TUV treatment into CAMx;



The Carnegie‐Mellon University, Department of Chemical Engineering, for providing full‐
science PM algorithms, assistance in incorporating them into CAMx, and testing the
implementation;



The Electric Power Research Institute (EPRI), for sponsoring the development and testing
of the Volatility Basis Set (VBS) organic aerosol algorithm.



The American Petroleum Institute (API), for sponsoring the development and testing of
improvements to the vertical advection algorithm;



The Utah Department of Environmental Quality (UDEQ), for sponsoring updates to the
CB6 chemistry mechanism, snow‐cover treatment, and surface chemistry model;



The University of Texas, Center for Energy and Environmental Resources, for assistance in
developing and testing the Open‐MP multi‐processor capability;



Atmospheric, Meteorological, and Environmental Technologies (ATMET), for providing
libraries and implementation support for the MPI parallel processing capability;



The Midwest Ozone Group (MOG), for co‐sponsoring the development and testing of the
MPI parallel processing capability;



Atmospheric and Environmental Research (AER), for developing the mercury chemistry
algorithm;



The San Francisco Bay Area Air Quality Management District (BAAQMD), for supporting
and testing the coupling of SAPRC gas‐phase chemistry to the PM chemistry algorithm.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

i

www.camx.com

March 2016

CAMx User’s Guide Version 6.3

CONTENTS
ACKNOWLEDGMENTS............................................................................................................ I
1. OVERVIEW ........................................................................................................................1
1.1 CAMX FEATURES ........................................................................................................2
1.2 CAMX EXTENSIONS AND PROBING TOOLS .................................................................4
1.3 NEW FEATURES AND MAJOR UPDATES IN CAMX VERSION 6.3 ..................................5
2. THE CAMX MODELING SYSTEM .........................................................................................6
2.1 CAMX PROGRAM STRUCTURE ...................................................................................7
2.1.1 Memory Management .......................................................................................... 8
2.1.2 Parallel Processing ................................................................................................. 9
2.2 COMPILING CAMX ................................................................................................... 10
2.2.1 A Note on Fortran Binary Input/Output Files ...................................................... 11
2.3 RUNNING CAMX ...................................................................................................... 12
2.3.1 Control File Namelist Input.................................................................................. 12
2.3.2 Using Scripts to Run CAMx .................................................................................. 20
2.4 BENCHMARKING MODEL RUN TIMES....................................................................... 24
2.5 CAMX PRE‐ AND POST‐PROCESSORS ........................................................................ 24
2.5.1 Emissions ............................................................................................................. 24
2.5.2 Meteorology ........................................................................................................ 25
2.5.3 Photolysis Rates ................................................................................................... 26
2.5.4 Initial and Boundary Conditions .......................................................................... 27
2.5.5 Landuse ................................................................................................................ 27
2.5.6 Post‐processors ................................................................................................... 28
3. CORE MODEL INPUT/OUTPUT STRUCTURES.................................................................... 29
3.1
3.2
3.3
3.4

CAMX CHEMISTRY PARAMETERS FILE ...................................................................... 30
PHOTOLYSIS RATES FILE ........................................................................................... 38
OZONE COLUMN FILE............................................................................................... 40
FORTRAN BINARY INPUT/OUTPUT FILES .................................................................. 43
3.4.1 What is a Fortran Binary File? ............................................................................. 43
3.4.2 CAMx Binary File Headers ................................................................................... 44
3.4.3 Input Files ............................................................................................................ 45
3.4.4 Output Files ......................................................................................................... 58

4. CORE MODEL FORMULATION ......................................................................................... 63
4.1 NUMERICAL APPROACH........................................................................................... 63
4.2 CAMX GRID CONFIGURATION .................................................................................. 65
4.2.1 Grid Cell Arrangement ......................................................................................... 65
4.2.2 Grid Nesting ......................................................................................................... 66
4.2.3 Flexi‐Nesting ........................................................................................................ 68
4.3 TREATMENT OF EMISSIONS ..................................................................................... 68
4.3.1 Gridded Emissions ............................................................................................... 69
4.3.2 Elevated Point Emissions ..................................................................................... 69
4.4 THREE‐DIMENSIONAL TRANSPORT .......................................................................... 72
4.4.1 Resolved Transport: Advection ............................................................................ 72
4.4.2 Sub‐Grid Turbulent Transport: Diffusion ............................................................. 75
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

ii

www.camx.com

March 2016

CAMx User’s Guide Version 6.3

4.5 WET DEPOSITION ..................................................................................................... 77
4.5.1 Precipitation Parameters..................................................................................... 78
4.5.2 Gas Scavenging .................................................................................................... 79
4.5.3 Aerosol Scavenging .............................................................................................. 82
4.6 DRY DEPOSITION ..................................................................................................... 84
4.6.1 The Wesely/Slinn Model ..................................................................................... 85
4.6.2 The Zhang Model ................................................................................................. 88
4.7 SNOW COVER AND SURFACE ALBEDO...................................................................... 91
4.8 SURFACE MODEL FOR CHEMISTRY AND RE‐EMISSION ............................................. 93
4.8.1 Surface Model Algorithms ................................................................................... 94
4.8.2 Running CAMx With the Surface Model.............................................................. 98
5. CHEMISTRY MECHANISMS ............................................................................................ 101
5.1 GAS‐PHASE CHEMISTRY ......................................................................................... 102
5.1.1 Carbon Bond ...................................................................................................... 102
5.1.2 SAPRC 2007........................................................................................................ 109
5.1.3 Implicit Gas‐Phase Species ................................................................................ 109
5.1.4 Photolysis Rates ................................................................................................. 109
5.1.5 Gas‐Phase Chemistry Solvers ............................................................................ 112
5.2 AEROSOL CHEMISTRY ............................................................................................ 114
5.2.1 Additional Gas‐Phase Species............................................................................ 114
5.2.2 Aerosol Processes .............................................................................................. 114
5.2.3 Aerosol Sectional Approach .............................................................................. 122
5.3 MERCURY CHEMISTRY ........................................................................................... 123
5.3.1 Gas‐Phase Chemistry ......................................................................................... 124
5.3.2 Adsorption of Hg(II) on PM................................................................................ 124
5.3.3 Aqueous‐Phase Chemistry................................................................................. 125
5.3.4 Implementation Approach ................................................................................ 127
5.3.5 Chemistry Parameters for Mercury ................................................................... 128
5.4 SIMPLE CHEMISTRY VIA MECHANISM 10 ............................................................... 128
6. PLUME‐IN‐GRID SUBMODEL ......................................................................................... 130
6.1 CAMX PIG FORMULATION ..................................................................................... 130
6.1.1 Basic Puff Structure and Diffusive Growth ........................................................ 130
6.1.2 Puff Transport .................................................................................................... 134
6.2 GREASD PIG ........................................................................................................... 136
6.3 PARTICULATE MATTER IN PIG ................................................................................ 138
6.4 IRON PIG ................................................................................................................ 138
6.5 PIG FEATURES ........................................................................................................ 139
6.5.1 Puff Layer Spanning (IRON and GREASD) .......................................................... 139
6.5.2 Puff Overlap and the Idea of Virtual Dumping (IRON only) .............................. 139
6.5.3 Multiple Puff Reactors (IRON only) ................................................................... 140
6.5.4 Puff Dumping (IRON and GREASD) .................................................................... 141
6.5.5 PiG Rendering (IRON and GREASD) ................................................................... 141
6.5.6 High Resolution Puff Sampling (IRON and GREASD) ......................................... 142
6.6 DEPOSITION ........................................................................................................... 142
6.6.1 Dry Deposition ................................................................................................... 142
6.6.2 Wet Deposition .................................................................................................. 144
6.7 PIG CONFIGURATION ............................................................................................. 144
6.7.1 Guidance on the Use of CAMx PiG .................................................................... 145
7. SOURCE APPORTIONMENT ........................................................................................... 149
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

iii

www.camx.com

March 2016

CAMx User’s Guide Version 6.3

7.1 OZONE SOURCE APPORTIONMENT ......................................................................... 150
7.1.1 OSAT Formulation.............................................................................................. 151
7.1.2 OSAT2 Formulation ........................................................................................... 152
7.1.3 OSAT3 Formulation ........................................................................................... 153
7.1.4 Alternative Ozone Apportionment Using APCA ................................................ 155
7.2 PARTICULATE SOURCE APPORTIONMENT ............................................................... 156
7.3 RUNNING CAMX WITH SA ...................................................................................... 160
7.3.1 CAMx Control File .............................................................................................. 160
7.3.2 Specifying Emission Groups ............................................................................... 162
7.3.3 Source Area Mapping ........................................................................................ 166
7.3.4 Receptor Definition ........................................................................................... 170
7.3.5 Output File Formats ........................................................................................... 170
7.3.6 Postprocessing.................................................................................................... 172
7.4 STEPS IN DEVELOPING INPUTS AND RUNNING SA.................................................. 172
8. DECOUPLED DIRECT METHOD FOR SENSITIVITY ANALYSIS ............................................ 176
8.1 IMPLEMENTATION ................................................................................................. 177
8.1.1 Tracking Sensitivity Coefficients Within CAMx.................................................. 179
8.1.2 Flexi‐DDM .......................................................................................................... 180
8.2 RUNNING CAMX WITH DDM AND HDDM .............................................................. 180
8.3 DDM OUTPUT FILES ............................................................................................... 185
8.4 DDM SENSITIVITY COEFFICIENT NAMES ................................................................. 185
8.4.1 Initial Condition Sensitivity Names .................................................................... 186
8.5 STEPS IN DEVELOPING INPUTS AND RUNNING DDM .............................................. 189
9. PROCESS ANALYSIS ....................................................................................................... 191
9.1 PROCESS ANALYSIS IN CAMX ................................................................................. 191
9.1.1 Integrated Process Rate Analysis ...................................................................... 192
9.1.2 Integrated Reaction Rate Analysis..................................................................... 193
9.1.3 Chemical Process Analysis ................................................................................. 193
9.2 RUNNING PROCESS ANALYSIS ................................................................................ 196
9.2.1 Setting CAMx Parameters.................................................................................. 198
9.2.2 Output File Formats ........................................................................................... 199
9.3 POSTPROCESSING .................................................................................................. 199
9.3.1 IPR Output Files ................................................................................................. 199
9.3.2 IRR Output Files ................................................................................................. 200
9.3.3 CPA Output Files ................................................................................................ 201
10. REACTIVE TRACERS ..................................................................................................... 202
10.1 DESCRIPTION OF RTRAC ....................................................................................... 202
10.1.1 RTRAC Gas‐Phase Chemistry ........................................................................... 204
10.2 DESCRIPTION OF RTCMC ...................................................................................... 206
10.2.1 RTCMC Gas‐Phase Chemistry .......................................................................... 206
10.3 REACTIVE TRACERS IN IRON PIG........................................................................... 214
10.4 RUNNING CAMX WITH REACTIVE TRACERS .......................................................... 215
10.4.1 CAMx Control File ............................................................................................ 215
10.4.2 User Adjustable Parameters ............................................................................ 217
11. REFERENCES ................................................................................................................ 219
APPENDIX A ....................................................................................................................... 232
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

iv

www.camx.com

March 2016

CAMx User’s Guide Version 6.3

CAMX MECHANISM 2: CB6R2 GAS‐PHASE CHEMISTRY ................................................. 232
APPENDIX B ....................................................................................................................... 241
CAMX MECHANISM 3: CB6R2 WITH HALOGEN CHEMISTRY .......................................... 241
APPENDIX C ....................................................................................................................... 245
CAMX MECHANISM 6: CB05 GAS‐PHASE CHEMISTRY ................................................... 245
APPENDIX D....................................................................................................................... 251
CAMX MECHANISM 5: SAPRC07TC GAS‐PHASE CHEMISTRY ......................................... 251
TABLES
Table 2‐1. Parameters and their defaults in Includes/camx.prm used to statically
dimension local arrays in low‐level subroutines. ............................................................... 8
Table 2‐2. CAMx output file suffixes and their corresponding file types. ............................................... 20
Table 2‐3. CAMx v6.20 speed performance with MPI and OMP parallelization from the
LADCo tests described above. ............................................................................................ 24
Table 3‐1. Data requirements of CAMx. .................................................................................................. 29
Table 3‐2. Description of the CAMx chemistry parameters file. The record labels exist in
columns 1‐20, and where given, the input data for that record start in
column 21. The format denoted “list” indicates a free‐format list of
numbers (comma or space‐delimited)............................................................................... 31
Table 3‐3a. Rate constant expression types supported in CAMx and order of expression
parameters for the chemistry parameters file. ................................................................. 39
Table 3‐3b. List of parameters that must be provided in the CAMx chemistry parameter
file for each of the seven types of rate constant expressions. Use
ppm/minute units for A and Kelvin for Ea and TR. The variable O is the
order of the reaction (1 to 3). ............................................................................................ 40
Table 3‐4. The 11 WESELY89 landuse categories, their default UV surface albedos, and
their surface roughness values (m) by season. Winter is defined for
conditions where there is snow present; winter months with no snow are
assigned to the Fall category. Roughness for water is calculated from the
function

z0  2106 w2.5 , where w is surface wind speed (m/s)................................... 46

Table 3‐5. The 26 ZHANG03 landuse categories, their UV albedos, default annual
minimum and maximum LAI and surface roughness (m) ranges, and
mapping to the Wesely scheme (Table 3‐4). Roughness for water is
6

calculated from the function z 0  2 10 w , where w is surface wind
speed (m/s). ....................................................................................................................... 47
2.5

Table 4‐1. Summary of the CAMx models and methods for key physical processes. ............................. 63
Table 4‐2. Relationships between season and month/latitude used in the CAMx
Wesely/Slinn dry deposition model. Exception: seasons for the area
within 50N‐75N and 15W‐15E are internally set to those of latitude band
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

v

www.camx.com

March 2016

CAMx User’s Guide Version 6.3

35‐50 to account for regions of Europe in which the climate is influenced
by the Gulf Stream. ............................................................................................................ 88
Table 4‐3. Description of CAMx surface model variables shown in Figure 4‐7. ...................................... 95
Table 4‐4(a). Wesely landuse categories and associated annual‐averaged soil/vegetation
split factors, UV albedo, and SWE Wc. ............................................................................... 96
Table 4‐4(b). Zhang landuse categories and associated annual‐averaged soil/vegetation
split factors, UV albedo, and SWE Wc. ............................................................................... 96
Table 5‐1. Gas‐phase chemical mechanisms currently implemented in CAMx v6.3. .............................. 101
Table 5‐2. Species names and descriptions common to all Carbon Bond Mechanisms in
CAMx. ................................................................................................................................. 103
Table 5‐3. Default dry extinction efficiency and single‐scattering albedo at 350 nm
(Takemura et al., 2002) in the distributed CAMx chemistry parameters file. ................... 111
Table 5‐4. List of inorganic PM species for the CAMx CF aerosol option. ............................................... 116
Table 5‐5. SOA precursor reactions included in the CAMx SOAP module. .............................................. 117
Table 5‐6. Properties of CG/SOA pairs in the CAMx SOAP module. ........................................................ 117
Table 5‐7. Molecular properties of the 1.5‐D VBS species. ..................................................................... 120
Table 5‐8. Input species for 1.5‐D VBS scheme........................................................................................ 121
Table 5‐9. Volatility distribution factors used to allocate POA emissions from five
different source types to the five PAP, PCP, and PFP volatility bins. ................................. 121
Table 7‐1. Numbers of emission file sets (i.e., gridded files and point source file) needed
for different model configurations. APCA requires at least two emission
groups, and the first group must be biogenic emissions. .................................................. 166
Table 7‐2. Format for the receptor definition file.................................................................................... 171
Table 8‐1. DDM output file suffix names. ................................................................................................ 185
Table 9‐1. Process information reported by the IPR option. ................................................................... 193
Table 9‐2. Chemical Process Analysis (CPA) variables calculated in CAMx for the CB05
and CB6r2 mechanisms. Concentrations are ppb; production and
destruction are ppb/hr; photolysis rates are hr‐1, ratios are unitless. ............................... 194
Table 9‐3. Process analysis keywords and associated CAMx output files. .............................................. 196
Table 10‐1. Keywords, options and default values for the Control section of the IMC file. .................... 208
Table 10‐2a. Recommended SCICHEM rate constant expression types for use in CAMx. ...................... 213
Table 10‐2b. Parameters required by SCICHEM rate constant expression types. ................................... 214
Table 10‐3. Determining the reaction order and consequent unit dimensions for rate
constants. ........................................................................................................................... 214
Table 10‐4. RTCMC parameters default settings in the Includes/rtcmcchm.inc
include file. ......................................................................................................................... 218
Table A‐1. Reactions and rate constant expressions for the CB6r2 mechanism. k298 is the
rate constant at 298 K and 1 atmosphere using units in molecules/cm3 and
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

vi

www.camx.com

March 2016

CAMx User’s Guide Version 6.3

1/s. For photolysis reactions k298 shows the photolysis rate at a solar
zenith angle of 60° and height of 600 m MSL/AGL. See Table 5‐2 for
species names. See Section 3.1 on temperature and pressure
dependencies. .................................................................................................................... 232
Table B‐1. Listing of the CB6r2 halogen mechanism (see Table A‐1 for a complete listing
of CB6r2). k298 is the rate constant at 298 K and 1 atmosphere using units
in molecules/cm3 and 1/s. For photolysis reactions k298 shows the
photolysis rate at a solar zenith angle of 60° and height of 600 m
MSL/AGL. See Table B‐2 for species names. See Section 3.1 on
temperature and pressure dependencies. ........................................................................ 241
Table B‐2. Chemical species included in CB6r2h...................................................................................... 244
Table C‐1. Reactions and rate constant expressions for the CB05 mechanism. k298 is the
rate constant at 298 K and 1 atmosphere using units in molecules/cm3 and
1/s. See Table 5‐2 for species names. See Section 3.1 on temperature
and pressure dependencies. .............................................................................................. 245
Table D‐1. Reactions and rate constants for the SAPRC07TC mechanism. k300 is the rate
constant at 300 K and 1 atmosphere using units in molecules/cm‐3 and
1/s. See Table D‐2 for species names. See Section 3.1 on temperature
and pressure dependencies. .............................................................................................. 251
Table D‐2. Explicit species in the SAPRC07TC mechanism. ...................................................................... 270

FIGURES
Figure 2‐1. Schematic diagram of the CAMx modeling system. See Table 3‐1 for a detailed
list of specific model input requirements for the five major data classes shown
at the top of the figure. Certain pre‐ and post‐processor programs shown in
the figure are described in this section. Third‐party models, processors, and
visualization software are not described in this User’s Guide and are not
distributed with CAMx. .........................................................................................................6
Figure 2‐2. A sample CAMx job script that generates a “CAMx.in” file and runs the model
with OMP parallelization. ................................................................................................... 21
Figure 2‐3. An example of global ozone column from the Ozone Monitoring Instrument
(OMI) platform. White areas denote missing data. From
ftp://toms.gsfc.nasa.gov/pub/omi/data/. ......................................................................... 26
Figure 3‐1a. Example CAMx chemistry parameters file for Mechanism 6 (CB05) with CF PM
scheme that includes the mercury species HG0, HG2, and HGP. ...................................... 33
Figure 3‐1b. Example inert chemistry parameters file (requires chemistry flag to be set false
– see the description of the CAMx control file). ................................................................ 37
Figure 3‐2. Example of the first several panels of lookup data in the photolysis rates input
file....................................................................................................................................... 41

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

vii

www.camx.com

March 2016

CAMx User’s Guide Version 6.3

Figure 3‐3. Example structure of a single‐grid ozone column input file showing panels for
the optional time‐invariant land‐ocean mask and time‐varying ozone column
field. ................................................................................................................................... 43
Figure 4‐1. A horizontal representation of the Arakawa C variable configuration used in
CAMx. ................................................................................................................................. 66
Figure 4‐2. An example of horizontal grid nesting, showing two telescoping nested grids
within a 10×10 cell master grid. The outer nest contains 10×12 cells (including
buffer cells to hold internal lateral boundary conditions), and the inner nest
contains 6×10 cells (including buffer cells). ....................................................................... 67
Figure 4‐3. Schematic representation of the turbulent exchange among layers within a
vertical grid column during convective adjustment in the ACM2 (taken from
Pleim [2007]). ..................................................................................................................... 76
Figure 4‐4. Comparison of monthly LAI data embedded in the Zhang dry deposition scheme
against episode‐specific LAI for June 2005. ....................................................................... 90
Figure 4‐5. Comparison of particle dry deposition velocities as a function of size and wind
speeds (m/s) for three models: black – Zhang et al. (2001); blue – Slinn and
Slinn (1980); orange – AERMOD (EPA, 1998). Results are shown for a forest
landuse category during daytime neutral stability. Particle density was set at
1.5 g/cm3. ........................................................................................................................... 92
Figure 4‐6. Example of grid‐cell albedo evolution for a hypothetical 20‐day springtime snow
event (assuming ablation conditions) for low and tall vegetation grid cells with
a terrestrial (non‐snow) albedo of 0.05. ............................................................................ 93
Figure 4‐7. Schematic of the CAMx surface model. ................................................................................. 94
Figure 4‐8. The portions of the CAMx chemistry parameters file (highlighted) to support the
surface model. In this example, 3 gases are treated, where nitric acid (HNO3)
and peroxynitric acid (PNA) react to form nitrous acid (HONO). All three are
subject to decay by soil leaching, plant penetration, and snow melt loss. The
values shown here are for illustrative purposes only and do not represent any
known surface chemistry mechanism.............................................................................. 100
Figure 5‐1. Relative humidity adjustment factor applied to the dry extinction efficiency for
hygroscopic aerosols (FLAG, 2000). ................................................................................. 112
Figure 5‐2. Schematic diagram of the CAMx VBS module. The model VBS species name
consists of 4 characters that indicate the phase (P – particle; V – vapor), the
source (A – anthropogenic; B – biogenic; C – cooking; F – fire), the formation
(P – primary; S – secondary), and the volatility bin number. The solid and
dashed arrows represent gas‐aerosol partitioning and chemical aging,
respectively. The thick colored arrows represent POA emissions or oxidation
of SOA precursors. ........................................................................................................... 119

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

viii

www.camx.com

March 2016

CAMx User’s Guide Version 6.3

Figure 6‐1. Schematic representation of CAMx PiG puff shape in the horizontal plane.
Directional orientation of the puff is arbitrary, and evolves according to wind
direction, shears and diffusive growth along its trajectory. ............................................ 131
Figure 6‐2. Plan‐view schematic representation of a chain of PiG puffs emitted from a point
source into an evolving gridded wind field. The red line connected by dots
represents puff centerlines, with dots representing leading and trailing points
of each puff. The CAMx computational grid is denoted by the blue lines. ..................... 135
Figure 6‐3. Example of a single point source PiG plume as depicted by a sampling grid with
200 m resolution (shown by the extent of the plot; 40 km by 32 km total
extent). This sampling grid was set within a CAMx computational grid with 4‐
km resolution. The source location is arbitrary and is emitting an inert tracer. ............ 143
Figure 7‐1. Example of the sub‐division of a CAMx domain into separate areas for
geographic source apportionment. ................................................................................. 150
Figure 7‐2. The original OSAT scheme for ozone apportionment. Information flows along
arrows. Changes in core model species are shown in blue, OSAT tracers are in
black, the diamond represents the OSAT algorithm that determines ozone
tracer changes. H2O2/HNO3 is the indicator ratio used to determine NOx‐
or VOC‐limited ozone production. ................................................................................... 151
Figure 7‐3. Daytime reactions of ozone with HOx (OH and HO2) showing potential for
reformation of ozone or ozone destruction via peroxide formation. ............................. 153
Figure 7‐4. The OSAT2 scheme for ozone apportionment. Information flows along arrows.
Changes in core model species are shown in blue, OSAT tracers are in black,
the diamond represents the OSAT algorithm that determines ozone tracer
producton. H2O2/HNO3 is the indicator ratio used to determine NOx‐ or
VOC‐limited ozone production. ....................................................................................... 153
Figure 7‐5. Correspondence between NOy species in CB6 and tracer families in OSAT3 with
conversions between species/tracers shown by arrows. ................................................ 154
Figure 7‐6. The OSAT3 scheme for ozone apportionment. Information flows along arrows.
Changes in core model species are shown in blue, OSAT tracers are in black,
the diamond represent the OSAT algorithms that determine ozone tracer
production. H2O2/HNO3 is the indicator ratio used to determine NOx‐ or
VOC‐limited ozone production. RGN apportions the nitrogen in NO2 whereas
OON and OOV apportion the odd‐oxygen in NO2. ........................................................... 156
Figure 7‐7a. An example of SA input records in the CAMx run control file. The options for
this OSAT run are as follows: this is a two‐grid run, master and nested grid
surface concentrations are written to file, a single tracer type is to be used for
all boundaries, 19 source regions, and one emission group (i.e., zero
additional emission files and no leftover group). This is the first day of the
simulation (i.e., restart is false), so no OSAT restart files are supplied. .......................... 163
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

ix

www.camx.com

March 2016

CAMx User’s Guide Version 6.3

Figure 7‐7b. As in Figure 7‐7a, but in this case the run is a continuation day of a run with
three emission groups. The three emission groups are defined by supplying
extra emission files for point and area sources for each grid (emission groups 1
and 2), and setting the “Use_Leftover_Group” flag to TRUE so that the model
calculates the third group internally. The point source group 1 filename is
blank because group 1 is a category with no point source emissions (e.g.,
biogenics). ........................................................................................................................ 164
Figure 7‐7c. This figure follows from Figure 7‐7b: it is a continuation day of a 2‐grid run with
three emission groups, and all three emission groups are defined explicitly by
supplying extra emission files; therefore, the “Use_Leftover_Group” flag is set
to FALSE. The point source group 1 filename is blank because group 1 is a
category with no point source emissions (e.g., biogenics). APCA is used to
attribute ozone sources, so biogenic emissions MUST be present as group 1.
PSAT will trace PM sulfate and nitrate species. ............................................................... 165
Figure 7‐8. Example of the original source area map file for the domain and source areas
shown in Figure 7‐1. ......................................................................................................... 167
Figure 7‐9. Example fractional area map file for a small (10x10) grid. This file is for source
category/group #3 and includes 2 map panels. The grid covers source region
#5 and #6 and these regions overlap in the middle of the domain. Panel 2
shows just the remaining overlap information for region #6. ......................................... 169
Figure 7‐10. Example receptor concentration file. Lines ending with “…” are truncated to fit
the page, and the file would continue with data for additional receptors and
hours in the same format................................................................................................. 173
Figure 8‐1. Example of DDM inputs in the CAMx control file. CAMx is run with two grids,
and DDM is configured to track emissions from four source regions and two
source groups. Sensitivity to ozone initial and boundary conditions are
tracked, while sensitivities to NOx and VOC emissions are tracked. Sensitivity
for a single rate constant group will be calculated involving mechanism
reaction numbers 120, 121, and 122. Three groups of second‐order
sensitivities to anthropogenic NOx and VOC emissions (from emissions group
2, source region 1) will be computed (d2/dNOx2, d2/dVOC2 and d2/dNOxdVOC).
No source region map is provided for the nested grid (the region assignments
on the nest are defined by the master grid). Only the group 2 point sources
are tracked (no biogenic point sources are available). .................................................... 184
Figure 8‐2. Example concordance of long and short sensitivity coefficient names from the
CAMx diagnostic output file. ............................................................................................ 186
Figure 9‐1. Example section of a CAMx control file specifying options for Process Analysis. ............... 198
Figure 9‐2. Example IPR time series analysis for PSO4; lateral boundary and chemistry terms
are not aggregated. .......................................................................................................... 200
Figure 10‐1. Example RTRAC chemistry input file for modeling specific toxic species. ......................... 203
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

x

www.camx.com

March 2016

CAMx User’s Guide Version 6.3

Figure 10‐2. Example RTRAC receptor input file identifying the grid cells with locations
where hourly decay rates will be output for subgrid‐scale point source
modeling (see format for SA receptor file in Table 7‐2). ................................................. 206
Figure 10‐3. Example free‐format RTCMC IMC chemistry input file. .................................................... 207
Figure 10‐4. Example input of RTRAC options and filenames within the CAMx control file. ................ 216

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

xi

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
1. Overview

1. OVERVIEW
The Comprehensive Air quality Model with extensions (CAMx) is an Eulerian regional
photochemical dispersion model that allows for integrated “one‐atmosphere” assessments of
tropospheric air pollution (ozone, particulates, air toxics) over spatial scales ranging from
neighborhoods to continents. It is a “state‐of‐the‐science” open‐source system that is
computationally efficient, flexible, and publicly available. The model’s Fortran source code is
modular and well‐documented. The Fortran binary input/output file formats are based on the
Urban Airshed Model (UAM) convention and are compatible with many existing pre‐ and post‐
processing tools. Meteorological fields are supplied to CAMx from separate weather prediction
models. All emission inputs are supplied from external pre‐processing systems.
CAMx simulates the emission, dispersion, chemical reaction, and removal of pollutants by
marching the Eulerian continuity equation forward in time (t) for each chemical species (l) on a
system of nested three‐dimensional grids. The continuity equation specifically describes the
time dependency of volume‐average species concentration within each grid cell as a sum of all
physical and chemical processes operating on that volume. This equation is expressed
mathematically in terrain‐following height (z) coordinates as follows:
cl
t

   H  VH cl




cl

t Emission

  cl 
 2h 
 cl


z t 
 z
cl
t Chemistry



    K cl /  
cl
t Removal

where cl is species concentration (mass/volume), VH is the horizontal wind vector,  is the net
vertical transport rate, h is the layer interface height,  is atmospheric density, and K is the
turbulent exchange (diffusion) coefficient. The first term on the right‐hand side represents
horizontal advection, the second term represents net resolved vertical transport across an
arbitrary space‐ and time‐varying height grid, and the third term represents sub‐grid scale
turbulent diffusion. Chemistry is treated by simultaneously solving a set of reaction equations
defined by specific chemical mechanisms. Pollutant removal includes both dry surface uptake
(deposition) and wet scavenging by precipitation.
CAMx can perform simulations on four types of Cartesian map projections: Lambert Conic
Conformal, Polar Stereographic, Mercator, and Universal Transverse Mercator. CAMx also
offers the option of operating on a geodetic latitude/longitude grid system. The vertical grid
structure is defined externally, so layer interface heights may be specified as any arbitrary
function of space and/or time. This flexibility in defining the horizontal and vertical grid
structures allows CAMx to be configured to match the grid of any meteorological model that is
used to provide environmental input fields.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

1

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
1. Overview

1.1 CAMx Features
Two‐Way Nested Grid Structure: CAMx can be run with variable grid spacing. Use a coarse grid
for regional domains where high spatial resolution is not particularly needed, while in the same
run, nest finer grids in specific areas of interest. Two‐way nesting propagates information both
up‐ and down‐scale across all grids. Nests may possess different meshing factors from their
parent grids, as long as they are common denominators of parent resolution. A “Flexi‐Nesting”
feature allows you to introduce and/or remove nested grids at any point during a simulation.
You can supply complete information for new grids (emissions, meteorology, surface
characteristics) or allow CAMx to interpolate any or all of these inputs from parent grids.
Multiple Photochemical Gas Phase Chemistry Mechanisms: CAMx offers several versions of
Carbon Bond chemistry (CB05 and CB6 variants) and the 2007 version of Statewide Air Pollution
Research Center chemistry (SAPRC07TC). These mechanisms are solved using the Euler‐
Backward Iterative (EBI) method, which is fast and accurate. CAMx also includes the fully
explicit Gear‐type Livermore Solver for Ordinary Differential Equations (LSODE), which we use
to "benchmark" new mechanisms and evaluate the performance of EBI. We do not
recommend LSODE for typical applications as the model will run much more slowly.
Particulate Matter (PM) Chemistry: CAMx includes algorithms for inorganic aqueous chemistry
(RADM‐AQ), inorganic gas‐aerosol partitioning (ISORROPIA), and two approaches for organic
gas‐aerosol partitioning and oxidation (VBS or SOAP). These algorithms use products from the
gas‐phase mechanisms for the production of sulfate, nitrate, and condensable organic gases.
CAMx provides two options to represent the particle size distribution: a static two‐mode
coarse/fine (CF) scheme, and an evolving multi‐section (CMU) scheme. The hybrid 1.5‐
dimensional (1.5‐D) Volatility Basis Set (VBS) describes the evolution of organics according to
oxidation state and volatility, and is implemented to provide a unified framework for gas‐
aerosol partitioning and chemical aging of both primary and secondary atmospheric organic
aerosols. VBS is compatible only with the CB05 and CB6r2 gas‐phase chemistry and the 2‐mode
CF aerosol option; it is not currently enabled for Source Apportionment or Decoupled Direct
Method Probing Tools. The original one‐dimensional (over volatility) Secondary Organic
Aerosol Partitioning (SOAP) treatment remains an option. SOAP is compatible with CF and CMU
aerosol options and works with all Probing Tools.
Mercury Chemistry: CAMx optionally treats the chemistry of five mercury species (two gases
and three particulates) via gas‐phase and aqueous pathways, including Hg(II) adsorption to PM.
The mercury chemistry module requires PM concentrations, so mercury must be modeled with
the “CF” two‐mode PM mechanism by including mercury species among the list of modeled
species. All of the rate and equilibrium constants for the mercury mechanism are hard‐coded
within the chemistry module.
User‐Defined Chemistry Mechanism: “Mechanism 10” provides a simple way to define your
own chemistry mechanism. This option is intended to define simple chemical decay or
transformations between gas and/or aerosol species. You must develop your own Mechanism
10 subroutine and chemistry parameters file.
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

2

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
1. Overview

Plume‐in‐Grid (PiG) Module: PiG treats the chemistry and dispersion of point source emission
plumes at sub‐grid scales using a Lagrangian puff model, until such time as the pollutant mass
can be adequately represented within the grid model framework. Both gas‐phase and PM
chemistry can be treated. PiG includes a “sampling grid” capability to passively sample plume
concentrations at any resolution, which is particularly useful to visualize near‐source sub‐grid
scale impacts.
Horizontal Advection Solver Options: CAMx offers the Piecewise Parabolic Method (PPM) of
Colella and Woodward (1984), and the area‐preserving advection solver of Bott (1989). Both
possess high‐order accuracy, little numerical diffusion, and are sufficiently quick for applications
on very large grids.
Vertical Diffusion (Mixing) Options: By default, CAMx employs a standard “K‐theory” approach
for vertical diffusion to account for sub‐grid scale mixing layer‐to‐layer. Version 2 of the
Asymmetric Convective Model (ACM2; Pleim, 2007) is available as an alternative to the K‐
theory approach. ACM2 is a hybrid of local K‐theory and non‐local convective transport
between the surface and layers aloft. ACM2 can increase CAMx runtime considerably relative
to the default K‐theory. ACM2 does not work with the Integrated Process Rate (IPR)
component of the Process Analysis (PA) tool.
Dry Deposition Options: CAMx offers two dry deposition options: an older approach based on
the models of Wesely (1989) and Slinn and Slinn (1980); and an updated approach based on the
algorithms of Zhang et al. (2001; 2003). The Wesely/Slinn model is formulated for 11 landuse
categories, while the Zhang model uses 26 landuse categories.
Surface Chemistry/Re‐emission Model: CAMx includes a simple surface sub‐model that treats
sorption and penetration of deposited pollutant mass into soils and vegetation, chemical
degradation and transformation, and volatilization back into the air (re‐emission). The surface
model treats any subset of species listed in the core model’s chemical mechanism. The surface
model can only be used with the Wesely (1989) dry deposition option; it cannot be used with
the Plume‐in‐Grid treatment.
Advanced Photolysis Model: The TUV radiative transfer and photolysis model, developed and
distributed by the National Center of Atmospheric Research (NCAR, 2011), is used as a CAMx
preprocessor to provide the air quality model with a multi‐dimensional lookup table of clear‐sky
photolysis rates. CAMx internally adjusts clear‐sky rates for the presence of clouds and aerosols
using a fast in‐line version of TUV.
Lateral and Top Boundary Conditions: Time‐ and space‐variable boundary conditions for the
master grid may be developed from down‐scaling three‐dimensional output from global
chemistry models like GEOS‐Chem and MOZART. Top boundary conditions improve the
characterization of chemicals entering vertically across the model top, which is particularly
important for common stratospheric constituents such as ozone and nitrogen oxides. A simpler
top boundary treatment remains available, which is not reliant on an input file and internally
assumes a “zero gradient” volume mixing ratio condition between the top model layer and the
environment above the model.
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

3

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
1. Overview

Parallel Processing: CAMx supports two types of parallelization: (1) OpenMP (OMP), which
allows parallel processing on shared‐memory (e.g., multi‐core) computers; and (2) Message
Passing Interface (MPI), which allows parallel processing across distributed memory
(networked) computer cluster environments. Both OMP and MPI can be used in combination
to maximize speed performance. To use OMP, your Fortran compiler must include libraries to
enable in‐code directives. To use MPI, you must have external MPI libraries installed on your
system.

1.2 CAMx Extensions and Probing Tools
Ozone and Particulate Source Apportionment Technology (OSAT/PSAT): Source apportionment
technology tracks emission contributions to predicted ozone and/or PM species concentrations
by source region and/or category. OSAT also reports information to determine whether each
ozone component formed in NOx or VOC sensitive conditions. OSAT/PSAT provides ozone/PM
attribution to source regions and categories for a given emissions matrix, but does not provide
quantitative information as to how ozone/PM contributions would change as emissions are
altered because chemical interactions are non‐linear. Source apportionment is available only
for CB05, CB6r2 and CB6r3 chemical mechanisms, the CF aerosol scheme and the SOAP organic
partitioning algorithm.
Decoupled Direct Method (DDM) and High‐Order DDM (HDDM) Source Sensitivity: This tool
calculates first‐order (DDM) and second‐order (HDDM) gas concentration sensitivity to changes
in emissions, initial conditions and boundary conditions. PM concentration sensitivity is limited
to first‐order DDM. (H)DDM estimates how pollutant concentrations respond to region‐ and
category‐specific emission changes, but does not provide information on source attribution.
(H)DDM can be run with any CB or SAPRC chemical mechanism, the CF aerosol scheme and the
SOAP organic partitioning algorithm.
Process Analysis (PA): This probing tool provides in‐depth information on the physical and
chemical processes occurring during a CAMx run. Through PA, one can more fully understand
the complex interactions of the different processes, explain simulation results within the
context of model formulation, and improve the design of control strategies. The integrated
process rates (IPR) option can be run with any CB or SAPRC chemical mechanism and any PM
aerosol treatment. Chemical process analysis (IRR and CPA) is fully available only for CB05; a
limited set of chemical process rates are available for CB6r2. PM rates are not tracked by PA.
Reactive Tracers (RTRAC): RTRAC provides a flexible add‐on to simulate the emission,
dispersion, chemistry, and deposition of multiple gas and particle tracers (such as specific
toxics) that are not included in the model’s core gas/PM chemistry mechanisms. Gas‐phase
chemistry may involve user‐defined linear decay (photolysis and/or oxidation) by species, or
complex non‐linear systems solved with the RTRAC Chemical Mechanism Compiler (RTCMC).
RTRAC can be run in combination with any CB or SAPRC chemical mechanism and is
independent from all aerosol treatments.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

4

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
1. Overview

1.3 New Features and Major Updates In CAMx Version 6.3
Speed Improvements: Several modifications were implemented to improve model speed. Our
tests without Probing Tools indicate speed increases by 15‐50%, depending on compiler,
chipset and model configuration (grid number and size, PiG, chemistry mechanism,
parallelization). These changes have minimal impacts on concentration results (e.g., ozone
differences < 1 ppb).
Source Apportionment Updates: OSAT has been expanded to track odd oxygen and nitrogen
through NOy chemistry to account for NOx recycling of ozone and to improve consistency with
PSAT nitrate chemistry. Many more OSAT tracers are necessary, and this affects memory
requirements and model speed. Expect to see slightly more ozone contributions from long
range transport, and commensurately less contributions from local emissions. Both OSAT and
PSAT can now use a new source region map format that supports fractional (partial) area
assignments for each grid cell. The Geographic Ozone Assessment Technology (GOAT) option
has been removed from OSAT as it is considered obsolete.
Additional Map Projections: CAMx can now run on Mercator and Polar projections using the
projection parameters/definitions from the Weather Research and Forecasting (WRF) model.
Snow and Surface Chemistry Model Updates: Surface albedo for snow‐covered grid cells is now
calculated according to land cover type and new additional input variables for snow cover and
age, following the approach used in the WRF/NOAH land surface model. Net surface albedo in
snow‐covered grid cells can be substantially different from the original constant value of 50%,
and this can have a large impact on photochemical activity. The surface chemistry model has
been extended to work with the Zhang dry deposition option and to add snow cover to the
original soil and vegetation compartments. A third set of surface chemical sorption, reaction
and loss rates have been implemented to represent these processes on and within the
snowpack.
Updated CB6 Chemistry Mechanism: CB6 “revision 3” (CB6r3) is now available as chemistry
mechanism 4. CB6r3 includes a temperature‐ and pressure‐dependent organic nitrate
branching ratio. This update was found to be important for photochemistry in cold and/or
elevated conditions, such as wintertime in the US inter‐mountain west. Generally, the effect of
this change is to reduce ozone production slightly relative to CB6r2 in cold conditions. There
are no effects in warm conditions that are more typical of the ozone season.
SAPRC07TC: SAPRC07 is a more recent gas‐phase chemistry mechanism that has replaced the
dated SAPRC99 mechanism. SAPRC07TC is a variant of SAPRC07 that includes extra model
species for toxics and uses numerical expressions of rate constants that are compatible with the
current chemistry mechanism solver.
Implicit‐Explicit Hybrid (IEH) Chemistry Solver: This rarely used gas‐phase chemistry solver was
removed as it possesses equivalent accuracy as the EBI solver, but runs much more slowly.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

5

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
2. The CAMx Modeling System

2. THE CAMX MODELING SYSTEM
CAMx comprises the core component of an overall air quality modeling system, as illustrated in
Figure 2‐1. CAMx inputs are developed using independent third‐party models and processing
tools that characterize meteorology, emissions, and various other environmental conditions
(land cover, radiative/photolysis properties, and initial/boundary conditions). Interface
programs are needed to translate the products of each of these models/processors into the
specific input fields and formats required by CAMx. After the air quality simulation is
completed, additional programs are used to post‐process the concentration fields, develop
model performance statistics and measures, manipulate Probing Tool output into various
reportable formats, and further translate raw results into forms necessary for regulatory
purposes. Commonly available graphical software can be used to view CAMx output files; some
like PAVE and VERDI can read CAMx files directly, others require reformatting CAMx files to
common data formats like NetCDF. While third‐party visualization software, meteorological
models, and emission processors are not distributed with CAMx, Ramboll Environ does
distribute many of the necessary interface programs and post‐processors on the CAMx web site
(www.camx.com). A brief description of each of these is provided at the end of this section.

Data

Emissions

Meteorology

Photolysis

Geographic

Emission Inventory,
Fire Activity

Analyses,
Observations,
Topography,
Landcover

Total
Atmospheric
Ozone Column

Land/Ocean Mask

Vegetative
Cover

Air Quality

Landcover
Leaf Area Index

Models &
Pre-Processors

SMOKE,
CONCEPT,
EPS3

WRF,
MM5,
RAMS

O3MAP

GIS Processing

ICBCPREP

TUV

MERGE LULAI

Global Models
(GEOS-CHEM,
MOZART, AM3)

Biogenic Models
SEASALT

Interface
Programs

PiGSET,
WINDOW,
MRGUAM

WRFCAMx,
MM5CAMx,
RAMSCAMx,
KVPATCH

GEOS2CAMx,
MOZART2CAMx

Core Model

Post-Processors

BNDEXT

CAMx

CAMxTRCT,
CAMx2IOAPI,
BIN2ASC

AVGDIF,
CAMxPOST,
EPASTAT

MATS

File Formatting

Performance

Regulatory

PA Tools,
User-Developed Post-Processors

Probing Tools

Figure 2‐1. Schematic diagram of the CAMx modeling system. See Table 3‐1 for a detailed list
of specific model input requirements for the five major data classes shown at the top of the
figure. Certain pre‐ and post‐processor programs shown in the figure are described in this
section. Third‐party models, processors, and visualization software are not described in this
User’s Guide and are not distributed with CAMx.
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

6

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
2. The CAMx Modeling System

2.1 CAMx Program Structure
The core CAMx model is written in Fortran, but includes some utilities written in C to interface
with MPI. The program is highly modular and well documented to ease code review,
modifications and substitution of alternate routines. The source code is arranged in several
directories, grouped according to function. The main source directory contains version release
notes, the GNU user license, the “Makefile” compile utility, and a control file namelist template.
Sub‐directories contain source code for the core model and ancillary routines according to the
following:
CAMx/

Source code for the main driver routine CAMx.f and core model
routines.

CF_AERO/

Source code for inorganic aerosol chemistry (aqueous and
thermodynamic partitioning) for the 2‐mode CF scheme.

CMC/

Source code for the gas‐phase chemical mechanism routines.

CMU_AERO/ Source code for inorganic aerosol chemistry (aqueous and
thermodynamic partitioning) for the multi‐section CMU scheme.
DDM/

Source code for the (H)DDM Probing Tool, consisting of I/O and core
routines that are unique to (H)DDM.

HG/

Source code for the mercury chemistry routines.

Includes/ Fortran “include” files, consisting of program parameters and memory
management code.
IO_bin/

Source code for Fortran binary (unformatted) I/O.

Mod_src/

Source code for F90 memory management modules.

MPI/

Source code for routines specific to MPI parallelization.

OSAT/

Source code for the OSAT/PSAT Probing Tools, consisting of I/O and core
routines that are unique to OSAT/PSAT.

PA/

Source code for the Process Analysis Probing Tool, consisting of I/O and
core routines that are unique to PA.

PiG/

Source code for the Plume‐in‐Grid sub‐model, consisting of I/O and core
routines that are unique to PiG.

RTRAC/

Source code for the Reactive Tracer Probing Tool, consisting of I/O and
core routines that are unique to RTRAC/RTCMC.

SOAP/

Source code for secondary organic aerosol thermodynamic partitioning.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

7

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
2. The CAMx Modeling System

2.1.1 Memory Management
All of the model’s global data structures are dynamically allocated when the model starts. The
data necessary to allocate memory space for a given model run are read from the CAMx control
file developed by the user (see Section 2.3). However, to alleviate compiler dependency on
speed performance, CAMx utilizes some hard‐coded Fortran parameters to statically allocate
local arrays in low‐level subroutines. All of these parameters are defined in the
Includes/camx.prm file. The distribution version of this “include” file sets key array
parameters to default values that should be sufficiently large to accommodate most
applications: see Table 2‐1 for a description of parameters and their default values. However,
you may want to customize these values to ensure that they are large enough to accommodate
all of your model configurations, or to exactly match your specific application, thus preventing
wasted memory.
If any parameter is set to a value that is too small to support your application the model will
stop, displaying an informative error message. To conserve memory, default values of 1 are
set for Probing Tool tracers, PiG sampling grids, and sampling grid dimensions. These must be
increased accordingly if Probing Tools or sampling grids are to be used.

Table 2‐1. Parameters and their defaults in Includes/camx.prm used to statically
dimension local arrays in low‐level subroutines.
Parameter Name
MXCELLS
MXLAYER
MXSPEC
MXREACT
MXGRID
MXPTSRC
MXTRSP
MXPIG
MXSAMPLE
MXCOLSMP
MXROWSMP

Description
Number of cells in X/Y direction for any grid
Number of layers
Number of species (could be number of radicals, number of input
species, or total number of model species)
Number of reactions (depends on the mechanism; see the user's
guide for the value for each mechanism)
Number of grids
Number of point sources
Number of Probing Tool tracer species
Number of PiG puffs
Number of PiG sampling grids
Number of PiG sampling grid columns
Number of PiG sampling grid rows

Default
Value
200
30
133
565
10
100000
1
50000
1
1
1

All of the parameters in the table above can be determined before starting a simulation except
for MXPIG. A value of 50,000 is usually sufficient for most applications in which PiG is used; set
this parameter to 1 if PiG is not used to conserve memory. If this parameter is exceeded during
a simulation, the model will stop with an informative error message. If this happens, simply
increase MXPIG, recompile the model executable, and restart the simulation. The other
parameters in camx.prm beyond those listed in Table 2‐1 will not normally need to be
changed and are not discussed further.
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

8

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
2. The CAMx Modeling System

2.1.2 Parallel Processing
Parallel processing refers to distributing a model application to multiple processors (CPUs) that
share the computational load. CAMx supports two types of parallelization: (1) OpenMP (OMP),
which allows parallel processing on shared‐memory (e.g., multi‐core) computers; and (2)
Message Passing Interface (MPI), which allows parallel processing across distributed memory
(networked) computer cluster environments. Both OMP and MPI can be used in combination
to maximize speed performance.
To use OMP, your Fortran compiler must include libraries to enable the in‐code parallelization
directives. OMP distributes calculations for individual processes, such as chemistry within a
single grid cell or advection/diffusion along a single row of cells, to a number of CPUs defined
by the user. Once each CPU has completed its calculations, it works on the next individual
process until all processes over the entire grid are completed.
To use MPI, you must have an external MPI library installed on your system. MPICH is a specific
open source MPI library widely used in the numerical modeling community; CAMx has been
specifically developed and tested using MPICH. With MPI, each CAMx grid is divided into sub‐
domains (“slices”) and each slice is assigned to a CPU on the user‐defined network. Each CPU
operates the entire model on its assigned slice and passes common information needed by
other CPUs via data “messages”.
MPI in CAMx is designed using a “master/slave” parallel processing approach. The CPU on
which the program is launched serves as the master node and will not conduct any model
computations on any part of the modeling domain. This process will perform all of the model
setup, the vast majority of I/O, and manage the communication between the slave or compute
nodes, which integrate the model forward for each grid slice. Since the master node handles
the important I/O it is the only CPU that needs access to the disk volume containing the input
files and the location of the output directory. This approach allows for a minimal amount of
network traffic to/among the compute nodes by eliminating the need for them to manage NFS
mounts. The master node may need access to the LAN for data access, but the compute nodes
only need access to the internal cluster network. However, the compute nodes will need access
to a copy of the executable program. This can be accomplished in a number of ways: (1) have
an NFS mount on the master node accessible to the internal cluster network and launch the
model from that location; or (2) port a copy of the executable program, using rcp or scp, to
the user’s home directory on each compute node and launch the model from the user’s home
directory on the master node.
During each model time step, when grid slice computations are performed by the compute
nodes, some information is written to the diagnostic and message output files. Rather than just
eliminate this information altogether, we decided to create node‐specific versions of each of
these two files and have each compute node write the information to its own version.
However, in order to prevent the need to have the output directory available to the compute
node across the network, we have designed the model so that the node‐specific files are
created in the current working directory. This means that if the model is launched from an NFS‐
mounted directory, all of the node‐specific files will all be created in that location. On the other
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

9

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
2. The CAMx Modeling System

hand, if the model is launched from a user's home directory on the compute nodes, you will
have to log in to the specific compute node to view the files.
When using a hybrid MPI/OMP approach, the grids will be divided into slices in the usual way as
part of MPI, but when operating on a particular slice, the host will spawn multiple OMP threads
to parallelize the portions of the code where OMP directives have been included.

2.2 Compiling CAMx
A “Makefile” script is provided in the main source directory. The Makefile will compile all CAMx
source code, link with necessary libraries, and build an executable program. It supports
platforms running Linux (Portland Group and Intel compilers) and Macintosh OSX (Absoft
compiler); it does not currently support compilers on platforms running Windows.
The choice for OMP and MPI parallelization, and the memory configuration for probing tools,
are set during model compilation. All other CAMx choices for chemical mechanism, model
algorithms, Probing Tools, and other options are selected at run time.
CAMx is compiled by issuing the following command at a shell prompt within the main source
directory:
make COMPILER=my_compiler  
where the text within the brackets “<>” is optional. To display an interactive help message,
you may type
make help
The mandatory COMPILER argument should be set to one of the following:
pgf or pgfomp (Portland Group compiler for Linux)
ifort or ifortomp (Intel compiler for Linux)
gfortran or gfortranomp (Gnu compiler for Linux)
absoft or absoftomp (Absoft compiler for Macintosh OSX)
These keywords inform the Makefile of the compiler being used to compile and run the model,
so that the Makefile can invoke the proper compiler‐specific commands and flags. If OMP is not
specified as part of the keyword then CAMx will not be able to run with OMP parallelization.
The optional CONFIG argument allows the CAMx executable program to be labeled for a
specific memory configuration as defined within the CAMx parameters file
(Includes/camx.prm) described above. You may want to customize some applications, for
example to configure the Probing Tool extensions, and it is convenient to be able to distinguish
between these executables. The Makefile will search for a CAMx parameters file called:
Includes/camx.prm.my_app
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

10

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
2. The CAMx Modeling System

If CONFIG is not set on the make command line, the Makefile will compile CAMx using the
default parameters file, Includes/camx.prm.v6.30.
The optional MPI argument will enable MPI parallel processing. This requires that third‐party
MPI libraries have been built and installed on the machine that is running this Makefile script
and compiling CAMx. If MPI is not set on the command line, the Makefile script will ignore the
MPI libraries and CAMx will not be able to run with MPI parallelization. The optional MPI
argument should be set to one of the following:
mpich (MPICH versions 1 or 2)
mpich3 (MPICH version 3)
mvapch
openmpi (PGF and IFORT compilers only)
You should check that the variable “MPI_INST” in the CAMx Makefile, and in the MPI
utilities Makefile (located in the MPI/util sub‐directory), are correctly set to your system's
MPI installation path.
CAMx supports the use of both OMP and MPI parallelization in a single run using PGF and IFORT
compilers. To utilize OMP in your MPI application, be sure to specify the appropriate OMP
compiler keyword.
The Makefile will generate a CAMx executable program named
CAMx.my_app.MPI_option.my_compiler
which will reside in the main source directory. For example, a default compilation using the
Portland Group compiler will result in an executable named
CAMx.v6.30.noMPI.pgf.
If you need to rebuild CAMx using different Makefile arguments we recommend typing “make
clean” between builds. Make clean will delete all existing object files and force a complete
re‐build.
2.2.1 A Note on Fortran Binary Input/Output Files
Large CAMx input and output data fields are contained within Fortran “unformatted” (binary)
files. This means that the data are read and written as represented in memory, without
translation between binary and ASCII character sets as is performed for “text” files. Binary files
reduce file volume and improve program read/write speed, but the user cannot directly view or
manually edit them. There are two ways to represent binary information in memory: “big
endian” and “little endian.” The difference between these is essentially the order of bits in a
word, and which order is used depends on the computer chipset. Historically, big endian has
been used in many Unix workstations (Sun, SGI, HP, and IBM). The x86 processors on personal
computer platforms (e.g., Intel and AMD) use little endian, while PowerPC chips are big endian.
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

11

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
2. The CAMx Modeling System

CAMx can be compiled and run on machines that use either big or little endian binary
representations, as long as the model and all of its pre‐ and post‐processors are consistently
compiled and run on the same type of platform. If any component of the modeling system is
compiled on a different platform using the opposite binary representation, I/O files will not be
properly read and will likely lead to a program crash.
A typical run‐time error message from trying to read the wrong binary format is “input record
too long,” so if you get this error message, check for big endian / little endian consistency
between your binary files and Fortran compiler options.
Compilers for little endian machines (e.g., x86 PC chipsets) provide compile‐time switches that
allow binary files to be read and written as big endian. The Portland Group compiler option is
“-byteswapio”, whereas the Intel compiler option is “-convert big_endian.” The
CAMx Makefile sets compiler flags to consistently use big endian to maximize platform
portability. Therefore, use of the CAMx Makefile will by default result in the model reading and
writing big endian binary files. In practice, users should use the default binary format that is
built into the CAMx Makefile and that is used for the CAMx distribution test case.

2.3 Running CAMx
2.3.1 Control File Namelist Input
CAMx reads a text run control file named “CAMx.in” that must exist locally in the directory
from which the model is run. This file must be in the Fortran “namelist” format, and contains
all user‐specified control parameters for a given simulation, including model configuration,
option‐specific inputs, and I/O filenames. The run control file must contain the primary
namelist module labeled “&CAMx_Control”, which provides all of the information to
configure the core model. Additional namelist modules may be provided in the run control file
to configure the various CAMx Probing Tool extensions. These optional namelist modules are
ignored if no Probing Tools are selected in the primary namelist.
Each record in the CAMx control file contains a variable name that is explicitly set to a
numerical, logical, or character value. The variable names are used by the program directly,
and therefore cannot be changed without source code modifications. Character strings must
be enclosed by single quotes, and all variable assignments must be delimited with commas.
The order of the records may be arranged in any fashion that the user prefers. Any number of
comment statements may be included anywhere within the namelists, provided that they do
not interrupt variable assignments (variable_name = value,). The “!” character is the
Fortran namelist comment delimiter.
Certain variables are multi‐dimension arrays; the user may provide a comma‐delimited list of
values to fill the array or assign values to specific array elements. Certain other variables are
optional or associated with option flags; these do not need to appear in the namelist if their
associated options are not invoked, and they will be ignored if they remain in the file.
If the user does not provide necessary inputs, the model will stop with a descriptive error
message.
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

12

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
2. The CAMx Modeling System

2.3.1.1 Common Errors When Creating a Namelist
Fortran programs ingest the entire contents of namelist modules using a single READ
statement. If the program experiences an error reading the namelist, it echoes a simple error
message like “error reading namelist” and the program stops. It is therefore difficult to
determine the cause of the read error, especially if the namelist is lengthy and contains a
variety of data types. When experiencing an error reading the CAMx control file namelist, you
must carefully inspect the file for any syntax errors. These errors can be subtle and difficult to
spot. Here are a few of the common reasons an error occurs when reading a namelist:


Mistyped variable name:
All variables to be assigned within a namelist must be recognized as a declared namelist
variable within the reading program. If a variable is misspelled or an unknown variable
is assigned a value, a read error will occur.



Incorrect data type for the assigned variable:
If the data type of the value assigned to a namelist variable does not match the
variable’s declared data type within the reading program, an error will occur. Some
compilers will allow real type variables to be assigned to integer values, but not the
converse.



Missing period around a logical value:
The logical values .true. and .false. must be surrounded by a period.



Missing quotes around a character variable:
Any character data type must be surrounded by quotes.



Overflow when assigning values to an array:
The values in an array can be assigned using array index notation. If the index used to
assign an array value exceeds the declared dimension of the array, a read error occurs.
Check the “MXNAM” parameter in the Includes/namelist.inc include file to see
if this value needs to be increased. Alternatively, check your namelist file to be sure all
of your array indices are correct.



Wrong number of dimensions when assigning values to a multi‐dimensional array:
When assigning values to an array using array index notation, the number of subscripts
in the assignment must match the declared dimensions of the array (e.g., assignments
to an array dimensioned var(i,j) must be referenced using two indices).



Missing comma following a variable definition:
A comma must be the last character in a variable assignment (variable = value,).
A comment may be placed after the comma (delimited using the “!” symbol, see below)
on the same file record. This restriction on the use of commas is ignored on some
compilers.



Too many commas following a scalar variable definition:
More than one comma following a scalar variable assignment will result in a read error.



Too many commas following the variable assignment list for an array:

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

13

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
2. The CAMx Modeling System

The entire contents of an array can be assigned using a single statement by listing the
values of each element separated by commas. The read will fail if there are more
commas than the dimension of the array.


Comment does not begin with !:
The character that delimits a comment in a namelist is the exclamation point.
Comments can appear anywhere within the namelist. However, all text in the namelist
must either be part of a namelist variable assignment or part of an identified comment.

We suggest that new CAMx users start with the “CAMx.namelist.template” that is
provided with the source code.
2.3.1.2 The Primary Namelist Module
This section describes the primary namelist module; detailed descriptions of each of the
Probing Tool modules are provided in their respective sections (Sections 7 through 10). A listing
of all namelist variables necessary to run the core model is presented on the following pages.
Description of CAMx Run Control File Variables
&CAMx_Control

Label for the primary namelist module that configures the core
model; it must begin in column 2

&

Flag ending a namelist module; it must be in column 2

Run_Message

60‐character simulation message, written to output files to label
the run

The short simulation “run message” is written to all output files to describe and label the run.
Model Clock Control
Time_Zone

Integer time zone (0=UTC, 5=EST, 6=CST, 7=MST, 8=PST)

Restart

Logical model restart flag (TRUE=read restart file, FALSE=read
initial conditions file)

Start_Date_Hour

Integer array start time (YYYY, MM, DD, HHmm)

End_Date_Hour

Integer array end time (YYYY, MM, DD, HHmm)

Maximum_Timestep

Real maximum allowable timestep (minutes)

Met_Input_Frequency

Real input frequency of environmental fields (minutes)

Ems_Input_Frequency

Real input frequency of emissions (minutes)

Output_Frequency

Real output frequency (minutes)

The user specifies the simulation start/end year, month, day, and hour; the model uses Julian
dates internally. All times must be given in military format (e.g., 1:30 PM must be given as
1330). The simulation time zone must match the time zone in which the emission and
environmental inputs are developed.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

14

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
2. The CAMx Modeling System

Map Projection Parameters
Map_Projection

Character map projection keyword (LAMBERT, POLAR, RPOLAR,
MERCATOR, UTM, LATLON)

UTM_Zone

Integer UTM zone

Longitude_Pole

Real longitude of projection pole or origin (degrees, west<0)

Latitude_Pole

Real latitude of projection pole or origin (degrees, south<0)

True_Latitude1

Real first true latitude of projection (degrees, south<0)

True_Latitude2

Real second true latitude of projection (degrees, south<0)

The grid projection may be selected as Cartesian (fixed physical distance coordinates on a flat
plane) or curvi‐linear geodetic (following the curved surface of the Earth). The Cartesian
options include Lambert Conic Conformal (LAMBERT), Polar Stereographic (POLAR), Rotated
Polar Stereographic (RPOLAR), Mercator (MERCATOR), and Universal Transverse Mercator
(UTM). The geodetic option performs the simulation on a latitude/longitude grid (LATLON). All
gridded input files must be defined on the grid projection specified for the CAMx simulation.
The LAMBERT, POLAR, and MERCATOR projections are all equivalent to the definitions used in
the WRF meteorological model, which assumes a spherical Earth with radius of 6370 km. The
RPOLAR projection is equivalent to the definition used in the RAMS meteorological model.
While the POLAR projection of WRF is defined to be tangent at (or secant around) the North
and South Poles, the RPOLAR projection of RAMS is defined to be only tangent to the Earth’s
surface at any user‐defined latitude/longitude.
If the LAMBERT projection is specified, the Longitude_Pole and Latitude_Pole must
be specified to define the projection origin (where LAMBERT coordinates are defined to be 0,0
km), and True_Latitude1 and True_Latitude2 must be specified to define the
projection true latitudes (they may be equal, which is a projection tanget at that latitude).
If the MERCATOR projection is specified, the Longitude_Pole and Latitude_Pole must
be specified to define the projection origin (where MERCATOR coordinates are defined to be
0,0 km), and True_Latitude1 must be specified to define the projection true latitude (it
may be zero, which is a projection tangent at the Equator).
If the POLAR projection is specified, the Longitude_Pole and Latitude_Pole must be
specified to define the projection origin (where coordinates are defined to be 0,0 km), and
True_Latitude1 must be specified to define the projection true latitude or secant (it may
be ±90 degrees, which is a projection tangent at the North or South Poles).
If the RPOLAR projection is specified, the Longitude_Pole and Latitude_Pole must be
specified to define the projection pole (where coordinates are defined to be 0,0 km). True
latitudes are not specified as RPOLAR is always tangent at the pole point.
If the UTM projection is specified, a UTM zone must be specified (1 through 60). Pole and true
latitude values are ignored for UTM and LATLON projections.
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

15

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
2. The CAMx Modeling System

Parameters For The Master (First) Grid
Number_of_Grids

Integer number of grids in simulation

Master_SW_XCoord

Real x‐coordinate of domain southwest corner (km, or degrees
for LATLON)

Master_SW_YCoord

Real y‐coordinate of domain southwest corner (km, or degrees
for LATLON)

Master_Cell_XSize

Real cell size in x (km, or degrees for LATLON)

Master_Cell_Ysize

Real cell size in y (km, or degrees for LATLON)

Master_Grid_Columns

Integer number of master grid columns (E‐W grid cells)

Master_Grid_Rows

Integer number of master grid rows (N‐S grid cells)

Number_of_Layers

Integer number of grid layers (applies to all grids)

The master grid is defined by its location (southwest corner of cell [1,1] in the coordinates of
the chosen projection space), number of grid cells (east‐west, north‐south, vertically), and
horizontal resolution. Vertical resolution is defined by the layer structure specified in the input
3D meteorological file.
Parameters For The Nested Grids
Nest_Meshing_Factor

Integer array (by grid) nested grid cell size relative to master
grid

Nest_Beg_I_Index

Integer array (by grid) master grid column containing western
edge of nest

Nest_End_I_Index

Integer array (by grid) master grid column containing eastern
edge of nest

Nest_Beg_J_Index

Integer array (by grid) master grid row containing southern edge
of nest

Nest_End_J_Index

Integer array (by grid) master grid row containing northern edge
of nest

The definition of nested grids is specified in the CAMx.in file in terms of the range of master
grid cells that each nested grid spans (see Section 4). The “meshing factor” sets the resolution
or cell size of the nested grids relative to the master grid. The CAMx diagnostic output file
provides information on the location and size of each nested grid to help ensure proper setup.
Model Options
Diagnostic_Error_Check

Logical model startup diagnostic flag (TRUE=stops before first
timestep indicating successful model initialization,
FALSE=continues with simulation after model initialization)

Flexi_Nest

Logical flexi‐nesting flag (TRUE=allow some/all nested input
fields to be interpolated from the parent grid, FALSE=all data
must be provided for all nests)

Advection_Solver

Character horizontal advection solver keyword (PPM, BOTT)

Chemistry_Solver

Character chemistry solver keyword (EBI, LSODE)

PiG_Submodel

Character PiG submodel keyword (NONE, GREASD, IRON)

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

16

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
2. The CAMx Modeling System

Probing_Tool

Character Probing Tool keyword (NONE, SA, DDM, HDDM, PA,
IPR, IRR, RTRAC, RTCMC)

Chemistry

Logical chemistry flag (TRUE=chemistry on, FALSE=chemistry
off)

Drydep_Model

Character dry deposition model keyword (NONE, WESELY89,
ZHANG03)

Wet_Deposition

Logical wet deposition flag (TRUE=deposition on,
FALSE=deposition off)

ACM2_Diffusion

Logical ACM2 vertical diffusion flag (TRUE=ACM2 on,
FALSE=standard K‐theory diffusion)

Surface_Model

Logical surface model flag (TRUE=surface model on,
FALSE=surface model off)

Super_Stepping

Logical super‐stepping flag (TRUE=use super‐stepping for
horizontal advection to maximize model speed, FALSE=do not
use super‐stepping).

Gridded_Emissions

Logical gridded emissions flag (TRUE=gridded emissions will be
used, FALSE=gridded emissions will be ignored)

Point_Emissions

Logical elevated point source flag (TRUE=point emissions will be
used, FALSE=point emissions will be ignored)

Ignore_Emission_Dates

Logical date‐insensitive emission flag (TRUE=dates on emission
files will be ignored, FALSE=dates on emission files will be
checked against simulation date)

The user has the option of selecting among the Bott or Piecewise Parabolic Method horizontal
advection solvers by specifying “BOTT” or “PPM” as keywords in the run control file. The user
also has the option to use the EBI or LSODE chemistry solvers for gas‐phase chemistry by
specifying these respective keywords. Probing Tools are selected by specifying one of the
allowed keywords; no Probing Tool will be run if this keyword is set to “None”. The description
of the PiG submodel is provided in Section 6.
Super stepping maximizes the model’s speed performance by setting the largest grid‐specific
driving time steps possible. This results in the need for potentially many sub‐steps to be
applied in horizontal advection on a layer‐by‐layer basis to maintain a stable solution. While
super stepping has little impact on surface concentrations in non‐MPI mode, larger differences
are seen using MPI. A “super‐stepping” flag was added to the control namelist that allows
users to specifically turn off super stepping when they wish to compare concentrations
between MPI and non‐MPI runs in the most consistent manner possible. Super stepping can
reduce the accuracy of the vertical transport solution, especially in high wind conditions over
complex terrain. Turning super stepping off will cause the model to run much more slowly.
Output Specifications
Root_Output_Name

Character root output path/filename (see Table 2‐2 for
description of file suffixes)

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

17

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
2. The CAMx Modeling System

Average_Output_3D

Logical 3‐D average output file flag (TRUE=output full 3‐D
concentration fields, FALSE=output surface layer concentration
fields)

Output_3D_Grid

Logical array (by grid) 3‐D average output file flag (TRUE=output
full 3‐D concentration fields for specified grid, FALSE=output
surface layer concentration fields)

Output_Species_Names

Character array (by output species) species names to be written
to average and deposition output files, or the single name “ALL”
to output all state gas and PM species (excluding radicals), or
the sinlge name “ALLR” to include radicals

PiG_Sampling_Grid

Logical sampling grid flag for IRON PiG output (TRUE=sampling
grids are specified, FALSE=sampling grids will not be generated)

Sample_Background

Logical flag to include background concentrations
(TRUE=background concentrations from the host computational
grid will be added to puff increments, FALSE=only puff
increments will be shown)

Number_of_Sampling_Grids

Integer number of sampling grids

SG_Beg_I_Index

Integer array (by sampling grid) master grid column containing
western edge of sampling grid

SG_End_I_Index

Integer array (by sampling grid) master grid column containing
eastern edge of sampling grid

SG_Beg_J_Index

Integer array (by sampling grid) master grid row containing
southern edge of sampling grid

SG_End_J_Index

Integer array (by sampling grid) master grid row containing
northern edge of sampling grid

SG_Mesh_Factor

Integer array (by sampling grid) cell size relative to master grid

The user specifies a “root” path and filename that will be used for all standard CAMx core
model output files. The model appends suffixes to these root names according to the file type
generated.
The types of CAMx output files are listed in Table 2‐2. A subset of state (gas or PM) and radical
species may be output to the average concentration output files; see the description of output
file formats in Section 3. By specifying a single output name “ALL”, the model will automatically
output fields for all state gas and PM species listed in the input chemistry parameters file,
excluding radicals (use “ALLR” to include radicals). If “ALL” or “ALLR” are specified, it must be
the only name listed; no species names may be listed before or after “ALL”. There are two flags
that control whether 3‐D average output files are generated. The first (original) flag will toggle
3‐D output for all grids in the run. The second is the “Output_3D_Grid” flag array, which
allows 3‐D average output to be set for specific grids. The original flag supersedes the grid‐
specific flag.
PiG sampling grids are set identically to the way nested grids are specified for the host model,
with one exception: there are no vertical levels to define (sampling grids are currently only 2‐D
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

18

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
2. The CAMx Modeling System

layer 1 fields). The same rules that apply for the specification of nested grids holds for the
specification of all sampling grids (see Section 4).
Input Files
Chemistry_Parameters

Character input chemistry parameters path/filename

Photolysis_Rates

Character input photolysis rates path/filename (optional
according to Chemistry flag and Chemistry_Parameters file)

Ozone_Column

Character input ozone column path/filename (ignored if
Chemistry=FALSE)

Initial_Conditions

Character input master grid initial conditions path/filename
(ignored if Restart=TRUE)

Boundary_Conditions

Character input master grid lateral boundary conditions
path/filename

Top_Concentrations

Character input master grid top boundary conditions
path/filename

Point_Sources

Character input elevated point source emissions path/filename
(ignored if Point_Emissions=FALSE)

Master_Grid_Restart

Character input master grid restart path/filename (ignored if
Restart=FALSE)

Nested_Grid_Restart

Character input nested grid restart path/filename (ignored if
Restart=FALSE or Number_of_Grids=1)

PiG_Restart

Character input PiG restart path/filename (ignored if
Restart=FALSE or PiG_Submodel=FALSE)

Srfmod_Grid

Character array (by grid) input surface model restart
path/filename (ignored if Restart=FALSE or
Surface_Model=FALSE)

Surface_Grid

Character array (by grid) input static 2D surface path/filename
(optional for nested grids)

Met2D_Grid

Character array (by grid) input time‐variant 2D surface
meteorology path/filename (optional for nested grids)

Met3D_Grid

Character array (by grid) input time‐variant 3D meteorology
path/filename (optional for nested grids)

Vdiff_Grid

Character array (by grid) input time‐variant 3D vertical
diffusivity path/filename (optional for nested grids)

Cloud_Grid

Character array (by grid) input time‐variant 3D cloud/rain
path/filename (optional but required if Wet_Deposition=TRUE,
optional for nested grids)

Emiss_Grid

Character array (by grid) input gridded emissions path/filename
(ignored if Gridded_Emissions=FALSE, optional for nested grids)

If CAMx cannot find or open a non‐blank input filename provided in the run control file, the
model will stop with an error. CAMx will accept blank input filenames for only those files that
are optional.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

19

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
2. The CAMx Modeling System

Table 2‐2. CAMx output file suffixes and their corresponding file types.
Suffix

CAMx File Type

.out

Text simulation tracking file (CPU, input files read, error/warning messages)

.diag

Text simulation diagnostic file (repeat of run control inputs, PiG diagnostics,
miscellaneous diagnostic output)

.mass

Text mass budget file for subsequent postprocessing

.inst

Fortran binary master grid 3‐D instantaneous concentration file at the end
of the simulation (used for restarts)

.finst

Fortran binary nested grid 3‐D instantaneous concentration file at the end
of the simulation (used for restarts)

.pig

Fortran binary PiG sub‐model file (used for restarts)
Standard CAMx Output Option

.avrg.grdnn

Fortran binary average concentration file for grid nn; optionally contains 2‐D
layer 1 concentration field or full 3‐D concentration field

.depn.grdnn

Fortran binary 2‐D surface deposition file for grid nn

.srf.grdnn

Fortran binary 2‐D surface model mass file for grid nn (optional)

.smpnn

Fortran binary 2‐D layer 1 average concentration file for PiG sampling grid
nn (optional)

2.3.2 Using Scripts to Run CAMx
The generation of the run control file is most easily accomplished in the job script that actually
runs the model; Figure 2‐2 shows an example of a CAMx job script that builds a “CAMx.in” file
and runs the model for each day to be simulated. Alternatively, the run control file could be
written separately with a name specific to a given simulation, then linked or copied to the
standard “CAMx.in” filename before the model is executed at a command line or in a job
script.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

20

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
2. The CAMx Modeling System

#!/bin/csh
#
# CAMx 6.30
#
setenv OMP_NUM_THREADS 4
setenv MPSTKZ 128M
limit stacksize unlimited
#
set EXEC
= "../../src/CAMx.v6.30.noMPI.pg_linuxomp"
#
set RUN
= "v6.30.midwest.36.12.noMPI"
set INPUT
= "../inputs"
set MET
= "../inputs/met"
set EMIS
= "../emiss"
set PTSRCE = "../ptsrce"
set OUTPUT = "../outputs"
#
mkdir -p $OUTPUT
#
# --- set the dates and times ---#
set RESTART = "NO"
foreach today (03.154 04.155)
set JUL = $today:e
set CAL = $today:r
set YESTERDAY = `echo ${CAL} | awk '{printf("%2.2d",$1-1)}'`
#
if( ${RESTART} == "NO" ) then
set RESTART = "false"
else
set RESTART = "true"
endif
#
# --- Create the input file (always called CAMx.in)
#
cat << ieof > CAMx.in
&CAMx_Control
Run_Message

= 'CAMx 6.30 Test Problem -- Mech6 CF CB05 $RUN',

!--- Model clock control --Time_Zone
Restart
Start_Date_Hour
End_Date_Hour

=
=
=
=

0,
.${RESTART}.,
2002,06,${CAL},0000,
2002,06,${CAL},2400,

Maximum_Timestep
Met_Input_Frequency
Ems_Input_Frequency
Output_Frequency

=
=
=
=

15.,
60.,
60.,
60.,

!
!
!
!

! (0=UTC,5=EST,6=CST,7=MST,8=PST)
! (YYYY,MM,DD,HHmm)
! (YYYY,MM,DD,HHmm)
minutes
minutes
minutes
minutes

Figure 2‐2. A sample CAMx job script that generates a “CAMx.in” file and runs the model
with OMP parallelization.
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

21

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
2. The CAMx Modeling System

!--- Map projection parameters --Map_Projection
Longitude_Pole
Latitude_Pole
True_Latitude1
True_Latitude2

= 'LAMBERT', ! (LAMBERT,POLAR,UTM,LATLON)
= -97.,
! deg (west<0,south<0)
= 40.,
! deg (west<0,south<0)
= 45.,
! deg (west<0,south<0)
= 33.,
! deg (west<0,south<0)

!--- Parameters for the master (first) grid --Number_of_Grids
Master_SW_XCoord
Master_SW_YCoord
Master_Cell_XSize
Master_Cell_YSize
Master_Grid_Columns
Master_Grid_Rows
Number_of_Layers

=
=
=
=
=
=
=
=

2,
-792.,
-1656.,
36.,
36.,
68,
68,
16,

!
!
!
!

km
km
km
km

or
or
or
or

deg, SW corner of cell(1,1)
deg, SW corner of cell (1,1)
deg
deg

!--- Parameters for the second grid --Nest_Meshing_Factor(2)
Nest_Beg_I_Index(2)
Nest_End_I_Index(2)
Nest_Beg_J_Index(2)
Nest_End_J_Index(2)

=
=
=
=
=

3,
22,
51,
22,
58,

=
=
=
=
=
=
=
=
=
=
=
=
=
=

.false.,
'PPM',
'EBI',
'None',
'None',
.true.,
‘WESELY89’,
.true.,
.false.,
.false.,
.true.,
.true.,
.true.,
.true.,

!
!
!
!
!

Cell size relative
Relative to master
Relative to master
Relative to master
Relative to master

to master grid
grid
grid
grid
grid

!--- Model options --Diagnostic_Error_Check
Advection_Solver
Chemistry_Solver
PiG_Submodel
Probing_Tool
Chemistry
Drydep_Model
Wet_Deposition
ACM2_Diffusion
Surface_Model
Super_Stepping
Gridded_Emissions
Point_Emissions
Ignore_Emission_Dates

!
!
!
!
!

True = will stop after 1st timestep
(PPM,BOTT)
(EBI,LSODE)
(None,GREASD,IRON)
(None,SA,DDM,HDDM,PA,IPR,IRR,RTRAC,RTCMC)

! (None, WESELY89, ZHANG03)

!--- Output specifications --Root_Output_Name
= '$OUTPUT/CAMx.$RUN.200206${CAL}',
Average_Output_3D
= .false.,
Output_Species_Names(1)
= 'NO',
Output_Species_Names(2)
= 'NO2',
Output_Species_Names(3)
= 'O3',
Output_Species_Names(4)
= 'SO2',
Output_Species_Names(5)
= 'H2O2',
Output_Species_Names(6)
= 'HNO3',
Output_Species_Names(7)
= 'NH3',
Output_Species_Names(8)
= 'PNO3',
Output_Species_Names(9)
= 'PSO4',
Output_Species_Names(10) = 'PNH4',
Output_Species_Names(11) = 'POA',

Figure 2‐2 (continued).
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

22

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
2. The CAMx Modeling System

Output_Species_Names(12)
Output_Species_Names(13)
Output_Species_Names(14)
Output_Species_Names(15)
Output_Species_Names(16)
Output_Species_Names(17)
Output_Species_Names(18)
Output_Species_Names(19)
Output_Species_Names(20)
Output_Species_Names(21)

=
=
=
=
=
=
=
=
=
=

'PEC',
'FPRM',
'CPRM',
'CCRS',
'FCRS',
'SOA1',
'SOA2',
'SOA3',
'SOA4',
'SOA5',

!--- Input files --Chemistry_Parameters
Photolysis_Rates
Ozone_Column
Initial_Conditions
Boundary_Conditions
Top_Concentrations
Point_Sources
Master_Grid_Restart
Nested_Grid_Restart
PiG_Restart
Srfmod_Grid(1)
Srfmod_Grid(2)
Flexi_Nest
Emiss_Grid(1)
Surface_Grid(1)
Met2D_Grid(1)
Met3D_Grid(1)
Vdiff_Grid(1)
Cloud_Grid(1)
Emiss_Grid(2)
Surface_Grid(2)
Met2D_Grid(2)
Met3D_Grid(2)
Vdiff_Grid(2)
Cloud_Grid(2)

=
=
=
=
=
=
=
=
=
=
=
=
=

=
=
=
=
=
=
=
=
=
=
=
=

'$INPUT/CAMx6.3.chemparam.6_CF',
'$INPUT/tuv.200206.STL.txt',
'$INPUT/o3col.200206.STL_36_68X68_12_92X113.txt',
'$INPUT/IC.vistas_2002gt2a_STL_36_68X68_16L.2002081',
'$INPUT/BC.vistas_2002gt2a_STL_36_68X68_16L.2002${JUL}',
' ',
'$PTSRCE/ptsrce.stl.36km.2002${JUL}.a0.bin',
'$OUTPUT/CAMx.$RUN.200206${YESTERDAY}.inst',
'$OUTPUT/CAMx.$RUN.200206${YESTERDAY}.finst',
' ',
' ',
' ',

.false.
'$EMIS/emiss.stl.36km.200206${CAL}.a1.bin',
'$INPUT/met/camx.lu.36k.bin',
'$INPUT/met/camx.2d.200206${CAL}.36k.bin',
'$INPUT/met/camx.3d.200206${CAL}.36k.bin',
'$INPUT/met/camx.kv.200206${CAL}.36k.bin',
'$INPUT/met/camx.cr.200206${CAL}.36k.bin',
'$EMIS/emiss.stl.12kmsmall.200206${CAL}.a1.bin',
'$INPUT/met/camx.lu.12ksmall.bin',
'$INPUT/met/camx.2d.200206${CAL}.12ksmall.bin',
'$INPUT/met/camx.3d.200206${CAL}.12ksmall.bin',
'$INPUT/met/camx.kv.200206${CAL}.12ksmall.bin',
'$INPUT/met/camx.cr.200206${CAL}.12ksmall.bin',

/
!------------------------------------------------------------------------------ieof
#
# --- Execute the model --#
if( ! { $EXEC } ) then
exit
endif
end

Figure 2‐2 (concluded).
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

23

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
2. The CAMx Modeling System

2.4 Benchmarking Model Run Times
Overall model speed and OMP/MPI scalability depends on several factors, including the number
and sizes of grids; the choice of chemistry mechanism; the number of point sources to be
treated with PiG and total PiG puffs that accumulate during a run; and the use of Probing Tools.
Parallelization is most advantageous for larger applications in which overhead processes (e.g.,
model setup, I/O, etc.) are a much smaller fraction of total model run time. In other words,
CAMx applications on multiple expansive grids, employing PiG, and/or including Probing Tools
would scale most effectively and thus benefit most from parallelization.
A set of systematic run time tests were recently conducted by the Lake Michigan Air Directors
Consortium (LADCo) using CAMx v6.20 with various combinations of OMP and MPI
parallelization. Note that model speed has been improved somewhat since v6.20; however,
results from LADCo’s test are informative. LADCo’s CAMx configuration was relatively simple,
employing a single large grid covering the entire US with 12 km grid spacing (396246, 25
layers), and using CB6r2 photochemistry with the CF PM treatment. PiG and Probing Tools
were not active. CAMx was compiled using Portland Group v15.7‐0 with OMP and MPICH
v3.1.4. CAMx was run on a 2.60 Ghz Intel Xeon chipsets with 12 physical cores (24 cores hyper‐
threaded). Run time results are shown in Table 2‐3.

Table 2‐3. CAMx v6.20 speed performance with MPI and OMP parallelization from the LADCo
tests described above.
OMP Threads
12
12
8
6
4
3
2
1

MPI Slices
0
2
3
4
6
8
12
24

Total Cores
12
24
24
24
24
24
24
24

Hours/Sim Day
4:05:12
3:30:47
2:59:08
2:36:10
2:35:17
2:27:34
2:25:25
3:06:57

2.5 CAMx Pre‐ And Post‐Processors
This section describes several important CAMx pre‐ and post‐processors that we make available
to the user community. Like CAMx itself, these programs are written in Fortran and distributed
as free software under the terms of the GNU General Public License. Each come with README
files, makefiles, and sample job scripts that document their purpose and usage. Ramboll
Environ occasionally posts updates for certain widely‐used programs when necessary, but does
not actively support or maintain every one. Users can e‐mail questions, comments, suggestions
or improvements to ask‐camx@environ.org.
2.5.1 Emissions
Certain emission models (those shown in Figure 2‐1) can provide speciated, temporally‐
allocated, gridded and point source emission input files in the CAMx‐ready format. Further
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

24

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
2. The CAMx Modeling System

processing may be required, however, to select point sources for the PiG treatment, translate
the point source files from a common text format to CAMx‐ready binary files, or to refine the
domain size/resolution for gridded emissions.
PIGSET: This program allows the user to select and set certain point sources for the Plume‐in‐
Grid (PiG) treatment in a CAMx simulation. It also converts text point source files
commonly generated by emission models such as SMOKE, CONCEPT, and EPS3 to
CAMx‐ready binary format. See the source code for more information, and the sample
job for usage. Also see Section 6 for guidance in selecting PiG point sources and
manipulating day‐specific point source files.
WINDOW: This program is used to “window” out a sub‐section of the surface emissions grid for
use on a smaller CAMx grid. It can also be used to aggregate or distribute surface
emissions to coarser or finer resolution, respectively. See the sample job for usage.
SEASALT: This program generates aerosol emissions of sodium and chloride, and gaseous
emissions of chlorine and halo‐methane compounds, using CAMx‐ready meteorological
and landuse files. A separate merging program is included that allows sea salt
emissions to be merged in with pre‐existing CAMx‐ready gridded emission files.
PREPVBS: This program converts the organic compound emissions (VOC precursors and POA)
prepared for the CAMx CF aerosol scheme to those compatible with the VBS scheme so
that the user can employ the VBS scheme without having to develop emission inputs for
the scheme from scratch. However, this approach should be used with caution because
significant uncertainties exist in the VBS emissions estimated by this. See the sample job
for usage.
REGNMAP: This program supports the development of source apportionment fractional
(partial) region maps with which to allocate gridded emissions to source regions. It
reads SMOKE spatial allocation reports for a specific modeling grid and source category
(or group of categories), extracts emissions data by grid cell and state/county FIPS code,
and generates a new CAMx input file that defines a fractional region map for that grid
and source category/group.
2.5.2 Meteorology
The recommended approach to develop meteorological inputs for CAMx is through the use of
prognostic meteorological models. Ramboll Environ distributes interface programs for three
specific models: WRF, MM5, and RAMS; This does not necessarily preclude other
meteorological models to be used, but users will need to develop interface programs on their
own.
WRFCAMx: This program generates CAMx v6 meteorological input files from WRF (ARW core)
v3 output files. See the README in the archive for a description of the program and
how it is applied. You will need NetCDF libraries to compile and run this program.
MM5CAMx: This program generates CAMx v6 meteorological input files from MM5 v3 output
files. See the README in the archive for a description of the program and how it is
applied.
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

25

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
2. The CAMx Modeling System

RAMSCAMx: This program generates CAMx v6 meteorological input files from RAMS v4/5/6
output files. See the README in the archive for a description of the program and how it
is applied. You will need HDF5 libaries to compile and run this program.
METCONVERT: This program converts older meteorological files from CAMx v4 and v5 to the
CAMx format introduced with CAMx v6.
KVPATCH: This program applies minimum limits on vertical diffusivity (Kv) within a user‐defined
surface layer depth based on an input landuse grid. It optionally allows Kv profiles to be
extended into daytime boundary‐layer capping convection as defined by input
cloud/rain files. See the source code for more information. Use of this program to
adjust Kv inputs is entirely optional.
2.5.3 Photolysis Rates
The development of photolysis rate inputs for CAMx is crucial for the photochemical
mechanisms, but is not needed for inert or simple chemistry (e.g., Mechanism 10) applications.
Two programs are available to assist the user in developing photolysis and ozone column input
files.
O3MAP: This program prepares ozone column input files for CAMx, and must be run prior to
running the TUV model as it defines the atmospheric ozone column intervals based on
input data. Ozone column data files (http://ozoneaq.gsfc.nasa.gov/data/ozone or
ftp://toms.gsfc.nasa.gov/pub/omi/data/) in latitude/longitude text format must be
supplied as input. O3MAP attempts to fill data gaps in day‐specific ozone column files
(Figure 2‐3) with an average determined from valid data processed for the extraction
domain. Alternatively, you may use monthly‐average ozone column files (no data
gaps). See the Readme file and job script in the archive for usage.

Figure 2‐3. An example of global ozone column from the Ozone Monitoring Instrument (OMI)
platform. White areas denote missing data. From ftp://toms.gsfc.nasa.gov/pub/omi/data/.
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

26

www.camx.com

March 2016

TUV:

CAMx User’s Guide Version 6.3
2. The CAMx Modeling System

This is a radiative transfer model that develops clear‐sky photolysis rate inputs for all
CAMx photochemical mechanisms. TUV is developed and distributed by NCAR (2011).
The program specifies default intervals for 5 surface UV albedos, 3 terrain heights, 11
altitudes above ground, 10 solar zenith angles, and 5 atmospheric ozone column
intervals (from O3MAP). See the sample job in the archive for usage.

2.5.4 Initial and Boundary Conditions
Ramboll Environ provides a few programs to develop initial and boundary conditions, but there
are many ways to generate these important inputs. The most common approach involves
“down‐scaling” (or extracting) the output from larger‐scale (e.g., global) models to the CAMx
domain. Users will need to develop their own programs if they choose an alternative
methodology or source of data to generate initial/boundary conditions.
ICBCPREP: This program prepares simple CAMx initial and lateral boundary condition files.
Values are constant in space and time, but unique values may be specified for each
chemical species to be modeled; they are defined in a text file. See the sample job in
the archive for usage.
GEOS2CAMx: This program generates CAMx initial, lateral boundary, and (optionally) top
boundary condition input files from GEOS‐Chem global model output. See the Release
Notes and job scripts in the archive for a description of the program and how it is
applied.
MOZART2CAMx: This program generates CAMx and CMAQ initial and lateral boundary
condition input files from MOZART4 output. This program will also process AM3
datasets if output is first translated to a geodetic (latitude/longitude) grid,
concentrations are provided as volume mixing ratio, and all needed state variables are
available. See the README in the archive for a description of the program and how it is
applied. You will need I/O‐API and NetCDF libraries to compile and run this program.
2.5.5 Landuse
Approaches for developing landuse/landcover inputs for CAMx include: (1) translating gridded
spatial allocation surrogates developed during emissions processing into the CAMx categories
described in Section 3; (2) translating the gridded landuse/landcover fields from the
meteorological model; or (3) separately developing landuse input fields from raw data (such as
from USGS, MODIS or NLCD) using GIS or other programs. Ramboll Environ distributes
meteorological interface programs (described above) that translate the meteorological model
landuse/landcover fields to the CAMx definitions and grid configuration.
MERGE_LULAI: This program merges independently‐developed landuse and/or LAI fields (for
example via GIS processing of common terrestrial datasets) with an existing CAMx 2D
surface file generated by the meteorological interface programs. See the sample job
and source code for a description of the program and how it is applied.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

27

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
2. The CAMx Modeling System

2.5.6 Post‐processors
Most of the CAMx post‐processors distributed by Ramboll Environ provide some manner of
concentration file manipulation, either to extract certain information from the raw output files,
to re‐format for use in other programs and applications, to concatenate files, etc. A few others
are distributed to assist in evaluating Probing Tool output. Ramboll Environ does not advocate
or support any particular visualization or graphics software.
AVGDIF: This program is used to compare two CAMx format output average files and print a
table of differences. This is useful for checking differences between CAMx runs on
different machines or different compile options for the test case. See the sample job in
the archive for usage.
BNDEXTR: Use this program to extract boundary conditions for a nested grid when you apply
CAMx in one‐way nesting mode. One‐way nesting means that CAMx is run successively
for each grid, with BNDEXTR as the interface between each run. This program is not
needed when CAMx is run in the more standard two‐way nesting mode, where all grids
are run in a single simulation. See the example job for usage.
CAMxPOST: This is a suite of post‐processing utilities designed to facilitate the evaluation of
model performance. It is used to combine observations and predictions, calculate
statistics, and plot time series. See the README file in the archive for usage.
CAMxTRCT: This program extracts a single chemical species for specified grids from the output
average concentration and deposition files, and from input emission files. Output from
this program can be written in the standard CAMx format, or alternatively to a text
format in Surfer® “GRD” format for subsequent plotting. It also has the capability to
convert units and combine species to yield certain hard‐coded bulk compounds like
NOx and VOC. See the sample job in the archive for usage.
CAMx2IOAPI: This program converts CAMx output average concentration and deposition files
to I/O‐API format. You will need I/O‐API and NetCDF libraries to compile and run this
program. This program allows you to use various third‐party manipulation and
visualization software that handle I/O‐API and NetCDF formats.
PA_Tools: This is a suite of post‐processing utilities designed to extract IPR, IRR, and CPA data
from CAMx Process Analysis output files and reformat the data suitable for subsequent
analysis (e.g. using spreadsheets).
XSPCMAP: Similar to CAMxTRCT yet more flexible, this program extracts any number of specific
chemical species or user‐defined combinations of species for specified grids from the
output average concentration and deposition files, and writes results to a new file in
CAMx format. See the sample job and species mapping table in the archive for usage.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

28

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
3. Core Model Input / Output Structures

3. CORE MODEL INPUT/OUTPUT STRUCTURES
Most CAMx input/output (I/O) files are Fortran binary and based on the Urban Airshed Model
(UAM) convention. This allows the user to employ widely available software designed
specifically for these formats to develop the input files or to post‐process and visualize the
output files.
CAMx requires input files that define the chemical mechanism and describe the photochemical
conditions, surface characteristics, initial/boundary conditions, emission rates, and various
meteorological fields over the entire modeling domain. Table 3‐1 summarizes the input data
requirements of CAMx. Preparing this information requires several preprocessing steps to
translate “raw” emissions, meteorological, air quality and other data into final input files for
CAMx. Prognostic meteorological models are used to generate all of the required time varying
three‐dimensional meteorological fields.

Table 3‐1. Data requirements of CAMx.
Data Types


Meteorology
Supplied by Prognostic Meteorological Models

Air Quality
Developed from Other Models or Measurement Data

Emissions
Supplied by Emissions Models and Processors









Geographic
Developed from Terrain, Landuse/Landcover, and
Vegetation Density Datasets
Photolysis
Derived from Satellite Measurements and Radiative
Transfer Models

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS



29

Data Fields
3‐Dimensional Gridded Fields:
‐ Vertical Grid Structure
‐ Horizontal Wind Components
‐ Temperature
‐ Pressure
‐ Water Vapor
‐ Vertical Diffusivity
‐ Clouds/Precipitation (optional)
‐ Snow Cover (optional)
Gridded Initial Concentrations
Gridded Lateral Boundary Concentrations
Gridded Top Boundary Concentrations (optional)
Elevated Point Sources (optional), e.g.:
‐ Industrial Facilities
‐ Prescribed, Agricultural, Wild Fires
‐ Lightning NOx
Combined Gridded Sources (optional), e.g.:
‐ Low‐Level Point
‐ On‐Road and Non‐Road Mobile
‐ Area
‐ Biogenic
‐ Oceanic
Gridded Surface Characteristics
‐ Landuse/LandCover
‐ Terrain Elevation (optional)
‐ Leaf Area Index (LAI; optional)
‐ Land/Ocean Mask (optional)
Atmospheric Radiative Properties
‐ Gridded Ozone Column Codes
‐ Photolysis Rates Lookup Table

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
3. Core Model Input / Output Structures

CAMx produces gridded time‐averaged concentration output files; the user selects the time
interval (usually hourly), the species to be output, and whether the output contains just two‐
dimensional surface layer fields or entire three‐dimensional fields. A separate average output
file is written for each grid employed in the simulation. Two‐dimensional surface deposition
fields for the same user‐selected species are also written to output files with the same structure
as the average concentration files. Gridded three‐dimensional instantaneous concentrations of
all species on all grids are written at the end of the simulation to allow for a model restart. The
CAMx Probing Tool options provide their own information in separate output files in the same
CAMx output format. Diagnostic output files include three files that track computer resources,
echo input selections, provide mass budget and diagnostic summaries, and provide
error/warning messages.

3.1 CAMx Chemistry Parameters File
Chemistry parameters are provided in a text file that specifies the chemical mechanism be used
and associated details on species properties and reaction types and rates. The chemistry
parameters file format is defined in Table 3‐2, and samples are given in Figure 3‐1. Some
records in this file are labeled (columns 1‐20) to indicate the type of information to supply on
that line (starting in column 21). For records that are not labeled, data start in column 1. Some
chemistry parameter records are optional, depending upon the logical flags indicated for such
records, and are shown in Table 3‐2 by a check in the third column; if the indicated option is not
invoked these records should not appear in the file. The first record of the chemistry
parameters file must contain the string “VERSION6.3”, which indicates that the file is specific
to this version of CAMx.
If the chemistry flag is set “true” on the CAMx.in file, CAMx checks that certain properties of
the selected mechanism are consistent with parameters supplied on the input file (e.g., number
of reactions, photolysis reactions and species). If any discrepancies are found, they are
reported in the output message file and the simulation is halted. The user may also specify an
inert simulation by setting the chemistry flag to “false”. In this case, any number of arbitrarily
named species may be listed, and chemistry reaction parameters are ignored.
Gas‐phase chemistry is selected by a Mechanism ID assigned to each photochemical
mechanism (see Section 5). Aerosol chemistry is selected by the keywords “NONE”, “INERT”,
“CF” or “CMU”. In the “INERT” case, the user can define any number of arbitrary particulate
names and sizes. The “CF” and “CMU” options invoke aerosol chemistry and treat aerosol size
using either static coarse and fine modes (CF) or an evolving size section model (CMU). Both CF
and CMU options require a minimum set of specific aerosol names with associated chemistry.
The chemistry parameters file controls how photolysis rates are calculated in CAMx. So‐called
“primary” photolysis rates are input to CAMx via the photolysis rates file. The primary
photolysis reactions are identified by number in the chemistry parameters file and the
photolysis rates file must match this declaration. So‐called “secondary” photolysis rates are set
by scaling factors to one of the primary reactions. Use of secondary rates requires at least one
primary photolysis reaction.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

30

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
3. Core Model Input / Output Structures

Table 3‐2. Description of the CAMx chemistry parameters file. The record labels exist in columns 1‐20, and where given, the
input data for that record start in column 21. The format denoted “list” indicates a free‐format list of numbers (comma or space‐
delimited).
Record Label
(columns 1‐20)
CAMx version
Mechanism ID
Aerosol Option
Description
No of gas species

Record
Optional

Format
A
list
A
A
list

No of aero species

list

No of reactions

list

Prim photo rxns

list

No of sec photo rxn

list

ID, prim ID, scale



SrfMod #spc, #rxns

list
list

Species Records
Gas Spec ...
5X,
A10,
E10.0,
E10.0,
F10.0,
F10.0,
F10.0
F10.0

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

Description
Model version keyword (VERSION6.3)
ID number of chemical mechanism (see Table 5‐1)
Keyword for aerosol scheme (NONE, INERT, CF, or CMU)1
Message record to describe this file
Number of radical and state gas species (NGAS ≥ 1)
Number of aerosol species (NAERO ≥ 0)
Aerosol chemistry timestep (min) (if NAERO > 0)
Number of size bins (NBIN ≥ 1) (if NAERO > 0)2
Aerosol diameter (m) for NBIN+1 cut points (if NAERO > 0)
Number of reactions (NREACT ≥ 0)
Number of primary photolysis reactions (NPHOT1 ≥ 0)
List of primary photolysis reaction ID numbers (must match the photolysis rates input file)
Number of secondary photolysis reactions (NPHOT2 ≥ 0)
If NPHOT2 > 0, repeat this record for each secondary photolysis reaction
ID number of the secondary photolysis reaction
ID number of the primary photolysis reaction used for scaling
Secondary reaction scale factor
Number of Surface Model species and reactions (see Section 4.8)
Set to 0,0 if not using the Surface Model
Heading
Heading
Repeat this record for each gas species (start in column 1)
Gas species name (radicals first, followed by state species)
Lower bound concentration (ppm)
Henry’s law constant (M/atm)
Henry’s law temperature dependence (K)
Molecular weight (g/mol)
Wesley’s reactivity parameter
Surface resistance scaling factor for strong acids (0‐1)

31

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
3. Core Model Input / Output Structures

Table 3‐2 (continued). Description of the CAMx chemistry parameters file. The record labels exist in columns 1‐20, and where
given, the input data for that record start in column 21. The format denoted “list” indicates a free‐format list of numbers (comma
or space‐delimited).
Record Label
(columns 1‐20)
Aero Spec ...

Record
Optional




Reaction Records
Rxn Typ Param ...

Format
5X,
A10,
E10.0,
F10.0
F10.0
I10
F10.0
I10





list

Description
Heading (if NAERO > 0)
If NAERO > 0, repeat this record for each aerosol species (start in column 1)
Aerosol species name
Lower bound concentration (g/m3)
Species density (g/cm3)
Dry extinction efficiency (m2/m)
Hygroscopic extinction adjustment (0 = no adjustment, 1 = RH‐dependent)
Single‐scattering albedo
Assigned size bin (INERT and CF aerosols only – references cut points in record 6)
Heading (if NREACT > 0)
Heading (if NREACT > 0)
If NREACT > 0, repeat this record for each gas‐phase reaction (start in column 1)
Reaction ID number
Rate constant expression ID number (1‐7, as shown in Table 3‐3)
Rate constant parameters (depending on expression type in Table 3‐3). For reactions identified as
photolysis reactions above, the rate constant is not used and is customarily set to zero.

1 NONE = gas‐phase chemistry only; INERT = user‐defined inert PM species; CF = Coarse/Fine aerosol chemistry scheme; CMU = multi‐sectional aerosol chemistry scheme;.
2 For the CF scheme, NBIN must be set to 2, and the user specifies the coarse/fine size ranges.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

32

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
3. Core Model Input / Output Structures

CAMx Version
|VERSION6.3
Mechanism ID
|6
Aerosol Option
|CF
Description
|CB05 plus PM (CF) with Hg: PNA rates set to 0; molecular units
No of gas species |68
#aero, dt, bins
|25 15.0 2 0.039 2.5 10.0
No of reactions
|156
Prim photo rxns
|19 1 8 9 14 15 25 36 51 52 53 62 65 75 76 87 91 102 143 152
No of sec photo rxn|4
ID, prim ID, scale |72 65 1.0
|97 65 0.0
|106 91 1.0
|138 75 9.0
SrfMod #spc, #rxns |0 0
Species Records
Gas Spec
lower bnd
H-law
T-fact
Molwt Reactvty
Rscale
1 O1D
1.00E-15 1.00E+00
0.
0.0
0.0
0.
2 O
1.00E-15 1.00E+00
0.
0.0
0.0
0.
3 OH
1.00E-15 1.00E+00
0.
0.0
0.0
0.
4 HO2
1.00E-15 1.00E+00
0.
0.0
0.0
0.
5 C2O3
1.00E-15 1.00E+00
0.
0.0
0.0
0.
6 XO2
1.00E-15 1.00E+00
0.
0.0
0.0
0.
7 XO2N
1.00E-15 1.00E+00
0.
0.0
0.0
0.
8 CXO3
1.00E-15 1.00E+00
0.
0.0
0.0
0.
9 MEO2
1.00E-15 1.00E+00
0.
0.0
0.0
0.
10 TO2
1.00E-15 1.00E+00
0.
0.0
0.0
0.
11 ROR
1.00E-15 1.00E+00
0.
0.0
0.0
0.
12 HCO3
1.00E-15 1.00E+00
0.
0.0
0.0
0.
13 CRO
1.00E-15 1.00E+00
0.
0.0
0.0
0.
14 AACD
1.00E-12 5.00E+03
-4000.
60.0
1.0
1.
15 ALD2
1.00E-12 6.30E+03
-6492.
44.0
1.0
1.
16 ALDX
1.00E-12 6.30E+03
-6492.
58.1
1.0
1.
17 CO
1.00E-04 1.00E-10
0.
28.0
0.0
1.
18 CRES
1.00E-12 2.70E+03
-6492.
108.1
1.0
1.
19 ETH
1.00E-12 1.00E-02
-4000.
28.0
0.0
1.
20 ETHA
1.00E-04 1.73E-03
-4000.
30.1
0.0
1.
21 ETOH
1.00E-12 2.20E+02
-4932.
46.1
1.0
1.
22 FACD
1.00E-12 5.68E+03
-6060.
46.0
1.0
1.
23 FORM
1.00E-12 6.30E+03
-6492.
30.0
1.0
1.
24 H2O2
1.00E-12 7.40E+04
-6643.
34.0
1.0
0.
25 HNO3
1.00E-12 2.10E+05
-8707.
63.0
0.0
0.
26 HONO
1.00E-12 5.90E+01
-4781.
47.0
1.0
1.
27 IOLE
1.00E-12 5.00E-03
-4000.
56.1
0.0
1.
28 ISOP
1.00E-12 1.00E-02
-4000.
68.1
0.0
1.
29 ISPD
1.00E-12 6.30E+03
-6492.
70.1
1.0
1.
30 MEOH
1.00E-12 2.20E+02
-4932.
32.0
1.0
1.
31 MEPX
1.00E-12 3.05E+02
-5250.
48.0
0.8
0.
32 MGLY
1.00E-12 2.70E+03
-6492.
72.0
1.0
1.
33 N2O5
1.00E-12 1.00E+05
-4000.
108.0
0.1
0.
34 NO
1.00E-09 1.90E-03
-1480.
30.0
0.0
1.
35 NO2
1.00E-12 1.00E-02
-2516.
46.0
0.8
1.
36 NO3
1.00E-15 1.00E+05
-4000.
62.0
0.1
0.
37 NTR
1.00E-12 9.40E+03
-8706.
119.1
0.0
1.
38 O3
1.00E-12 1.10E-02
-2415.
48.0
1.0
1.
39 OLE
1.00E-12 5.00E-03
-4000.
42.1
0.0
1.
40 OPEN
1.00E-12 2.70E+03
-6492.
84.0
1.0
1.
41 PACD
1.00E-12 5.00E+03
-4000.
76.0
1.0
1.
42 PAN
1.00E-12 3.60E+00
-5910.
121.0
0.6
1.
43 PANX
1.00E-12 3.60E+00
-5910.
135.0
0.6
1.
44 PAR
1.00E-04 1.00E-03
-4000.
72.1
0.0
1.
45 PNA
1.00E-12 2.00E+04
-5910.
79.0
1.0
1.

Figure 3‐1a. Example CAMx chemistry parameters file for Mechanism 6 (CB05) with CF PM
scheme that includes the mercury species HG0, HG2, and HGP.
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS
www.camx.com
33

March 2016

CAMx User’s Guide Version 6.3
3. Core Model Input / Output Structures

46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68

ROOH
1.00E-12 1.00E+02
-4000.
90.1
SO2
1.00E-09 1.22E+00
-3156.
64.0
SULF
1.00E-12 1.00E+10
0.
98.0
TERP
1.00E-12 4.90E-02
-4000.
136.2
TOL
1.00E-12 1.20E+00
-4000.
92.1
XYL
1.00E-12 1.40E+00
-4000.
106.2
NH3
1.00E-09 5.76E+01
-4100.
17.0
HCL
1.00E-12 1.00E+05
-4000.
36.5
TOLA
1.00E-12 1.20E+00
-4000.
92.0
XYLA
1.00E-12 1.40E+00
-4000.
106.0
BNZA
1.00E-12 1.80E-01
-4000.
78.0
ISP
1.00E-12 1.00E-02
-4000.
68.0
TRP
1.00E-12 4.90E-02
-4000.
136.0
SQT
1.00E-12 4.90E-02
-4000.
204.0
CG1
1.00E-12 1.00E+05
-4000.
150.0
CG2
1.00E-12 1.00E+05
-4000.
150.0
CG3
1.00E-12 1.00E+05
-4000.
130.0
CG4
1.00E-12 1.00E+05
-4000.
130.0
CG5
1.00E-12 1.00E+05
-4000.
180.0
CG6
1.00E-12 1.00E+05
-4000.
180.0
CG7
1.00E-12 1.00E+05
-4000.
210.0
HG0
1.00E-12 1.11E-01
-4970.
200.6
HG2
1.00E-12 2.00E+05
-4000.
253.1
Aero Spec lower bnd
Density Dry Bext RH Adjust
1 PNO3
1.00E-09
1.5
7.0
1
2 PSO4
1.00E-09
1.5
7.0
1
3 PNH4
1.00E-09
1.5
7.0
1
4 POA
1.00E-09
1.0
7.0
0
5 SOA1
1.00E-09
1.0
7.0
0
6 SOA2
1.00E-09
1.0
7.0
0
7 SOA3
1.00E-09
1.0
7.0
0
8 SOA4
1.00E-09
1.0
7.0
0
9 SOA5
1.00E-09
1.0
7.0
0
10 SOA6
1.00E-09
1.0
7.0
0
11 SOA7
1.00E-09
1.0
7.0
0
12 SOAH
1.00E-09
1.0
7.0
0
13 SOPA
1.00E-09
1.0
7.0
0
14 SOPB
1.00E-09
1.0
7.0
0
15 PEC
1.00E-09
2.0
18.0
0
16 FPRM
1.00E-09
3.0
0.4
0
17 FCRS
1.00E-09
3.0
0.4
0
18 CPRM
1.00E-09
3.0
0.4
0
19 CCRS
1.00E-09
3.0
0.4
0
20 NA
1.00E-09
2.0
1.5
1
21 PCL
1.00E-09
2.0
1.5
1
22 PH2O
1.00E-09
1.0
0.0
0
23 HGP
1.00E-15
8.0
0.0
0
24 HGIIP
1.00E-20
8.0
0.0
0
25 HGIIPC
1.00E-20
8.0
0.0
0
Reaction Records
Rxn Typ Order Parameters (1 to 12, depending upon Typ)
1
1 1 0.000E+00
2
3 3 6.000E-34
0.0
-2.40
300.0
3
3 2 3.000E-12
1500.0
0.00
300.0
4
3 2 5.600E-12
-180.0
0.00
300.0
5
4 2 2.500E-31
0.0
-1.80
300.0
0.70
300.0
0.60
1.00
6
4 2 9.000E-32
0.0
-1.50
300.0
300.0
0.60
1.00
7
3 2 1.200E-13
2450.0
0.00
300.0
8
1 1 0.000E+00

0.8
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
SSA
0.99
0.99
0.99
0.80
0.80
0.80
0.80
0.80
0.80
0.80
0.80
0.80
0.80
0.80
0.25
0.70
0.70
0.70
0.70
0.99
0.99
0.99
0.99
0.99
0.99

0.
1.
0.
1.
1.
1.
0.
0.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
1.
0.
SizeBin
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
1
1
1
1
1
2

2.200E-11

0.0

-

3.000E-11

0.0

0.00

Figure 3‐1a (continued).
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

34

www.camx.com

March 2016

9
1
10 3
11 1
12 3
13 3
14 1
15 1
16 3
17 3
18 4
0.70
19 1
20 1
21 4
300.0
22 3
23 1
24 4
0.10
25 1
26 3
27 1
28 4
300.0
29 6
300.0
30 3
31 1
32 1
33 1
34 7
300.0
35 7
300.0
36 1
37 3
38 1
39 3
40 3
41 3
42 4
300.0
43 3
44 3
45 3
46 1
47 1
48 1
49 1
50 3
51 1
52 1
53 1
54 3
55 3
56 3
57 3
58 1
59 1
60 1
61 3

1
2
2
2
2
1
1
2
2
2

CAMx User’s Guide Version 6.3
3. Core Model Input / Output Structures

2
2
1
2
2

0.000E+00
2.100E-11
2.200E-10
1.700E-12
1.000E-14
0.000E+00
0.000E+00
1.500E-11
4.500E-14
2.000E-30
300.0
2.500E-22
1.800E-39
1.000E-03
0.45
3.300E-39
5.000E-40
7.000E-31
300.0
0.000E+00
1.800E-11
1.000E-20
2.000E-30
0.60
2.400E-14
6.500E-34
3.500E-12
0.000E+00
0.000E+00
0.000E+00
2.300E-13

3

3.220E-34

1
2
2
2
2
2
2

0.000E+00
2.900E-12
1.100E-10
5.500E-12
2.200E-11
4.200E-12
6.900E-31
0.60
4.800E-11
3.000E-11
1.400E-12
1.000E-11
2.200E-11
3.500E-12
1.000E-17
8.500E-13
0.000E+00
0.000E+00
0.000E+00
2.600E-12
2.600E-12
7.500E-13
7.500E-13
6.800E-14
6.800E-14
6.800E-14
5.900E-13

2
3
1
3
3
2
1
2
2
2
2

2
2
2
2
2
2
2
2
1
1
1
2
2
2
2
2
2
2
2

-102.0

0.00

300.0

940.0
490.0

0.00
0.00

300.0
300.0

-170.0
1260.0
0.0
0.60

0.00
0.00
-4.40
1.00

300.0
300.0
300.0

1.400E-12

0.0

-

11000.0
1.00
-530.0

-3.50

300.0

9.700E+14

11080.0

0.10

0.00

300.0

0.0
0.60

-2.60
1.00

300.0

3.600E-11

0.0

-

390.0

0.00

300.0

0.0
1.00
-460.0
-1335.0
-250.0

-3.00

300.0

2.500E-11

0.0

0.00

0.00
0.00
0.00

300.0
300.0
300.0

2.700E-17

-2199.0

0.00

-600.0

0.00

300.0

1.700E-33

-1000.0

0.00

-2800.0

0.00

300.0

2.380E-54

-3200.0

0.00

160.0

0.00

300.0

2000.0
-120.0
240.0
0.0
1.00
-250.0
-200.0
2000.0

0.00
0.00
0.00
-1.00

300.0
300.0
300.0
300.0

2.600E-11

0.0

0.00

0.00
0.00
0.00

300.0
300.0
300.0

2450.0

0.00

300.0

-365.0
-365.0
-700.0
-700.0

0.00
0.00
0.00
0.00

300.0
300.0
300.0
300.0

360.0

0.00

300.0

Figure 3‐1a (continued).
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

35

www.camx.com

March 2016

62 1
63 4
300.0
64 3
65 1
66 7
300.0
67 3
68 3
69 3
70 3
71 3
72 1
73 3
74 1
75 1
76 1
77 3
78 1
79 3
80 3
81 1
82 3
83 1
84 3
85 3
86 3
87 1
88 3
89 4
0.90
90 4
300.0
91 1
92 3
93 3
94 3
95 3
96 3
97 1
98 3
99 3
100 3
101 1
102 1
103 3
104 4
0.90
105 4
300.0
106 1
107 1
108 3
109 3
110 3
111 3
112 3
113 3
114 3
115 1
116 3

1
2
2
1
2
2
2
2
2
2
1
2
2
1
1
2
2
2
1
2
2
2
2
2
2
1
2
2
1
1
2
2
2
2
2
1
2
2
2
2
1
2
2
1
1
2
2
2
2
2
2
2
2
2
1

CAMx User’s Guide Version 6.3
3. Core Model Input / Output Structures
0.000E+00
3.000E-31
0.60
3.010E-12
0.000E+00
1.440E-13
2.450E-12
2.800E-12
4.100E-13
9.500E-14
3.800E-12
0.000E+00
7.300E-12
9.000E-12
0.000E+00
0.000E+00
3.400E-11
5.800E-16
9.700E-15
2.400E+12
5.600E-12
5.600E-15
4.000E-13
1.800E-11
5.600E-12
1.400E-12
0.000E+00
8.100E-12
2.700E-28
300.0
4.900E-03
0.30
0.000E+00
4.300E-13
2.000E-12
4.400E-13
2.900E-12
4.000E-13
0.000E+00
4.000E-13
1.300E-11
5.100E-12
6.500E-15
0.000E+00
6.700E-12
2.700E-28
300.0
4.900E-03
0.30
0.000E+00
3.000E-13
4.300E-13
2.000E-12
4.400E-13
2.900E-12
2.900E-12
8.700E-12
6.900E-12
8.100E-13
1.000E+15

0.0
1.00
-190.0

-3.30

300.0

1.500E-12

0.0

0.00

0.00

300.0

0.0

0.00

300.0

3.430E-33

0.0

0.00

1775.0
-300.0
-750.0
-390.0
-200.0

0.00
0.00
0.00
0.00
0.00

300.0
300.0
300.0
300.0
300.0

620.0

0.00

300.0

1600.0

0.00

300.0

-625.0
7000.0

0.00
0.00

300.0
300.0

-2300.0

0.00

300.0

1100.0
-270.0
1900.0

0.00
0.00
0.00

300.0
300.0
300.0

-270.0
0.0
0.30
12100.0
1.00

0.00
-7.10
1.00
0.00

300.0
300.0

1.200E-11

0.0

-

300.0

5.400E+16

13830.0

0.00

-1040.0
-500.0
-1070.0
-500.0
-200.0

0.00
0.00
0.00
0.00
0.00

300.0
300.0
300.0
300.0
300.0

-200.0
870.0
-405.0

0.00
0.00
0.00

300.0
300.0
300.0

-340.0
0.0
0.30
12100.0
1.00

0.00
-7.10
1.00
0.00

300.0
300.0

1.200E-11

0.0

-

300.0

5.400E+16

13830.0

0.00

-1040.0
-500.0
-1070.0
-500.0
-500.0
1070.0
230.0

0.00
0.00
0.00
0.00
0.00
0.00
0.00

300.0
300.0
300.0
300.0
300.0
300.0
300.0

8000.0

0.00

300.0

Figure 3‐1a (continued).
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

36

www.camx.com

March 2016

117 1
118 1
119 3
120 1
121 3
122 3
123 3
124 4
300.0
125 3
126 3
127 1
128 3
129 3
130 3
131 3
132 1
133 1
134 1
135 1
136 1
137 1
138 1
139 1
140 3
141 3
142 1
143 1
144 1
145 3
146 3
147 3
148 1
149 1
150 1
151 1
152 1
153 1
154 3
155 3
156 3

1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
1
2
2
2
2
1
2
2
2
2
1
2
2
2
2
2
2
2
2
1
2
2
2
2

CAMx User’s Guide Version 6.3
3. Core Model Input / Output Structures
1.600E+03
1.500E-11
1.000E-11
3.200E-11
6.500E-15
7.000E-13
1.040E-11
1.000E-28
0.60
1.200E-14
3.300E-12
2.300E-11
1.000E-11
8.400E-15
9.600E-13
1.800E-12
8.100E-12
4.200E+00
4.100E-11
2.200E-11
1.400E-11
5.500E-12
0.000E+00
3.000E-11
5.400E-17
1.700E-11
1.700E-11
0.000E+00
3.600E-11
2.540E-11
7.860E-15
3.030E-12
1.500E-19
3.360E-11
7.100E-18
1.000E-15
0.000E+00
3.600E-11
1.500E-11
1.200E-15
3.700E-12

280.0

0.00

300.0

1900.0
2160.0
792.0
0.0
1.00
2630.0
2880.0

0.00
0.00
0.00
-0.80

300.0
300.0
300.0
300.0

0.00
0.00

300.0
300.0

-550.0
1100.0
270.0
-355.0

0.00
0.00
0.00
0.00

300.0
300.0
300.0
300.0

500.0
-116.0

0.00
0.00

300.0
300.0

-407.6
1912.0
448.0

0.00
0.00
0.00

300.0
300.0
300.0

-449.0
821.0
-175.0

0.00
0.00
0.00

300.0
300.0
300.0

8.800E-12

0.0

0.00

Figure 3‐1a (concluded).
CAMx Version
|VERSION6.3
Mechanism ID
|0
Aerosol Option
|NONE
Description
|inert test
No of gas species |1
No of aero species |0
No of reactions
|0
Prim photo rxns
|0
No of sec photo rxn|0
SrfMod #spc, #rxns |0 0
Species Records
Gas Spec
lower bnd
H-law
1 TRACER
1.00E-09 1.00e-10

T-fact
0.

Molwt
1.00

Reactvty
0.0

Rscale
1.

Figure 3‐1b. Example inert chemistry parameters file (requires chemistry flag to be set false –
see the description of the CAMx control file).
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

37

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
3. Core Model Input / Output Structures

The section of the file that lists gas species must include the names of radical and state species,
according to the selected chemical mechanism, along with each species’ lower bound value,
Henry’s Law solubility parameters, molecular weight, and surface reactivity and resistance
scaling parameters for use in the dry deposition calculations. The lower bound values set the
minimum concentration “floor” for any chemical or physical process that reduces
concentrations to “zero”. The lower bound is also used to set initial and/or boundary
conditions for species that are omitted from the initial and/or boundary condition files.
The section of the file that lists aerosol species names must include a lower‐bound value,
particle density, dry extinction efficiency, a hygroscopic flag, and single‐scattering albedo. In
the case of INERT or CF options, the user must also specify the size bin assigned to each species
at the end of each particulate species record (CMU automatically applies each species to all size
bins). Since the effect of aerosol water on optical parameters is taken into account through an
internal relative humidity adjustment, the dry extinction efficiency for particle species PH2O
must be set to zero.
CAMx supports several equations for specifying gas‐phase rate constants, as shown in Table 3‐
3a. The type of equation used for each reaction is identified by the second parameter specified
for each reaction – a number between 1 and 7 (Table 3‐3a). The number of additional
parameters required depends upon the expression type and varies between 2 and 13, as shown
in Table 3‐3b. Expression type 4 (Troe expression) allows for a complete description of
dependencies on temperature and pressure; background information on Troe expressions may
be found in the NASA and IUPAC rate constant compilations (NASA, 1997; IUPAC, 1992).
Rate constants can be specified in molecular units (e.g., cm3 molecule‐1 s‐1) or ppm units (e.g.,
ppm‐1 min‐1). All the rate constants must be in a single units system; CAMx will determine
which units system is being used from the magnitude of the rate constants. Diagnostic
information on the rate constants and units system is output by CAMx at run‐time.

3.2 Photolysis Rates File
The rates for the primary photolysis reactions are supplied via the photolysis rates file in units
of minute‐1. This file must be supplied if chemistry is invoked. The photolysis rates file
comprises a large look‐up table of clear‐sky photolysis rates specific to the gas‐phase chemistry
mechanism to run. Rates are arranged in a matrix of five dimensions, including variations over
10 solar zenith angles, 5 ultraviolet (UV) surface albedos, 3 terrain heights, 11 altitudes above
ground, and 5 total ozone column values. The look‐up table is generated using the TUV
preprocessor, which internally specifies the ranges of solar zenith (0, 10, 20, 30, 40, 50, 60, 70,
78, 86), surface UV albedo (0.04, 0.1, 0.2, 0.5, 0.9.), and terrain heights (0, 1, 3 km). The ranges
of altitude above ground are controlled by the user, while the ranges of ozone column are
taken from the ozone column file (Section 3.3). TUV is run with a typical aerosol profile defined
by Elterman (1968).

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

38

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
3. Core Model Input / Output Structures

Table 3‐3a. Rate constant expression types supported in CAMx and order of expression
parameters for the chemistry parameters file.
Expression
Type

Description

1

Constant

2

UAM (Arrhenius expression)

Expression

k  k298
  1
1 
k  k298 exp Ea 
 
  298 T 
B

3

 T 
 E 
 exp  a 
k  A 
 T 
 TR 

General temperature dependence


k 0 M 
k 
 1  k 0 M  / k 


 G
F



B

4

Troe‐type temperature and pressure
dependence

 T 
 E 
 exp  a 
k 0  A 
 T 
 TR 
k



 T 

 A 
 TR 

B

 E 
exp  a 
 T 



5

Equilibrium with a previously defined
reaction (kref)

6

Lindemann ‐ Hinshelwood as used for OH +
HNO3

7

Simple pressure dependence used for OH +
CO

 


 
 

1

  T B
 E 
k  k ref  A   exp  a 
 T 
  TR 

1

  log k 0 [ M ] / k 
G  1  
n
 

k  k0 

2

k3[M ]
1  k3[M ] / k2

k  k1  k 2 [ M ]

Notes:
T is temperature (K)
TR is reference temperature of 300 K
Ea is an Arrhenius activation energy (K)
k0 is the low pressure limit of the rate constant
k is the high pressure limit of the rate constant
[M] is the concentration of air

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

39

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
3. Core Model Input / Output Structures

Table 3‐3b. List of parameters that must be provided in the CAMx chemistry parameter file
for each of the seven types of rate constant expressions. Use ppm/minute units for A and
Kelvin for Ea and TR. The variable O is the order of the reaction (1 to 3).
Expression

Parameters

Type
1
2
3
4
5
6
7

1
O
O
O
O
O
O
O

2

3

4

5

6

7

8

9

10

11

k298
k298
A
A
kref
Ao
A1

Ea
Ea
Ea
A
E ao
E a1

B
B
Ea
Bo
B1

TR
TR
B
TRo
TR1

A’
TR
A2
A2

E a’

B’

TR’

F

n

E a2
E a2

B2
B2

TR2
TR2

A3

E a3

12

13

B3

TR3

The photolysis rates file is a readable text format and it has the following structure:
TUV4.8CAMx6
Loop from 1 to nozn ozone column intervals:
Loop from 1 to nalb UV albedo intervals:
Loop from 1 to ntht terrain height intervals:
ozcl,albcl,trncl
(12X,f7.3,8X,f7.3,11X,f7.3)
Loop from 1 to nalt altitudes above ground:
height
(*)
Loop from 1 to nphot photolysis reactions:
(pk(n),n=1,nsol)
(1X,10F12.0)

where the first record labels the version of TUV used to generate the file, and where variables
have the following definitions:
ozcl
albcl
trncl
height
pk

Ozone column value for the current interval (Dobson units)
UV albedo value for the current interval (unitless)
Terrain height value for the current interval (km MSL)
Altitude (km AGL)
Photolysis rates (min‐1) for nsol solar zenith angles

Figure 3‐2 presents an example of a photolysis rates file for the first several panels of data.

3.3 Ozone Column File
This file defines the intervals of total atmospheric ozone column to be used by TUV, as well as
its spatial and temporal distributions for a specific CAMx domain and episode. This parameter
is essential for photochemical simulations as it determines the spatial and temporal variation of
photolysis rates. Therefore, this file must be supplied if chemistry is invoked. Additionally, the
ozone column file may also provide an optional field defining a land/ocean mask (for mercury
chemistry).
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS
www.camx.com
40

March 2016

TUV4.8CAMx6
O3 Column = 0.290 Albedo= 0.040 Terrain = 0.000
0.000
km above surface
4.562E-01
4.526E-01
4.416E-01
4.224E-01
3.934E-01
1.670E-03
1.643E-03
1.563E-03
1.427E-03
1.236E-03
2.211E-03
2.186E-03
2.107E-03
1.973E-03
1.776E-03
2.078E-03
2.007E-03
1.801E-03
1.485E-03
1.097E-03
3.131E-04
3.058E-04
2.841E-04
2.491E-04
2.027E-04
8.728E-05
8.638E-05
8.364E-05
7.891E-05
7.189E-05
0.150
km above surface
4.866E-01
4.830E-01
4.717E-01
4.521E-01
4.223E-01
1.855E-03
1.827E-03
1.742E-03
1.598E-03
1.394E-03
2.443E-03
2.416E-03
2.334E-03
2.193E-03
1.986E-03
2.327E-03
2.249E-03
2.024E-03
1.677E-03
1.250E-03
3.489E-04
3.410E-04
3.177E-04
2.799E-04
2.297E-04
9.553E-05
9.460E-05
9.176E-05
8.684E-05
7.952E-05
0.360
km above surface
5.050E-01
5.013E-01
4.900E-01
4.702E-01
4.402E-01
1.974E-03
1.945E-03
1.857E-03
1.708E-03
1.497E-03
2.587E-03
2.560E-03
2.477E-03
2.333E-03
2.120E-03
2.494E-03
2.411E-03
2.174E-03
1.806E-03
1.353E-03
3.722E-04
3.640E-04
3.396E-04
3.002E-04
2.475E-04
1.006E-04
9.970E-05
9.681E-05
9.180E-05
8.433E-05
0.640
km above surface
5.230E-01
5.193E-01
5.080E-01
4.882E-01
4.581E-01
2.092E-03
2.062E-03
1.972E-03
1.819E-03
1.601E-03
2.731E-03
2.703E-03
2.618E-03
2.471E-03
2.254E-03
2.664E-03
2.577E-03
2.327E-03
1.939E-03
1.459E-03
3.958E-04
3.872E-04
3.618E-04
3.207E-04
2.656E-04
1.057E-04
1.047E-04
1.018E-04
9.672E-05
8.912E-05
0.980
km above surface
5.409E-01
5.373E-01
5.260E-01
5.062E-01
4.760E-01
2.210E-03
2.179E-03
2.087E-03
1.930E-03
1.705E-03
2.873E-03
2.845E-03
2.758E-03
2.609E-03
2.387E-03
2.837E-03
2.746E-03
2.483E-03
2.074E-03
1.567E-03
4.195E-04
4.106E-04
3.842E-04
3.414E-04
2.840E-04
1.107E-04
1.097E-04
1.068E-04
1.016E-04
9.392E-05
1.380
km above surface
5.593E-01
5.556E-01
5.444E-01
5.247E-01
4.947E-01
2.329E-03
2.298E-03
2.203E-03
2.042E-03
1.811E-03
3.016E-03
2.988E-03
2.900E-03
2.749E-03
2.524E-03
3.015E-03
2.920E-03
2.644E-03
2.215E-03
1.681E-03
4.437E-04
4.345E-04
4.071E-04
3.627E-04
3.029E-04
1.157E-04
1.148E-04
1.118E-04
1.066E-04
9.883E-05
1.840
km above surface
5.782E-01
5.746E-01
5.635E-01
5.440E-01
5.141E-01
2.451E-03
2.419E-03
2.322E-03
2.157E-03
1.920E-03
3.162E-03
3.133E-03
3.045E-03
2.892E-03
2.664E-03
3.200E-03
3.101E-03
2.811E-03
2.361E-03
1.799E-03
4.685E-04
4.589E-04
4.307E-04
3.846E-04
3.224E-04
1.209E-04
1.199E-04
1.169E-04
1.117E-04
1.039E-04
2.350
km above surface
5.974E-01
5.939E-01
5.829E-01
5.636E-01
5.340E-01
2.572E-03
2.540E-03
2.441E-03
2.273E-03
2.031E-03
3.307E-03
3.278E-03
3.190E-03
3.036E-03
2.805E-03
3.389E-03
3.285E-03
2.983E-03
2.512E-03
1.922E-03
4.934E-04
4.836E-04
4.544E-04
4.068E-04
3.424E-04
1.261E-04
1.251E-04
1.221E-04
1.169E-04
1.090E-04
2.910
km above surface
6.169E-01
6.134E-01
6.026E-01
5.836E-01
5.543E-01
2.694E-03
2.661E-03
2.561E-03
2.390E-03
2.143E-03
3.452E-03
3.423E-03
3.335E-03
3.180E-03
2.948E-03
3.582E-03
3.474E-03
3.159E-03
2.667E-03
2.049E-03
5.187E-04
5.086E-04
4.786E-04
4.294E-04
3.628E-04
1.312E-04
1.303E-04
1.273E-04
1.221E-04
1.142E-04
3.530
km above surface
6.368E-01
6.334E-01
6.228E-01
6.041E-01
5.754E-01
2.818E-03
2.784E-03
2.683E-03
2.510E-03
2.258E-03
3.599E-03
3.571E-03
3.482E-03
3.328E-03
3.094E-03
3.783E-03
3.670E-03
3.343E-03
2.830E-03
2.184E-03
5.446E-04
5.343E-04
5.034E-04
4.528E-04
3.840E-04
1.365E-04
1.355E-04
1.326E-04
1.274E-04
1.195E-04
4.210
km above surface
6.569E-01
6.536E-01
6.433E-01
6.250E-01
5.969E-01
2.943E-03
2.909E-03
2.807E-03
2.632E-03
2.376E-03
3.747E-03
3.719E-03
3.631E-03
3.477E-03
3.243E-03
3.992E-03
3.875E-03
3.534E-03
3.000E-03
2.325E-03
5.711E-04
5.605E-04
5.289E-04
4.769E-04
4.060E-04
1.417E-04
1.408E-04
1.379E-04
1.327E-04
1.249E-04
O3 Column = 0.290 Albedo= 0.040 Terrain = 1.000
0.000
km above surface
4.655E-01
4.615E-01
4.493E-01
4.279E-01
3.958E-01
1.688E-03
1.660E-03
1.576E-03
1.435E-03
1.237E-03
2.240E-03
2.212E-03
2.129E-03
1.987E-03
1.780E-03
2.098E-03
2.025E-03
1.815E-03
1.492E-03
1.098E-03
3.162E-04
3.086E-04
2.863E-04
2.503E-04
2.029E-04
8.857E-05
8.761E-05
8.467E-05
7.961E-05
7.213E-05
0.150
km above surface
5.004E-01
4.963E-01
4.839E-01
4.623E-01
4.296E-01
1.870E-03
1.841E-03
1.752E-03
1.604E-03
1.394E-03
2.473E-03
2.445E-03
2.359E-03
2.211E-03
1.994E-03
2.338E-03
2.259E-03
2.030E-03
1.677E-03
1.245E-03
3.509E-04
3.428E-04
3.189E-04
2.803E-04
2.292E-04
9.713E-05
9.614E-05
9.312E-05
8.789E-05
8.013E-05

CAMx User’s Guide Version 6.3
3. Core Model Input / Output Structures

3.516E-01
9.889E-04
1.506E-03
6.925E-04
1.482E-04
6.212E-05

2.921E-01
6.936E-04
1.152E-03
3.382E-04
9.100E-05
4.892E-05

2.065E-01
3.747E-04
7.108E-04
1.034E-04
4.015E-05
3.166E-05

1.148E-01
1.502E-04
3.352E-04
2.201E-05
1.264E-05
1.580E-05

3.074E-02
2.570E-05
7.560E-05
2.056E-06
1.517E-06
3.934E-06

3.794E-01
1.130E-03
1.701E-03
8.008E-04
1.704E-04
6.928E-05

3.179E-01
8.102E-04
1.322E-03
4.035E-04
1.075E-04
5.530E-05

2.284E-01
4.566E-04
8.415E-04
1.337E-04
5.024E-05
3.672E-05

1.308E-01
1.956E-04
4.159E-04
3.383E-05
1.755E-05
1.910E-05

3.760E-02
3.701E-05
1.000E-04
4.166E-06
2.556E-06
5.068E-06

3.968E-01
1.222E-03
1.826E-03
8.740E-04
1.851E-04
7.382E-05

3.343E-01
8.862E-04
1.432E-03
4.474E-04
1.183E-04
5.940E-05

2.426E-01
5.096E-04
9.258E-04
1.536E-04
5.679E-05
3.999E-05

1.412E-01
2.245E-04
4.673E-04
4.129E-05
2.067E-05
2.122E-05

4.171E-02
4.415E-05
1.152E-04
5.492E-06
3.210E-06
5.763E-06

4.143E-01
1.315E-03
1.951E-03
9.497E-04
2.000E-04
7.839E-05

3.510E-01
9.632E-04
1.544E-03
4.928E-04
1.294E-04
6.354E-05

2.574E-01
5.633E-04
1.011E-03
1.738E-04
6.343E-05
4.331E-05

1.520E-01
2.535E-04
5.191E-04
4.872E-05
2.378E-05
2.336E-05

4.577E-02
5.126E-05
1.304E-04
6.809E-06
3.860E-06
6.448E-06

4.321E-01
1.409E-03
2.077E-03
1.028E-03
2.152E-04
8.299E-05

3.682E-01
1.041E-03
1.656E-03
5.396E-04
1.407E-04
6.775E-05

2.727E-01
6.178E-04
1.099E-03
1.944E-04
7.017E-05
4.673E-05

1.635E-01
2.827E-04
5.718E-04
5.611E-05
2.689E-05
2.556E-05

4.989E-02
5.837E-05
1.454E-04
8.121E-06
4.509E-06
7.133E-06

4.507E-01
1.505E-03
2.207E-03
1.109E-03
2.310E-04
8.772E-05

3.863E-01
1.122E-03
1.773E-03
5.887E-04
1.524E-04
7.213E-05

2.892E-01
6.743E-04
1.190E-03
2.158E-04
7.716E-05
5.031E-05

1.760E-01
3.128E-04
6.270E-04
6.364E-05
3.009E-05
2.789E-05

5.427E-02
6.563E-05
1.609E-04
9.454E-06
5.169E-06
7.835E-06

4.702E-01
1.605E-03
2.341E-03
1.196E-03
2.473E-04
9.263E-05

4.056E-01
1.206E-03
1.894E-03
6.407E-04
1.647E-04
7.671E-05

3.069E-01
7.338E-04
1.286E-03
2.385E-04
8.451E-05
5.412E-05

1.897E-01
3.445E-04
6.856E-04
7.141E-05
3.341E-05
3.039E-05

5.904E-02
7.312E-05
1.769E-04
1.082E-05
5.848E-06
8.570E-06

4.904E-01
1.706E-03
2.477E-03
1.285E-03
2.641E-04
9.763E-05

4.257E-01
1.292E-03
2.019E-03
6.952E-04
1.774E-04
8.142E-05

3.258E-01
7.955E-04
1.386E-03
2.622E-04
9.214E-05
5.811E-05

2.046E-01
3.775E-04
7.476E-04
7.936E-05
3.685E-05
3.305E-05

6.425E-02
8.077E-05
1.934E-04
1.221E-05
6.538E-06
9.333E-06

5.111E-01
1.810E-03
2.615E-03
1.379E-03
2.814E-04
1.027E-04

4.466E-01
1.382E-03
2.148E-03
7.528E-04
1.906E-04
8.630E-05

3.458E-01
8.602E-04
1.492E-03
2.874E-04
1.002E-04
6.231E-05

2.209E-01
4.124E-04
8.140E-04
8.761E-05
4.045E-05
3.593E-05

7.003E-02
8.867E-05
2.106E-04
1.363E-05
7.246E-06
1.014E-05

5.328E-01
1.917E-03
2.759E-03
1.478E-03
2.996E-04
1.080E-04

4.687E-01
1.476E-03
2.283E-03
8.148E-04
2.046E-04
9.142E-05

3.673E-01
9.296E-04
1.605E-03
3.147E-04
1.088E-04
6.681E-05

2.389E-01
4.505E-04
8.871E-04
9.644E-05
4.434E-05
3.911E-05

7.670E-02
9.704E-05
2.291E-04
1.512E-05
7.989E-06
1.103E-05

5.550E-01
2.029E-03
2.906E-03
1.585E-03
3.186E-04
1.135E-04

4.917E-01
1.574E-03
2.424E-03
8.818E-04
2.195E-04
9.674E-05

3.902E-01
1.004E-03
1.725E-03
3.447E-04
1.181E-04
7.161E-05

2.588E-01
4.925E-04
9.679E-04
1.060E-04
4.859E-05
4.262E-05

8.447E-02
1.060E-04
2.493E-04
1.668E-05
8.775E-06
1.202E-05

3.499E-01
9.833E-04
1.498E-03
6.883E-04
1.473E-04
6.180E-05

2.855E-01
6.824E-04
1.132E-03
3.334E-04
8.961E-05
4.802E-05

1.957E-01
3.626E-04
6.858E-04
1.009E-04
3.896E-05
3.042E-05

1.048E-01
1.414E-04
3.159E-04
2.101E-05
1.193E-05
1.483E-05

2.871E-02
2.142E-05
6.495E-05
1.716E-06
1.254E-06
3.488E-06

3.828E-01
1.123E-03
1.697E-03
7.929E-04
1.690E-04
6.934E-05

3.163E-01
7.976E-04
1.306E-03
3.960E-04
1.056E-04
5.476E-05

2.220E-01
4.424E-04
8.179E-04
1.294E-04
4.859E-05
3.574E-05

1.237E-01
1.847E-04
3.956E-04
3.180E-05
1.651E-05
1.823E-05

3.1E-02
3.188E-05
8.840E-05
3.572E-06
2.187E-06
4.615E-06

Figure 3‐2. Example of the first several panels of lookup data in the photolysis rates input file.
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

41

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
3. Core Model Input / Output Structures

There are two mandatory and one optional header records in the ozone column file. The first
record contains an arbitrary file label. The second record defines the intervals for 5 ozone
column values for the domain and temporal period to be simulated. These intervals must
exactly match those defined in preparing the photolysis rates file, so the ozone column file is
also read by the TUV preprocessor to define the photolysis rates lookup table (Section 3.2).
If the optional land/ocean mask is included, then a third header record must be added to
inform CAMx that this field is to be read. The time‐invariant land/ocean mask is used for
mercury chemistry to define profiles of ambient halogens; it is simply a map of 0 (land and fresh
water bodies) and 1 (ocean) that must be located directly under its header record. A value
must be supplied for each cell of the master grid and optionally any nested grids.
Gridded fields of time‐varying ozone column follow the header records and optional land/ocean
mask data. The gridded fields are maps of the respective “codes” for each interval, as defined
in the header. For example, 5 ozone column intervals are specified in TUV and in the ozone
column header record, so the map must consist of a distribution of integers ranging from 1 to 5.
Ozone column is supplied for the master grid only; CAMx internally assigns master cell values to
all nested grids cells. Multiple maps of these codes may be provided for arbitrary time intervals
that span the entire simulation period.
The ozone column file is a readable text format and it has the following structure:
text
ozname,(ozncl(n),n=1,nozn)
loname,igrd,nx,ny
Loop from j = ny grid rows to 1
(jocn(i,j),i=1,nx)
loname,igrd,nx,ny
ozname,idt1,tim1,idt2,tim2
Loop from j = ny master grid rows to 1
(jozn(i,j),i=1,nx)

(A)
(A10,5F10.0)
(A10,3I10)

‐‐ Optional
‐‐ Optional
(999I1)
‐‐ Optional
(A10,3I10)
‐‐ Optional
(A10,I10,F10.0,I10,F10.0)
(9999I1)

where the variables in the ozone column file have the following definitions:
text
ozname
ozncl
loname
igrd
nx
ny
jocn
idt1
tim1
idt2

Text identifying file and any messages
Text string “OZONE COL”
Ozone column (Dobson units) for each of nozn ozone values
Text string “OCEAN”
Grid index (1 = master grid, 2+=nested grid, 0 = end of data)
Number of grid columns for this grid index
Number of grid rows for this grid index
Grid igrd, row j land/ocean codes for nx grid columns
Beginning date (YYJJJ) of time span
Beginning hour (HHMM) of time span
Ending date of time span

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

42

www.camx.com

March 2016

tim2
jozn

CAMx User’s Guide Version 6.3
3. Core Model Input / Output Structures

Ending hour of time span
Master grid row j ozone column codes for nx master grid columns

An example of a small ozone column file is given in Figure 3‐3.
Sample ozone column file with optional land ocean mask
OZONE COL
0.285
0.315
0.345
0.375
0.405
OCEAN
1
64
10
0000000000000000000011111111100000000000000000000000000000000000
0000000000000000000000000001111111110000000000000000000000000000
0000000000000000000000001111111111111100000000000000000000000000
0000000000000000000000000001111111111110000000000000000000000000
0000000000000000000000011111111111111100000000000000000000000000
0000000000000000000000111111111111111110000000000000000000000000
0000000000000000000000111111111111111111000000000000000000000000
0000000000000000000001111111000000000111111110000000000000000000
0000000000000000000011111111100000000011111111100000000000000000
0000000000000000000000111111111100000000011111111110000000000000
OCEAN
0
0
0
OZONE COL
05213
0.00
05213
2400.00
3333333333333333333333333333333333333333333333333333333333333333
3333333333333333333333333333333333333333333333333333333333333333
2222222222222333333333333333333333333333333333333333333333333333
2222222222222223333333333333333333333333333333333333333333333333
2222222222222223333333333333333333333333333333333333333333333333
2222222222222223333333333333333333333333333333333333333333333333
2222222222222223333333333333333333333333333333333333333333333333
2222222222222223333333333333333333333333333333333333333333333333
2222222222222223333333333333333333333333333333333333333333333333
2222222222222222222222333333333333333333333333333333333333333333

Figure 3‐3. Example structure of a single‐grid ozone column input file showing panels for the
optional time‐invariant land‐ocean mask and time‐varying ozone column field.

3.4 Fortran Binary Input/Output Files
3.4.1 What is a Fortran Binary File?
Large CAMx input and output data fields are contained within Fortran “unformatted” (binary)
files. This means that the data are read and written as represented in memory, without
translation between binary and ASCII character sets as done for “text” files. Binary files reduce
file volume and improve program read/write speed, but the user cannot directly view or
manually edit them. There are two ways to represent binary information in memory: “big
endian” and “little endian.” The difference between these is essentially the order of bits in a
word, and which order is used depends on the computer chipset. Historically, big endian has
been used in many Unix workstations (Sun, SGI, HP, and IBM). The x86 processors on personal
computer platforms (e.g., Intel and AMD) use little endian, while PowerPC chips are big endian.
CAMx can be compiled and run on machines using their native big or little endian binary
representations, as long as the model and all of its pre‐ and post‐processors are consistently
compiled and run on the same type of platform. If any component of the modeling system is
compiled for a different platform using the opposite binary representation, I/O files will not be
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

43

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
3. Core Model Input / Output Structures

properly read and will likely lead to a program crash. A typical run‐time error message from
trying to read the wrong binary format is “input record too long,” so if you get this error
message, check for consistency between your binary files and Fortran compiler options.
Compilers for little endian machines provide compile‐time switches that allow binary files to be
read and written as big endian. The “makefile” compiler script that is distributed with CAMx
sets compiler flags to consistently use big endian to maximize platform portability. Therefore,
use of the CAMx Makefile will by default result in the model reading and writing big endian
binary files. Additional information on this topic is provided in Section 2.
3.4.2 CAMx Binary File Headers
The format of all binary CAMx I/O files follows the convention established by the Urban Airshed
Model (EPA, 1990). CAMx binary files contain a set of time‐invariant header records, followed
by a set of data records that contain time‐ and variable‐specific fields. All input gridded
emissions, initial and top boundary conditions, meteorology1, and output concentration and
deposition files share the same basic format. The input lateral boundary condition and point
source emissions files are similar but include additional records specific to their data structures.
The input 3D meteorological file may provide wind fields in a staggered or un‐staggered grid
arrangement. A flag to indicate the wind staggering is included in the second header record
and is checked only when the 3D meteorological file is read.
The four header records within all CAMx binary files have the following structure:
name,note,itzon,nvar,ibdate,btime,iedate,etime
plon,plat,iutm,xorg,yorg,delx,dely,nx,ny,nz,iproj,istag,tlat1,tlat2,rdum
ione,ione,nx,ny
(namvar(l),l=1,nvar)

The header variables have the following definitions:
Record 1
name

note
itzon
nvar
1

Text string describing file contents (character*4(10) array):
AIRQUALITY
Initial and top boundary conditions
BOUNDARY
Boundary conditions
EMISSIONS
Gridded emissions
PTSOURCE
Point source emissions
AVERAGE
Average output concentrations/deposition and
input meteorology/surface variables
INSTANT
Instantaneous output concentrations
Text string containing user note (character*4(60) array)
Integer time zone (0=UTC, 5=EST, etc.)
Integer number of variables on file

NOTE: Starting with CAMx v6.00, all binary meteorological files have been converted to the UAM convention.
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS
www.camx.com
44

March 2016

ibdate
btime
iedate
etime
Record 2
plon
plat
iutm
xorg
yorg
delx
dely
nx
ny
nz
iproj

istag
tlat1
tlat2
rdum
Record 3
ione
ione
nx
ny
Record 4
namvar

CAMx User’s Guide Version 6.3
3. Core Model Input / Output Structures

Integer beginning date span on file (YYJJJ)
Real beginning decimal hour (e.g., 1:30 PM = 13.5)
Integer ending date span on file (YYJJJ)
Real ending decimal hour (e.g., 1:30 PM = 13.5)
Real projection pole/origin longitude (degrees; west is negative)
Real projection pole/origin latitude (degrees; south is negative)
Integer UTM zone (ignored for other projections)
Real x‐coordinate at southwest corner of grid (m or degrees longitude)
Real y‐coordinate at southwest corner of grid (m or degrees latitude)
Real cell size in x‐direction (m or degrees longitude)
Real cell size in y‐direction (m or degrees latitude)
Integer number of grid columns (east‐west)
Integer number of grid rows (north‐south)
Integer number of vertical layers
Integer projection index:
0 = geodetic (LATLON)
1 = Universal Transverse Mercator (UTM)
2 = Lambert Conic Conformal (LAMBERT)
3 = Rotated Polar Stereographic (RPOLAR)
4 = Polar Stereographic (POLAR)
5 = Mercator (MERCATOR)
Integer flag to indicate wind staggering (1=staggered, 0=not staggered)
Real LCP first true latitude (degrees; south is negative)
Real LCP second true latitude (degrees; south is negative)
Real dummy variable
Integer dummy variable (=1)
Integer dummy variable (=1)
Integer number of grid columns (east‐west)
Integer number of grid rows (north‐south)
Text names for nvar variables (character*4(10,nvar) array)

3.4.3 Input Files
The Fortran binary input files include initial/boundary conditions, gridded and elevated point
source emissions, and several meteorological files. All times on input files must match the time
zone specified in the CAMx control file (CAMx.in).
Initial/boundary condition files may include a single time interval covering the entire simulation
period, or more detailed hour‐by‐hour (or other interval) variations. The time intervals are
allowed to be entirely arbitrary to maximize flexibility in defining these inputs. A subset of the
pollutant species to be simulated may be defined in the initial/boundary condition files; any
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

45

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
3. Core Model Input / Output Structures

species that are missing will be set to “lower bound” values as defined in the chemistry
parameters file.
Emission inputs are usually developed at one hour intervals, but the file structures allow for any
other interval as long as a consistent interval is used throughout the file (all one hour, or all
three hour, but no mixing of these). The time interval on all emission files must match the
emission update frequency defined in the CAMx control file. A subset of species to be
simulated may be included in the emission files; zero emission rates will be applied for any
species that are missing.
Meteorological fields are separated into two‐dimensional and three‐dimensional fields. The
core meteorological file contains the basic parameters needed for most model processes, and
includes winds and basic thermodynamic properties. Separate vertical diffusivity and cloud files
allow for easy substitution of alternative inputs without the need to re‐generate all of the core
fields. The time interval on all meteorological files must match the meteorological update
frequency defined in the CAMx control file.
3.4.3.1 Input Static 2‐D Surface File
The static 2‐D surface file contains time‐invariant gridded fields of landuse and topographic
elevation, and optionally leaf area index (LAI). This file must be developed for the master grid,
and optionally any of the nested fine grids. The fractional distribution of 26 landuse categories,
consistent with the “ZHANG03” dry deposition scheme, is supplied for each grid cell. If the
“WESELY89” dry deposition option is invoked, CAMx internally maps the 26 categories to the 11
Wesely categories. The landuse categories are described in Tables 3‐4 and 3‐5. Landuse is used
to define surface UV albedo, surface resistances for dry deposition calculations, and to set
seasonal default surface roughness lengths and LAI values (if LAI is not specified in the file).
Topographic elevation is used to define terrain heights for photolysis calculations.
Table 3‐4. The 11 WESELY89 landuse categories, their default UV surface albedos, and their
surface roughness values (m) by season. Winter is defined for conditions where there is snow
present; winter months with no snow are assigned
to the Fall category. Roughness for water
6 2.5
is calculated from the function z0  210 w , where w is surface wind speed (m/s).
Land Cover Category
1 Urban
2 Agricultural
3 Rangeland
4 Deciduous forest
5 Coniferous forest, wetland
6 Mixed forest*
7 Water
8 Barren land
9 Non‐forested wetlands
10 Mixed agricultural/range**
11 Rocky (with low shrubs)

Surface Roughness (meters)
Spring
1.0
0.03
0.05
1.0
1.3
1.15
f(w)
0.002
0.2
0.04
0.3

Summer
1.0
0.2
0.1
1.3
1.3
1.3
f(w)
0.002
0.2
0.15
0.3

Fall
1.0
0.05
0.01
0.8
1.3
1.05
f(w)
0.002
0.2
0.03
0.3

Winter
1.0
0.01
0.001
0.5
1.3
0.9
f(w)
0.002
0.05
0.006
0.15

UV
Albedo
0.08
0.05
0.05
0.05
0.05
0.05
0.04
0.08
0.05
0.05
0.05

* Roughness for mixed forest is the average of deciduous and coniferous forest.
* Roughness for mixed ag/range is the average of agricultural and rangeland.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

46

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
3. Core Model Input / Output Structures

Table 3‐5. The 26 ZHANG03 landuse categories, their UV albedos, default annual minimum
and maximum LAI and surface roughness (m) ranges, and mapping to the Wesely
scheme
6 2.5
(Table 3‐4). Roughness for water is calculated from the function z0  210 w , where w is
surface wind speed (m/s).
Land Cover Category
1 Water
2 Ice
3 Inland lake
4 Evergreen needleleaf trees
5 Evergreen broadleaf trees
6 Deciduous needleleaf trees
7 Deciduous broadleaf trees
8 Tropical broadleaf trees
9 Drought deciduous trees
10 Evergreen broadleaf shrubs
11 Deciduous shrubs
12 Thorn shrubs
13 Short grass and forbs
14 Long grass
15 Crops
16 Rice
17 Sugar
18 Maize
19 Cotton
20 Irrigated crops
21 Urban
22 Tundra
23 Swamp
24 Desert
25 Mixed wood forest
26 Transitional forest

Wesely
Mapping
7
8
7
5
5
4
4
5
4
3
3
3
3
10
2
2
2
2
2
2
1
11
9
8
6
6

Roughness (meters)
Min
Max
f(w)
f(w)
0.01
0.01
f(w)
f(w)
0.9
0.9
2.0
2.0
0.4
0.9
0.4
1.0
2.5
2.5
0.6
0.6
0.2
0.2
0.05
0.2
0.2
0.2
0.04
0.04
0.02
0.1
0.02
0.1
0.02
0.1
0.02
0.1
0.02
0.1
0.02
0.2
0.05
0.05
1.0
1.0
0.03
0.03
0.1
0.1
0.04
0.04
0.9
0.9
0.9
0.9

LAI
Min
0.0
0.0
0.0
5.0
6.0
0.1
0.1
6.0
4.0
3.0
0.5
3.0
1.0
0.5
0.1
0.1
0.1
0.1
0.1
1.0
0.1
0.1
4.0
0.0
3.0
3.0

Max
0.0
0.0
0.0
5.0
6.0
5.0
5.0
6.0
4.0
3.0
3.0
3.0
1.0
2.0
4.0
6.0
5.0
4.0
5.0
1.0
1.0
2.0
4.0
0.0
5.0
5.0

UV
Albedo
0.04
0.5
0.04
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.05
0.08
0.05
0.05
0.08
0.05
0.05

The data records for the static 2‐D surface file have the following structure:
ibdate,btime,iedate,etime
Loop from l = 1 to nvar variables:
ione,namvar(l),((var(i,j),i=1,nx),j=1,ny)

The variables have the following definitions:
Record 1
ibdate
btime
iedate
etime

Integer beginning date span on file (YYJJJ)
Real beginning decimal hour (e.g., 1:30 PM = 13.5)
Integer ending date span on file (YYJJJ)
Real ending decimal hour (e.g., 1:30 PM = 13.5)

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

47

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
3. Core Model Input / Output Structures

Record 2 through nvar+1
ione
Integer dummy variable (=1)
namvar
Text names for nvar variables (character*4(10,nvar) array):
WATER
Water fraction (Zhang 1)
ICE
Ice fraction (Zhang 2)
LAKE
Lake fraction (Zhang 3)
ENEEDL
Evergreen needle leaf forest fraction (Zhang 4)
EBROAD
Evergreen broad leaf forest fraction (Zhang 5)
DNEEDL
Deciduous needle leaf forest fraction (Zhang 6)
DBROAD
Deciduous broad leaf forest fraction (Zhang 7)
TBROAD
Tropical broad leaf forest fraction (Zhang 8)
DDECID
Drought deciduous tree fraction (Zhang 9)
ESHRUB
Evergreen shrub fraction (Zhang 10)
DSHRUB
Deciduous shrub fraction (Zhang 11)
TSHRUB
Thorn shrub fraction (Zhang 12)
SGRASS
Short grass fraction (Zhang 13)
LGRASS
Long grass fraction (Zhang 14)
CROPS
Cropland fraction (Zhang 15)
RICE
Rice crop fraction (Zhang 16)
SUGAR
Sugar crop fraction (Zhang 17)
MAIZE
Corn crop fraction (Zhang 18)
COTTON
Cotton crop fraction (Zhang 19)
ICROPS
Irrigated cropland fraction (Zhang 20)
URBAN
Urban fraction (Zhang 21)
TUNDRA
Tundra fraction (Zhang 22)
SWAMP
Swamp fraction (Zhang 23)
DESERT
Desert fraction (Zhang 24)
MWOOD
Mixed woodland fraction (Zhang 25)
TFOREST
Transitional forest fraction (Zhang 26)
TOPO_M
Topographic elevation above sea level (m)
LAI
Optional Leaf Area Index
var
Real variable field values for nx grid columns and ny grid rows
3.4.3.2 Input Time‐Variant 2‐D Surface File
The time‐variant 2‐D surface file contains gridded fields of surface temperature and snow
cover. This file must be developed for the master grid, and optionally any of the nested fine
grids. The surface temperature is used for dry deposition calculations and to establish surface‐
layer atmospheric stability. Snow cover includes snow depth and age, which are used to
calculate surface albedo for photochemistry, adjust surface resistances for dry deposition, and
define the snow compartment for the surface chemistry model.
The data records for the time‐variant 2‐D surface file have the following structure and are
repeated for each time interval on file:
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

48

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
3. Core Model Input / Output Structures

ibdate,btime,iedate,etime
Loop from l = 1 to nvar variables:
ione,namvar(l),((var(i,j),i=1,nx),j=1,ny)

The variables have the following definitions:
Record 1
ibdate
Integer beginning date of time interval (YYJJJ)
btime
Real beginning decimal hour (e.g., 1:30 PM = 13.5)
iedate
Integer ending date of time interval (YYJJJ)
etime
Real ending decimal hour (e.g., 1:30 PM = 13.5)
Record 2 through nvar+1
ione
Integer dummy variable (=1)
namvar
Text names for nvar variables (character*4(10,nvar) array):
TSURF_K
Surface temperature (K)
SNOWEW_M
Snow water equivalent depth (m)
SNOWAGE_HR
Snow age since last snowfall (hr)
var
Real variable field values for nx grid columns and ny grid rows
CAMx time‐interpolates surface temperature to each model timestep for each grid (but holds
snow cover constant), and so the model requires that data be available on file for an additional
update time at the end of the simulation. For example, in the case of hourly fields, a 24‐hour
simulation requires 25 input fields on file. The time interval of the data records must match the
time zone and input frequency of the meteorology as specified in the CAMx.in file.
CAMx is backward‐compatible with older 2‐D meteorological files that may contain the snow
cover variable (SNOWCOVER), which is a simple map of 0 or 1 to indicate the presence of snow
in each grid cell. If the SNOWCOVER variable is found, CAMx arbitrarily assumes a snow water
equivalent depth of 0.025 m (~25 cm snow depth) and snow age of 5 days.
3.4.3.3 Input Time‐Variant 3‐D Meteorological File
The time‐variant 3‐D meteorological file contains gridded fields of state meteorological
parameters. This file must be developed for the master grid and optionally any fine grid nest
specified for a given simulation. The layer interface heights define the vertical grid structure for
each grid. The number of vertical layers and the vertical grid definition must be consistent
among all grids in a simulation; otherwise CAMx will stop with an error message if this
condition is not met. The layer interface heights may be specified to vary in space and/or time
(e.g., to follow the layer structure of meteorological models), or they may be set to a constant
field. CAMx allows the user to optionally supply wind components at cell center, in which case
the model will interpolate the components to their respective positions on cell interfaces, or
the user may supply these components directly on the staggered Arakawa C configuration
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

49

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
3. Core Model Input / Output Structures

(recommended). In any case, the user must supply a full nx×ny×nz array of wind values for each
component (even though the CAMx Arakawa C configuration uses only (nx‐1)×(ny‐1) values in
the horizontal). The wind staggering flag is set in the second header record. The pressure,
wind, temperature, and humidity fields are used for transport, plume rise, PiG, dry and wet
deposition, and chemistry calculations.
The data records for the time‐variant 3‐D meteorological file have the following structure and
are repeated for each time interval on file:
ibdate,btime,iedate,etime
Loop from l = 1 to nvar variables:
Loop from k = 1 to nz layers:
ione,namvar(l),((var(i,j,k),i=1,nx),j=1,ny)

The variables have the following definitions:
Record 1
ibdate
Integer beginning date of time interval (YYJJJ)
btime
Real beginning decimal hour (e.g., 1:30 PM = 13.5)
iedate
Integer ending date of time interval (YYJJJ)
etime
Real ending decimal hour (e.g., 1:30 PM = 13.5)
Record 2 through nvarnz+1
ione
Integer dummy variable (=1)
namvar
Text names for nvar variables (character*4(10,nvar) array):
ZGRID_M
Layer interface heights (m AGL)
PRESS_MB
Pressure (mb)
TEMP_K
Temperature (K)
HUMID_PPM
Humidity as mixing ratio (ppm)
UWIND_MpS
U‐component (east‐west) wind (m/s)
VWIND_MpS
V‐component (north‐south) wind (m/s)
Real layer k variable field values for nx grid columns and ny grid rows
var
CAMx time‐interpolates these meteorological variables to each model timestep for each grid,
and so the model requires that data be available on file for an additional update time at the end
of the simulation. For example, in the case of hourly fields, a 24‐hour simulation requires 25
input fields on file. The time interval of the data records must match the time zone and input
frequency of the meteorology as specified in the CAMx.in file.
3.4.3.4 Input Time‐Variant 3‐D Vertical Diffusivity File
The time‐variant 3‐D vertical diffusivity file contains gridded fields of layer‐interface diffusivity
(i.e., turbulent exchange or diffusion coefficients). This file must be developed for the master
grid, and optionally any fine grid nests. This file is kept separate from the main meteorological
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

50

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
3. Core Model Input / Output Structures

data file to facilitate the substitution of alternative vertical mixing rates. Diffusivity is used for
vertical diffusion and PiG puff growth calculations.
The data records for the time‐variant 3‐D meteorological file have the following structure and
are repeated for each time interval on file:
ibdate,btime,iedate,etime
Loop from l = 1 to nvar variables:
Loop from k = 1 to nz layers:
ione,namvar(l),((var(i,j,k),i=1,nx),j=1,ny)

The variables have the following definitions:
Record 1
ibdate
Integer beginning date of time interval (YYJJJ)
btime
Real beginning decimal hour (e.g., 1:30 PM = 13.5)
iedate
Integer ending date of time interval (YYJJJ)
etime
Real ending decimal hour (e.g., 1:30 PM = 13.5)
Record 2 through nvarnz+1
ione
Integer dummy variable (=1)
namvar
Text names for nvar variables (character*4(10,nvar) array):
KV_M2pS
Vertical diffusivity (m2/s)
var
Real layer k variable field values for nx grid columns and ny grid rows
CAMx time‐interpolates the diffusivity to each model timestep for each grid, and so the model
requires that data be available on file for an additional update time at the end of the
simulation. For example, in the case of hourly fields, a 24‐hour simulation requires 25 input
fields on file. The time interval of the data records must match the time zone and input
frequency of the meteorology as specified in the CAMx.in file.
3.4.3.5 Input Time‐Variant 3‐D Cloud/Precipitation File
The time‐variant 3‐D cloud/precipitation file contains gridded fields of cloud and precipitation
parameters to be used for photochemistry, aqueous chemistry, and wet/dry deposition
calculations. Note that precipitation rate is not explicitly provided to the model; instead, it is
internally calculated from the three precipitation water content forms provided on the
cloud/rain file. This file also contains layer‐specific cloud optical depth to scale down photolysis
rates for layers within or below clouds to account for UV attenuation, or to scale up the rates
for layers above clouds to account for UV reflection. This file must be developed for the master
grid, and optionally any fine grid nests, if chemistry, dry, and/or wet deposition are invoked.
The data records for the time‐variant 3‐D cloud/precipitation file have the following structure
and are repeated for each time interval on file:
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

51

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
3. Core Model Input / Output Structures

ibdate,btime,iedate,etime
Loop from l = 1 to nvar variables:
Loop from k = 1 to nz layers:
ione,namvar(l),((var(i,j,k),i=1,nx),j=1,ny)

The variables have the following definitions:
Record 1
ibdate
Integer beginning date of time interval (YYJJJ)
btime
Real beginning decimal hour (e.g., 1:30 PM = 13.5)
iedate
Integer ending date of time interval (YYJJJ)
etime
Real ending decimal hour (e.g., 1:30 PM = 13.5)
Record 2 through nvarnz+1
ione
Integer dummy variable (=1)
namvar
Text names for nvar variables (character*4(10,nvar) array):
CLODW_GpM3
Cloud water content (g/m3)
RAINW_GpM3
Rain water content (g/m3)
SNOWW_GpM3
Snow water content (g/m3)
GRPLW_GpM3
Graupel water content (g/m3)
CLOUDOD
Layer‐specific cloud optical depth
var
Real layer k variable field values for nx grid columns and ny grid rows
The CAMx cloud fields are assumed to be time‐averaged, so the model does not require an
additional update time at the end of the simulation. For example, in the case of hourly fields, a
24‐hour simulation requires only 24 cloud input fields on file. The time interval of the data
records must match the time zone and input frequency of the meteorology as specified in the
CAMx.in file.
3.4.3.6 Input 3‐D Initial Conditions File
The input 3‐D initial conditions file contains gridded concentration fields on the master grid.
Initial concentration fields may be specified for a sub‐set of the total number of modeled
species. An initial condition file must be developed for the master grid, and contain
concentration fields for at least one species. For those species not on the initial condition file,
CAMx sets up uniform fields using the “lower bound” values specified in the chemistry
parameters file. CAMx then interpolates all master grid initial conditions to each fine grid nest
at the start of the simulation.
The data records for the initial conditions file have the following structure and are repeated for
each time interval on file:

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

52

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
3. Core Model Input / Output Structures

ibdate,btime,iedate,etime
Loop from l = 1 to nspec species:
Loop from k = 1 to nz layers:
ione,namspec(l),((conc(i,j,k,l),i=1,nx),j=1,ny)

The variables have the following definitions:
Record 1
ibdate
Integer beginning date of time interval (YYJJJ)
btime
Real beginning decimal hour (e.g., 1:30 PM = 13.5)
iedate
Integer ending date of time interval (YYJJJ)
etime
Real ending decimal hour (e.g., 1:30 PM = 13.5)
Record 2 through nspecnz+1
ione
Integer dummy variable (=1)
namspec
Text names for nspec species (character*4(10,nvar) array):
conc
Real layer k concentration field values (ppm for gases, µg/m3 for aerosols)
for nx grid columns and ny grid rows
The time interval of the data records must match the time zone as specified in the CAMx.in file.
3.4.3.7 Input 3‐D Lateral Boundary Conditions File
The input 3‐D lateral boundary conditions file contains gridded concentration fields on the
lateral faces of the master grid boundary. Boundary concentration fields may be specified for a
sub‐set of the total number of modeled species. However, if a boundary concentration is
specified for a given species, it must be supplied for all four boundaries. A boundary
concentration file must be developed for the master grid, and contain concentration fields for
at least one species. For those species not on the boundary conditions file, CAMx sets up
uniform fields using the “lower bound” values specified in the chemistry parameters file. The
time span of each set of boundary data records may be set arbitrarily; e.g., a set of boundary
conditions may be specified for a six hour span, followed by a set spanning just an hour.
The boundary conditions file adds an additional set of four header records, resulting in a total
of eight header records altogether (note that first four records are identical to the header
records described above):
name,note,ione,nspec,ibdate,btime,iedate,etime
plon,plat,iutm,xorg,yorg,delx,dely,nx,ny,nz,iproj,istag,tlat1,tlat2,rdum
ione,ione,nx,ny
(namspec(l),l=1,nspec)
Loop from 1 to 4 boundaries:
ione,iedge,ncell,(icell(n),idum,idum,idum,n=1,ncell)

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

53

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
3. Core Model Input / Output Structures

The additional header variables have the following definitions:
Records 5‐8
ione
iedge
ncell
icell
idum

Integer dummy variable (=1)
Integer boundary edge number (1=west, 2=east, 3=south, 4=north)
Integer number of rows or columns on this edge
Integer index of first cell modeled (edges 1,3), or last cell modeled (edges
2,4): if “0”, this row/column is omitted from the simulation
Integer dummy variable

The data records for the boundary conditions file have the following structure, and are
repeated for each time interval on file:
ibdate,btime,iedate,etime
Loop from l = 1 to nspec species:
Loop from iedge = 1 to 4 boundaries:
ione,namspec(l),iedge,((bc(i,k,iedge,l),k=1,nz),i=1,ncell)

The variables have the following definitions:
Record 1
ibdate
Integer beginning date of time interval (YYJJJ)
btime
Real beginning decimal hour (e.g., 1:30 PM = 13.5)
iedate
Integer ending date of time interval (YYJJJ)
etime
Real ending decimal hour (e.g., 1:30 PM = 13.5)
Record 2 through nspec4+1
ione
Integer dummy variable (=1)
namspec
Text names for nspec species (character*4(10,nvar) array):
bc
Real edge iedge boundary concentrations (ppm for gases, µg/m3 for
aerosols) for ncell grid rows/columns, and nz layers
The time interval of the data records must match the time zone as specified in the CAMx.in file.
3.4.3.8 Input 2‐D Top Boundary Conditions File
The input 2‐D top boundary conditions file contains gridded concentration fields above the top
of the master grid boundary. Boundary concentration fields may be specified for a sub‐set of
the total number of modeled species; the sub‐set of species may differ from the lateral
boundary conditions. The top boundary concentration file is optional, but if supplied it must
contain concentration fields for at least one species. For those species not on the boundary
conditions file, CAMx sets up uniform fields using the “lower bound” values specified in the
chemistry parameters file. The time span of each set of top boundary data records may be set
arbitrarily; e.g., a set of boundary conditions may be specified for a six hour span, followed by a
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

54

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
3. Core Model Input / Output Structures

set spanning just an hour. The time span of the top boundary conditions may differ from the
lateral boundary conditions.
If the top boundary condition file is not supplied, CAMx will default to internally employing the
original “zero‐gradient” mixing ratio assumption, whereby the concentrations of each species in
the top model layer (as mole pollutant per mole air) is assumed to also exist above the model
top.
The data records for the top boundary conditions file have the following structure and are
repeated for each time interval on file:
ibdate,btime,iedate,etime
Loop from l = 1 to nspec species:
ione,namspec(l),((tc(i,j,l),i=1,nx),j=1,ny)

The variables have the following definitions:
Record 1
ibdate
Integer beginning date of time interval (YYJJJ)
btime
Real beginning decimal hour (e.g., 1:30 PM = 13.5)
iedate
Integer ending date of time interval (YYJJJ)
etime
Real ending decimal hour (e.g., 1:30 PM = 13.5)
Record 2 through nspecnz+1
ione
Integer dummy variable (=1)
namspec
Text names for nspec species (character*4(10,nvar) array):
tc
Real concentration field values (ppm for gases, µg/m3 for aerosols)
for nx grid columns and ny grid rows
The time interval of the data records must match the time zone as specified in the CAMx.in file.
3.4.3.9 Input Elevated Point Source File
The input elevated point source emissions file contains stack parameters and emission rates for
all elevated point sources, and for all emitted species, to be modeled. If elevated point sources
are to be modeled, only one point source emissions file must be developed for the entire
modeling domain. The point source file also flags the individual stacks to be treated by the
CAMx PiG sub‐model by setting the stack diameter as a negative value. The file offers the
ability to optionally specify the effective plume height or the vertical plume distribution for
each point source and to bypass the internal plume rise calculation.
The elevated point source file adds two additional set of header records that specify time‐
invariant stack parameters, resulting in a total of six header records altogether (note that first
four records are identical to the header records described above):
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

55

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
3. Core Model Input / Output Structures

name,note,ione,nspec,ibdate,btime,iedate,etime
plon,plat,iutm,xorg,yorg,delx,dely,nx,ny,nz,iproj,istag,tlat1,tlat2,rdum
ione,ione,nx,ny
(namspec(l),l=1,nspec)
ione,nstk
(xstk(n),ystk(n),hstk(n),dstk(n),tstk(n),vstk(n),n=1,nstk)

The additional header variables have the following definitions:
Record 5
ione
nstk
Record 6
xstk
ystk
hstk
dstk
tstk
vstk

Integer dummy variable (=1)
Integer number of elevated point source stacks
Real stack x‐coordinate (m or degrees longitude)
Real stack y‐coordinate (m or degrees latitude)
Real stack height (m)
Real stack diameter (m); negative value flags source for PiG
Real stack exit temperature (K)
Real stack exit velocity (m/hr)

The time‐variant data records for the elevated point source file have the following structure,
and are repeated for each time interval on file:
ibdate,btime,iedate,etime
ione,nstk
(idum,idum,kcell(n),flow(n),plmht(n),n=1,nstk)
Loop from l = 1 to nspec species:
ione,namspec(l),(ptems(n,l),n=1,nstk)

The variables have the following definitions:
Record 1
ibdate
btime
iedate
etime
Record 2
ione
nstk

Integer beginning date of time interval (YYJJJ)
Real beginning decimal hour (e.g., 1:30 PM = 13.5)
Integer ending date of time interval (YYJJJ)
Real ending decimal hour (e.g., 1:30 PM = 13.5)
Integer dummy variable (=1)
Integer number of elevated point source stacks

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

56

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
3. Core Model Input / Output Structures

Record 3
idum
kcell

Integer dummy variable
Zero or positive: Ignored
Negative: integer flag for OSAT/PSAT source region override
flow
Zero: ignored (plume rise calculation uses time‐invariant exit velocity)
Positive: real stack flow rate (m3/hr) for plume rise calculations
Negative: real plume bottom (m) for vertical plume distribution override
plmht
Zero or positive: ignored (plume rise calculation is performed)
Negative (flow ≥ 0): real effective plume rise override (m)
Negative (flow < 0): real plume top (m) for vertical plume distribution
override
Record 4 through nspec+4
ione
Integer dummy variable (=1)
namspec
Text names for nspec species (character*4(10,nvar) array):
ptems
Real point emission rate (mol/time period for gases, g/time period for
aerosols) for nstk point sources
Note that the emission time interval (the denominator for the emissions rate) is normally, but
not necessarily, 1 hour. The time interval of the emission records must match the time zone
and input frequency of the emissions as specified in the CAMx.in file.
3.4.3.10 Input Gridded Emissions File
The input gridded emissions file contains gridded fields of low‐level (i.e., surface) emission rates
for all emitted species to be modeled. If gridded emissions are to be modeled, a gridded
emissions file must be developed for the master grid and optionally any nested fine grids.
The data records of the gridded emissions file have the following structure, and are repeated
for each time interval on file:
ibdate,btime,iedate,etime
Loop from l = 1 to nspec species:
ione,namspec(l),((emiss(i,j,l),i=1,nx),j=1,ny)

The variables have the following definitions:
Record 1
ibdate
Integer beginning date of time interval (YYJJJ)
btime
Real beginning decimal hour (e.g., 1:30 PM = 13.5)
iedate
Integer ending date of time interval (YYJJJ)
etime
Real ending decimal hour (e.g., 1:30 PM = 13.5)
Record 2 through nspecnz+1
ione
Integer dummy variable (=1)
namspec
Text names for nspec species (character*4(10,nvar) array):
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS
www.camx.com
57

March 2016

emiss

CAMx User’s Guide Version 6.3
3. Core Model Input / Output Structures

Real gridded emission field values (mol/time period for gases, g/time
period for aerosols) for nx grid columns and ny grid rows

Note that the emission time interval (the denominator for the emissions rate) is normally, but
not necessarily, 1 hour. The time interval of the emissions record must match the time zone
and input frequency of the emissions as specified in the CAMx.in file.
3.4.4 Output Files
Master and nested grid instantaneous files contain full three‐dimensional fields of all species
modeled, while the gridded average and deposition files contain only those species specified in
the run control file. For flexibility, CAMx offers the option to write full three‐dimensional
average concentration fields, or just surface layer two‐dimensional fields (see the “3‐D average
file” option in the CAMx control file). It is permissible to change the number of species on the
average file, or change between 2‐D and 3‐D average files, from one CAMx simulation period to
the next (provided the periods are configured as separate CAMx runs). As the instantaneous
concentration files are used for CAMx restarts, the model only writes instantaneous fields at
the end of the simulation. PiG output files possess a unique format, and are used primarily for
model restarts.
3.4.4.1 Output Concentration Files
The output average files for all grids, and the coarse (master) grid instantaneous file, are all
written in the CAMx Fortran binary format as described earlier. There are three differences
between the output concentration files and the input initial concentration files. First, the file
name given in the file description header record (header record #1) is “AVERAGE” for the
average output file, “INSTANT” for the instantaneous output file, and “AIRQUALITY” for the
input initial concentration file. Second, the “note” in the file description header record of the
output concentration files is the message supplied in the first line of the CAMx run control file,
whereas the “note” in the air quality file is set as part of the input file preparation. Third, the
species lists can be different among the files: the output instantaneous file contains all species
modeled (as specified in the chemistry parameters file), the average output file contains only
the species specified in the run control file, and the input initial concentration file may contain
any subset of modeled species as determined when that file is prepared.
Two other differences exist between the average and instantaneous output files. As noted
above, the average file may contain only surface‐level fields or the entire three‐dimensional
fields, as selected by the user. Also, gas concentration fields are output as ppm in average files,
but as µmol/m3 in instantaneous files (aerosols are in µg/m3 in both files). Because of these
differences, and because average files usually do not contain all modeled species, CAMx does
not allow the average output concentration file to be used for simulation restarts.
3.4.4.2 Output Deposition Files
The output deposition file is identical in format to the two‐dimensional surface‐level output
average concentration file. The file name given on the first record of the deposition file is
“AVERAGE” so that existing post‐processing software will recognize the format. However, the
output deposition file differs from the output average concentration file in one important way.
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

58

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
3. Core Model Input / Output Structures

The species list is identical to the list on the average concentration output files, except that four
parameters are output for each species:
species_DV
species_DD
species_WD
species_LC

Real 2‐D dry deposition velocity field for species (m/s)
Real 2‐D dry deposited mass field for species (mol/ha for
gases, g/ha for aerosols)
Real 2‐D wet deposited mass field for species (mol/ha for
gases, g/ha for aerosols)
Real 2‐D precipitation liquid concentration for species (mol/l for
gases, g/l for aerosols)

3.4.4.3 Output Surface Mass Files
The output surface model mass file is identical in format to the two‐dimensional surface‐level
output average concentration and deposition files. The file name given on the first record of
the deposition file is “AVERAGE” so that existing post‐processing software will recognize the
format. However, the contents of the surface mass file differ from the other files in two ways.
First, the species list is defined from the section of the chemistry parameters file that explicitly
lists the species to be tracked by the surface model (see Section 4.8). Second, two parameters
are output for each surface model species:
Sspecies
Vspecies

Real 2‐D dry mass on soil or snow for species (mol/ha)
Real 2‐D dry mass on vegetation for species (mol/ha)

This file is also used for restarts to re‐initialize the surface model with accumulated species
mass on soil/snow and vegetative surfaces from a previous run.
3.4.4.4 Nested (Fine) Grid Instantaneous Output File
The nested (or “fine”) grid Fortran binary output instantaneous file is unique and contains the
three‐dimensional concentration fields for all nested grids together, as opposed to separate
files for each grid. All grid definition parameters given in these files are referenced relative to
the master grid, so specific absolute information about grid cell size or projection coordinates
for each nested grid must be determined from master grid parameters. If the user utilizes the
Flexi‐nesting capability of CAMx, then the gridded fields output to the nested fine grid files will
change according to how nests are altered, added, and/or removed during the course of a
simulation.
The header portion of the nested grid output files contain 3+nnest records with the following
structure:
message
nnest,nspec
(mspec(l),l=1,nspec)
Loop from 1 to nnest grid nests
ibeg,jbeg,iend,jend,mesh,ione,nx,ny,nz,iparnt,ilevel
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

59

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
3. Core Model Input / Output Structures

The time‐variant portion of the nested grid output files have the following structure, where the
following records are repeated for each output time:

time,idate
Loop from 1 to nnest grid nests
Loop from l = 1 to nspec species:
Loop from k = 1 to nz layers:
((conc(i,j,k,l),i=1,nx),j=1,ny)

The variables on the nested grid output concentration files have the following definitions:
message
nnest
nspec
mspec
ibeg
jbeg
iend
jend
mesh
ione
nx
ny
nz
iparnt
ilevel
time
idate
conc

Text string containing file description (character*60)
Number of fine grid nests on file
Number of species on file
Species names for nspec species
Grid n x‐direction starting index of grid (master grid cell)
Grid n y‐direction starting index of grid (master grid cell)
Grid n x‐direction ending index of grid (master grid cell)
Grid n y‐direction ending index of grid (master grid cell)
Grid n meshing factor (number of nested cells per master)
Dummy integer = 1
Grid n number of grid rows
Grid n number of grid columns
Grid n number of layers
Grid n’s parent grid (grid index within which this fine grid is nested; 0 =
master grid)
Grid n’s grid level (depth at which this grid is nested; 1=master grid is
parent)
Time of output (HHMM); ending hour for average output
Date of output (YYJJJ)
Grid n, species l, layer k concentrations (ppm for average gases, µg/m3
for average aerosols, µmol/m3 for instantaneous gas species) for nx grid
columns, and ny grid rows

3.4.4.5 PiG Restart File
When the PiG option is invoked, CAMx outputs all puff parameters each hour for model restart
capabilities. This file is Fortran binary and is analogous to the instantaneous gridded
concentration output files in that it represents a “snapshot” of data at the top of each hour.
The file format is unique and contains information for each puff, including coordinates, grid
location, size specifications, age, and mass of each of the chemical species carried. While this
file contains PiG information for the entire simulation, it would be of limited use for certain
analyses such as plotting puff trajectories. This is because the instantaneous nature of the data,
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

60

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
3. Core Model Input / Output Structures

and the dynamic memory allocation utilized in the PiG submodel, leads to insufficient
information to identify and track individual puffs hour to hour.
The PiG restart file contains two records with the following structure, and these are repeated
for each output time:
idatpig,timpig,npig,nreactr
(ingrd(n),idpig(n),xpigf(n),xpigb(n),ypigf(n),ypigb(n),zpig(n),
&
axisy(n),axisz(n),sigy(n),sigx(n),sigz(n),pufftop(n),puffbot(n),
&
htfms(n),htfmb(n),vtfms(n),vtfmb(n),agepigf(n),agepigb(n),fmspig(n),
&
ipufmap(n),ipufgrp(n),
&
((puffrad(i,nr,n),i=1,nrad),nr=1,nreactr),
&
((puffmass(i,nr,n),i=1,nspec),nr=1,nreactr),n=1,npig

The variables on the PiG restart file have the following definitions:
idatpig
timpig
npig
nreactr
ingrd
idpig
xpigf
xpigb
ypigf
ypigb
zpig
axisy
axisz
sigy
sigx
sigz
pufftop
puffbot
htfms
htfmb
vtfms
vtfmb
agepigf
agepigb
fmspig
ipufmap
ipufgrp

Date of output (YYJJJ)
Time of output (HHMM)
Number of PiG puffs active at this output time
Number of chemical reactors in each puff
Grid index for npig puffs
Point source index for npig puffs
x‐coordinate of puff front (km from master grid SW corner) for npig puffs
x‐coordinate of puff back (km from master grid SW corner) for npig puffs
y‐coordinate of puff front (km from master grid SW corner) for npig puffs
y‐coordinate of puff back (km from master grid SW corner) for npig puffs
Puff height (m AGL) for npig puffs
Puff lateral width (m) for npig puffs
Puff vertical depth (m) for npig puffs
Puff Gaussian lateral dimension (m) for npig puffs
Puff Gaussian longitudinal dimension (m) for npig puffs
Puff Gaussian vertical dimension (m) for npig puffs
Puff top height (m AGL) for npig puffs
Puff bottom height (m AGL) for npig puffs
Puff horizontal turbulent flux moment, shear (m2/s)
Puff horizontal turbulent flux moment, buoyancy (m2/s)
Puff vertical turbulent flux moment, shear (m2/s)
Puff vertical turbulent flux moment, buoyancy (m2/s)
Puff front age since release (s) for npig puffs
Puff back age since release (s) for npig puffs
Puff volume parameter (unitless) for npig puffs
Puff OSAT/PSAT region map pointer (unitless) for npig puffs
Puff OSAT/PSAT group pointer (unitless) for npig puffs

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

61

www.camx.com

March 2016

puffrad
puffmass

CAMx User’s Guide Version 6.3
3. Core Model Input / Output Structures

Puff radical concentrations (ppm) for nrad species, nreactr reactors, and
npig puffs
Puff pollutant mass (µmol) for nspec species, nreactr reactors, and npig
puffs

3.4.4.6 PiG Sample Grid Files
The optional PiG sampling grid concentrations are time‐averaged in the same manner as the
output average concentrations provided on the computational grids. The same user‐defined
set of output species are written to the sampling grid files, but only two‐dimensional surface
layer concentrations are reported. The sampling grid file format is identical to the CAMx
average and deposition files, with one file generated per sampling grid, so that they may be
readily viewed and manipulated with CAMx post‐processing software.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

62

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
4. Core Model Formulation

4. CORE MODEL FORMULATION
This section outlines the numerical approach employed in the core CAMx model, and describes
the technical formulation of the emissions, transport and removal algorithms. The specific
chemical mechanisms and their numerical solvers are discussed in detail in Section 5.
Descriptions of Plume‐in‐Grid and each Probing Tool are provided in Sections 6 through 10.

4.1 Numerical Approach
The physical representations and the numerical methods used for each term of the pollutant
continuity equation (described in Section 1) are summarized in Table 4‐1. CAMx includes peer‐
accepted algorithms and component formulations, and its modular framework eases the
addition and/or substitution of alternative algorithms in the future.

Table 4‐1. Summary of the CAMx models and methods for key physical processes.
Process

Physical Models

Numerical Methods
 Bott (1989)
 PPM (Colella and Woodward,1984)
Explicit simultaneous 2‐D solver
Implicit backward‐Euler (time) hybrid
centered/upstream (space) solver
 Implicit backwards‐Euler (time)
centered (space) solver
 Explicit ACM2 non‐local
convection/diffusion (Pleim, 2007)

Horizontal advection

Eulerian continuity equation

Horizontal diffusion

K‐theory 1st order closure

Vertical advection

Eulerian continuity equation

Vertical diffusion

 K‐theory 1st order closure
 Non‐local mixing

Gas‐Phase Chemistry

 Carbon Bond 2005 (Yarwood et al., 2005b)
 Carbon Bond 6 (Yarwood et al., 2010,
2012a, 2014; Hildebrandt Ruiz and
Yarwood, 2013; Emery et al., 2015)
 SAPRC07TC (Carter, 2010; Hutzell et al.,
2012)
 Inorganic/organic aerosol precursors

 EBI (Hertel et al., 1993)
 LSODE (Hindmarsh, 1983)






 RADM‐AQ (Chang et al., 1987)
 ISORROPIA (Nenes et al., 1998)
 SOAP (Strader et al., 1999) or
VBS (Koo et al., 2014)
 Coarse/Fine (CF) 2‐mode model
 CMU sectional model (Pandis et al.,
1993)

Aerosol Chemistry

Dry deposition

Wet deposition

Aqueous inorganic chemistry
Inorganic thermodynamics/partitioning
Organic thermodynamics/partitioning
Static 2‐mode or multi‐section size models

 Resistance model for gases (Wesely, 1989)
and aerosols (Slinn and Slinn, 1980)
 Resistance model for gases (Zhang et al.,
2003) and aerosols (Zhang et al., 2001)
Scavenging model for gases and aerosols
(Seinfeld and Pandis, 1998)

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

63

Deposition velocity as surface
boundary condition in vertical
diffusion solver
Exponential decay as a function of
scavenging coefficient

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
4. Core Model Formulation

The continuity equation is numerically marched forward in time over a series of time steps. At
each step, the continuity equation is integrated by way of an operator‐splitting approach that
calculates the separate contribution of each major process (emission, advection, diffusion,
chemistry, and removal) to concentration change within each grid cell. The specific equations
that are solved individually in the operator‐splitting process are shown in order below:

cl
t Emission

 m2

cl
t X advection

m 2   uAyz cl

 
Ayz x  m

cl
t Y advection

 

cl
t Z transport



 cl η
 2h
 cl
z
z t

cl
t Z diffusion



 cl /ρ  
 
ρK v

z 
z 

cl
t XY diffusion


 m
 x

cl
t Wet Scavenging

   l cl

cl
t Chemistry

 Mechanism  specific reaction equations

El
x y z






m 2   vAxz cl 


Axz y  m 

 cl /ρ   
 cl /ρ  
 

m
K
m
K




X
Y


y 
y  
x 


where cl is species concentration (mol/m3 for gasses, g/m3 for aerosols), El is the local species
emission rate (mol/s for gasses, g/s for aerosols), t is timestep length (s), u and v are the
respective east‐west (x) and north‐south (y) horizontal wind components (m/s), Ayz and Axz are
cell cross‐sectional areas (m2) in the y‐z and x‐z planes, respectively, m is the ratio of the
transformed distance on the various map projections to true distance (m=1 for curvi‐linear
latitude/longitude coordinates), and Λl is the wet scavenging coefficient (s‐1).
A master driving time step for the model is dynamically determined during the simulation for
the largest and coarsest (master) grid. Time steps typically range from 5‐15 minutes for grid
cell sizes of 10‐50 km, to a minute or less for small cell sizes of 1‐2 km. As a result, nested grids
require multiple driving time steps per master step depending on their grid sizes relative to the
master grid spacing. Furthermore, multiple transport and chemistry time steps per driving step
are used as necessary to ensure accurate solutions for these processes on all grids.
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

64

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
4. Core Model Formulation

The first process in each time step for a given grid is the injection of emissions from all sources.
CAMx then performs horizontal and vertical advection, vertical diffusion coupled to dry
deposition, horizontal diffusion, chemistry, and finally wet scavenging.
Although advection is performed separately in the x (east‐west), y (north‐south), and z (vertical)
directions, the numerical linkage between these components has been developed in a mass
consistent fashion to preserve the density field at each time step. This maintains the flexibility
to allow many types of meteorological models, and modeling grid resolutions, projections, and
layer structures, to characterize transport in CAMx.
Trace gases and particles are removed from the atmosphere via deposition to the ground. Dry
deposition refers to the direct sedimentation and/or diffusion of material to various terrestrial
surfaces and uptake into biota. Dry deposition velocities for each species are used as the
surface boundary condition for vertical diffusion, which appropriately couples surface removal
of pollutants through each column of cells via the vertical mixing process. Wet deposition
refers to the uptake of material into cloud water and precipitation, and its subsequent transfer
to the surface. The efficiency with which wet and dry deposition processes can remove
pollutants from the air depends upon the physical and chemical properties of the pollutants,
local meteorological conditions, the type of surface on which they are being deposited, and on
the frequency, duration, and intensity of precipitation events.

4.2 CAMX Grid Configuration
4.2.1 Grid Cell Arrangement
CAMx carries pollutant concentrations at the center of each grid cell volume, representing the
average concentration over the entire cell. Meteorological fields are supplied to the model to
quantify the state of the atmosphere in each grid cell for the purposes of calculating transport,
chemistry, and removal. CAMx internally carries these variables in an arrangement known as
an “Arakawa C” grid configuration (Figure 4‐1). State variables such as temperature, pressure,
water vapor, and cloud water are located at cell center along with pollutant concentration, and
represent grid cell average conditions. Wind components and diffusion coefficients are carried
at cell interfaces to describe the transfer of mass in and out of each cell face. Note in Figure 4‐
1, for example, that horizontal wind components u and v are staggered from each other. This
facilitates the solving of the transport equations in “flux form”.
Depending upon the source of meteorological data, it is recommended that the user directly
provide the gridded horizontal wind fields in the staggered Arakawa C configuration. However,
this is not always feasible, and so CAMx offers the option for the user to supply all
meteorological variables, including horizontal wind components, at cell center; in this case
CAMx internally interpolates the winds to cell interfaces. Note that this leads to a slight
smoothing effect on the horizontal wind fields.
Figure 4‐1 also describes the horizontal cell indexing convention used in CAMx. Each cell is
defined by the index pair (i,j), where i ranges from 1 to nx (the number of cells in the east‐west
direction), and j ranges from 1 to ny (the number of cells in the north‐south direction). The
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

65

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
4. Core Model Formulation
v, Ky

(i, j)

u, Kx

(i-1, j)

T, p, q, Kv

(i, j)

u, Kx

(i, j)

(i, j-1)

v, Ky

Figure 4‐1. A horizontal representation of the Arakawa C variable configuration used in
CAMx.
eastern and northern faces of the cell are indexed (i,j), while the western and southern faces
are indexed (i‐1,j) and (i,j‐1), respectively.
In the vertical, most variables are carried at each layer midpoint (defined as exactly half way
between layer interfaces) to represent layer averages. Again the exceptions are those variables
that describe the rate of mass transport across the layer interfaces, which include the vertical
diffusion coefficient KV and the vertical transport rate . These variables are carried in the
center of each cell horizontally, but are located at the top of the layer (i.e., the interface)
vertically.
4.2.2 Grid Nesting
CAMx incorporates two‐way grid nesting, which means that pollutant concentration
information propagates into and out of all grid nests during model integration. Any number of
grid nests can be specified in a single run, where horizontal grid spacing can vary from one grid
nest to another (note that vertical grid structures must be consistent among all grids). The
nested grid capability of CAMx allows cost‐effective application to large regions in which
regional transport occurs, yet at the same time providing fine resolution to address small‐scale
impacts in selected areas.
Each grid nest is defined over a subset of master (coarsest) grid cells. The range of master grid
row and column indices that define the coverage of each nested grid must be specified in the
run control file. An integer number of nested grid cells must span one master grid cell; this
number is referred to as a “meshing factor”. “Buffer” cells are added around the perimeter of
each nested grid to hold lateral boundary conditions. Buffer cells are added automatically
within CAMx and should not be specified by the user in the run control file. All nested grid
output files contain data for the entire array of computational and buffer cells; however, buffer
cell concentrations are considered invalid and should be ignored. Additionally, all nested grid
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

66

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
4. Core Model Formulation

Coarse Grid Boundary Cells

10

9
Fine Buffer Cells

4

Coarse Grid Boundary Cells

5

Fine Buffer Cells

6

Fine Buffer Cells

7

Coarse Grid Boundary Cells

8

Boundary of Fine Grid 2

Fine Buffer Cells

3

Boundary of Fine Grid 1

2

Coarse Grid Boundary Cells

1
1

2

3

4

5

6

7

8

9

10

Figure 4‐2. An example of horizontal grid nesting, showing two telescoping nested grids
within a 10×10 cell master grid. The outer nest contains 10×12 cells (including buffer cells to
hold internal lateral boundary conditions), and the inner nest contains 6×10 cells (including
buffer cells).
input files must contain data for the entire array of computational and buffer cells. An example
of a horizontal nesting arrangement is shown in Figure 4‐2. Here, two telescoping fine grid
nests are defined: one with a meshing factor of 2 spanning master grid cells (5,4) to (8,8), and
one with a meshing factor of 4 spanning master grid cells (6,6) to (6,7).
Restrictions on specifying the size and resolution of all grid nests include the following:
1) The ratio of master grid cell size to nested grid cell size must be an integer (e.g., a
“meshing factor” of 3 means that 3 nested cells span the distance of 1 master cell,
resulting in an area of 9 nested cells per master cell);
2) For telescoping grids (a nested grid containing an even finer grid), the cell size of the finest
grid must be a common denominator for all parent grids above it (e.g., a 36‐12‐4 km or
36‐12‐2 km arrangement is allowed, but a 36‐12‐9 km is not);
3) The restriction in (2) above does not apply to parallel nested grids of the same generation
(e.g., 4 km and 5 km grids can be located in different areas of a master grid provided that
the master cell size is some multiple of 20 km);
4) Nested grids cannot overlap, although they may share a common lateral boundary or
edge;
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

67

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
4. Core Model Formulation

5) Nested grids cannot extend into a lateral boundary, or non‐modeled, area of the master
grid;
6) CAMx is currently configured to allow four “generations” of nests (e.g., four levels of
telescoping grids); this can be extended in the code if more than four levels of nests are
required;
7) The total vertical depth of each nested grid must exactly match the depth of the master
grid, and nested grid vertical layer structures must be consistent with the master grid in
terms of the number of layers and their thicknesses;
4.2.3 Flexi‐Nesting
The following Fortran binary I/O files must be provided for the master grid, and optionally
provided for each nested grid:



2‐D surface emissions;



2‐D time‐invariant surface characteristics (landuse and LAI distribution);



2‐D time‐variant surface meteorology (surface temperature, snow cover);



3‐D time‐variant state meteorology (wind, temperature, pressure, moisture, vertical grid);



3‐D time‐variant cloud and precipitation variables;



3‐D time‐variant vertical diffusivities

Any of these input files may be supplied for each nested grid, or none at all. If any of these files
are not supplied for a particular nested grid, the Flexi‐Nest algorithm within CAMx interpolates
the missing fields from the parent grid. Clearly it is desirable to provide nested grid data
whenever possible. However, the ability to interpolate data is useful for testing sensitivity to
grid configurations or for situations when it is not possible to run a meteorological model for all
grid nests.
The Flexi‐Nest option also allows users to redefine the nested grid configuration at any point in
a simulation. Nested grids can be introduced or removed only at the time of a model restart
since a new CAMx user control file must be used to redefine the grid configuration. For
example, the user may wish to “spin‐up” the model over the first two days using just the master
grid. On the third day, the user might introduce one or more nests for more detailed analysis.
This would require that the model be restarted on the third day with a new control file that
defines the position of the new nests and (optionally) provides any additional input fields for
these grids. CAMx will internally reconcile the differences in grid structure between the restart
files and the new user control file, and then interpolate any data fields not supplied to CAMx for
the new nests from the parent grid(s).

4.3 Treatment of Emissions
Pollutant emissions are treated in two basic ways within CAMx: low‐level (gridded) emissions
that are released into the lowest (surface) layer of the model; and elevated stack‐specific
(point) emissions with buoyant plume rise that can be emitted into any model layer. Emission
rates are held constant (not time interpolated) between reading intervals (usually 1 hour) but
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

68

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
4. Core Model Formulation

are injected at every grid‐specific time step during the simulation. Gridded and point emissions
are provided to CAMx in separate input files. External emission processing systems are used to
develop gridded and point, time‐ and space‐resolved, chemically‐speciated input files for CAMx.
These external programs are not discussed in this User’s Guide; see Section 2 for more
information on emission systems that are used to support CAMx applications.
4.3.1 Gridded Emissions
Two‐dimensional gridded low‐level emissions are defined by space‐ and time‐varying rates for
each individual gas and PM species to be modeled. Gridded emissions represent sources that
emit near the surface and that are not sufficiently buoyant to reach into the upper model
layers. Such emission categories include:



Low‐level stack (point) emissions that are too small to result in plume rise above the
model surface layer;



Other non‐point industrial sources (fugitive leaks, tanks, etc.);



Mobile sources (cars, trucks, non‐road vehicles, railroad, marine, aircraft, etc.);



Residential sources (heating, cooking, consumer products);



Commercial sources (bakeries, refueling stations, dry cleaners);



Biogenic sources;



Natural sources (small fires, wind‐blown dust).

The spatial distribution of each individual source within these categories is defined by the
modeling grid. Information such as population distribution, housing density, roadway
networks, vegetative cover, etc. is typically used as a surrogate to distribute regional emission
estimates for each source to the grid system. Processing tools are used to combine emissions
from all sources into a single input file for each grid (see Sections 2 and 3).
4.3.2 Elevated Point Emissions
Similarly to gridded emissions, elevated point emissions are defined by space‐ and time‐varying
rates for each individual gas and PM species to be modeled. The only difference is that these
sources emit from individual stacks with buoyant rise that may take them into upper model
layers. These types of sources are almost always associated with large industrial processes,
such as electric generators, smelters, refineries, large factories, etc. but can also represent
natural elevated sources such as wildfires and lightning NOx. The spatial distribution of these
points is specifically given by the coordinates of the stacks themselves (grid locations are
determined within CAMx). Plume rise is determined within CAMx as a function of stack
parameters (height, diameter, exit velocity and temperature) and ambient meteorological
conditions, so the point source file provides speciated time‐resolved emission rates and stack
parameters for each individual source. A single point source file provides the definition of all
stacks and their emissions over the entire modeling domain (see Sections 2 and 3).
Plume rise is calculated using the multi‐layer stability‐dependent algorithm of Turner et al.
(1986). This approach calculates the momentum and buoyant plume rise energy from the
stack, takes the larger of these two values, and determines the dissipation of that energy via
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS
www.camx.com
69

March 2016

CAMx User’s Guide Version 6.3
4. Core Model Formulation

mixing with ambient air according to the meteorological conditions through the host model
layer. If sufficient energy remains to reach into the next model layer, the calculation for
buoyant rise repeats for the meteorological conditions of that layer, and so on, until a layer is
found where the plume cannot rise any farther. All emissions from this source are then
injected into the grid cell directly above the stack at this layer height. This algorithm was
adopted for CAMx because it provides a more realistic handling of stable layers aloft that can
trap plume rise, whereas this effect would not be realized based on meteorological conditions
at stack top alone.
Under neutral/unstable conditions, momentum rise at the stack top is calculated from

3 d s vs
w



H mu
while buoyancy rise is the lesser of

30  f / w 

3/ 5



H bu

 zb

and
H bu





24 f / w 3

 h
3/5

s



 200 f / w 3



2/5

 zb

In these expressions ds is stack diameter (m), vs is stack exit velocity (m/s), hs is stack height (m),
w is ambient wind speed (m/s), and zb is the distance between the stack top and the base of the
current model layer. A minimum wind speed of 1 m/s is specified to avoid unrealistically large
plume rise. Buoyancy flux f is initially calculated from stack parameters, but is set to residual
flux entering the bottom of any higher layer. The initial buoyancy flux at stack top is given by

T T 

g vs d s2  s
4
T
s 




f0

where g is gravitational acceleration (9.8 m2/s), Ts is stack exit temperature (K), and T is
ambient temperature (K). The residual flux calculation into the next higher layer depends on
which neutral/unstable buoyancy rise equation was originally used in the current layer:

 H  zt 
 w  bu

 30 

f
or

f

 5.5  10 w H bu
3

3

5/3


hs
 zt 1 
H bu  zt






2 / 3

where zt is the distance between stack top and the top of the current model layer.
Under stable conditions, momentum rise at the stack top is calculated from the lesser of Hmu
and
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

70

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
4. Core Model Formulation

1/ 3

 0.646 T

H ms

1/ 2

 vs2 d s2    
 


 Ts w   z 

1 / 6

while buoyancy rise is the lesser of

1.8 f T    1

3
 

  zb 

 w  z 

H bs

1/ 3

and

 4.1 f T    1

8/3
  1/ 3 
  zb 
 z 
 f 0


H bs

3/8

In these expressions, /z is the potential temperature gradient (a measure of atmospheric
stability). Here also, buoyancy flux is initially calculated from stack parameters, but is set to
residual flux if the plume extends into the next higher layer. The residual flux calculation
depends on which stable buoyancy rise equation was originally used:



f

f  0.56

or

f



f  0.24

 w 3
zt  zb3
z T





 f 01 / 3 8 / 3
z t  z b8 / 3
z T





When final plume rise is reached using stable buoyancy rise, it is adjusted downward to two‐
thirds of the rise through the stable depth. After final plume rise is determined, the rise is
further adjusted downward by stack tip downwash according to a critical Froude number and
ambient wind speed. The stack Froude number is given by

F



T vs2
g d s Ts  T 

For F<3, no downwash adjustment is made to final plume rise. Above that value, the following
downwash factors (D) are applied depending upon the ambient wind speed at stack top:

D  0,
vs  w no plume rise 
vs  w
2
vs  w  v s
D  3
,
vs
3
2
D  1,
w  vs
3
CAMx injects point source emissions into all model layers spanned by the plume depth at final
(adjusted) rise. Plume depth is determined as a function of stack diameter, plume
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

71

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
4. Core Model Formulation

temperature, plume velocity, time of plume rise, and ambient wind and temperature
conditions. A uniform mass distribution through plume depth is assumed. If this depth is
wholly contained within a single layer, that layer receives all emission mass. If this depth spans
several layers, then fractions of the emissions are injected into these layers according to the
fraction of plume depth spanning those layers. We apply the “rule‐of‐thumb” that plume depth
equals plume rise (Turner and Schulze, 2007) as a maximum limit for plume depth.
The following equations are used to define the plume depth after reaching final rise. These are
based on the approach used in the SCIPUFF model (EPRI, 2000) and were developed for use in
the CAMx plume‐in‐grid (PiG) submodel. The plume depth Dp at final rise is given by



 3 2 Ds 2  2 K t

Dp

1/ 2

where Ds is stack diameter, K is plume diffusivity during rise, and t is the time of rise. The time
of rise is determined by dividing final plume rise by the mean plume rise speed Vp; the latter is
set to half the stack exit velocity. A lower limit of 1 m/s is applied to the exit velocity, so the
minimum value of Vp is 0.5 m/s. The plume diffusivity is determined by scaling initial plume
width (according to stack diameter) by the turbulent flux moment qp2:
K

 0.15



2 Ds



q 2p

where

q 2p







2 
 v  V p 


3v
f p V p2 0.4  
2




2

The turbulent flux moment is a function of the mean plume rise speed Vp, the ambient wind
speed v taken at the level of final rise, and a plume entrainment coefficient fp:

fp

 Tp  T 

 1  4 2 Ds g 
 TV2 
p 






where g is the gravitational constant (9.8 m2/s), T is ambient temperature at the level of final
rise, and Tp is the mean plume temperature, taken as the mean of the stack exit temperature
and the ambient temperature at final rise.

4.4 Three‐Dimensional Transport
4.4.1 Resolved Transport: Advection
The CAMx advection algorithm is both mass conservative and mass consistent. Mass
conservation refers to the ability to accurately account for all sources and sinks of mass in the
model, with no spurious loss or gain of mass during model integration. To be mass
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

72

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
4. Core Model Formulation

conservative, CAMx internally carries concentrations of each species as a density (µmol/m3 for
gases, µg/m3 for aerosols), and solves the advection and diffusion equations in flux form. This
also serves to simplify mass budget accounting, which is used by the various source
apportionment and process analysis options. Gas concentrations are internally converted to
volumetric mixing ratio (parts per million, or ppm) for the chemistry step, and when they are
written to the average output files.
Mass consistency refers to the model’s ability to transport pollutant mass exactly equivalent to
the input atmospheric momentum field. For example, a model that is perfectly mass consistent
will preserve a unity pollutant mixing ratio field in a divergent momentum field given constant
unity boundary and initial conditions and zero sources and sinks. Sources of poor mass
consistency in air quality models are typically related to (1) supplying meteorology that is
inherently inconsistent (e.g., from an interpolative objective analysis or diagnostic model); (2)
spatially interpolating or averaging meteorological model fields to a different air quality map
projection or grid resolution; and (3) employing different numerical and/or physical methods in
the air quality and meteorological models.
It is expected that CAMx users will prepare high quality, mass consistent meteorological fields
using advanced prognostic models so as to minimize inconsistencies in the inputs themselves.
The practice of developing meteorological input fields using objective analysis or “diagnostic”
approaches is highly discouraged.
CAMx operates on the map projections and grid systems employed in several widely used
prognostic meteorological models (e.g., WRF, MM5, and RAMS) so that translation of
meteorological data to CAMx requires as little manipulation as possible. However, CAMx
provides a very important flexibility that allows the air quality grid to differ in projection and
resolution from the source of meteorological data. This, of course, leads to a potentially large
external source of mass consistency error. The ability to drive CAMx with the output from any
prognostic meteorologic model guarantees a difference in numerical methods between the two
models, leading to an internal source of mass consistency error. The three diminensional
advection algorithm in CAMx is designed to compensate for both external and internal sources
of consistency error.
Horizontal advection uses input horizontal winds fields and is solved using the area preserving
flux‐form advection solver of Bott (1989) or the Piecewise Parabolic Method (PPM) of Colella
and Woodward (1984) as implemented by Odman and Ingram (1993). These two finite
difference schemes were incorporated into CAMx because they provide higher order accuracy
with minimal numerical diffusion, yet are equivalent in execution speed compared to other
simpler advection algorithms when operating on equivalent time steps. In CAMx, the Bott
scheme is allowed to take larger time steps than PPM because Bott remains stable for Courant‐
Friedrichs‐Levy (CFL) numbers up to 1 (i.e., the ratio of wind speed to grid spacing). Time steps
are determined for Bott using a CFL number of 0.9, while time steps for PPM are restrained by a
CFL number of 0.5. Therefore, the Bott option results in a faster simulation than the PPM
option, perhaps at the price of some accuracy. We recommend testing both for your specific
application.
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

73

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
4. Core Model Formulation

CAMx internally calculates an instantaneous vertical velocity field from grid‐ and timestep‐
specific horizontal momentum fields as a way to balance the local atmospheric continuity
equation. To calculate a vertical velocity profile for a given grid column, the divergent
atmospheric continuity equation
ρ
t

    ρV

is locally integrated through the depth of the column by imposing zero vertical velocity at the
ground:
ρw  z  

z

 ρ

 
  H  ρV H  dz
t

0

where the local time‐rate change of atmospheric density /t in each grid cell is known from
the input meteorological fields. The horizontal flux divergence of atmospheric density is
calculated in a manner that is numerically consistent with the procedure used to horizontally
advect pollutants:
 H  ρV H



m 2   uA yz  

 
Ayz x  m 

m 2   vAxz  


Axz y  m 

and this equation is solved using either the Bott or PPM advection solvers, as described above.
In this approach, a vertical velocity profile w(z) is constructed that provides a balance between
the imposed density tendency and the resolved horizontal momentum divergence in each grid
cell at each time step. Total three‐dimensional advection thus includes resolved momentum
convergence/divergence rates as well as any artificial divergences caused by the horizontal and
vertical grid specifications (e.g., spatially varying vertical grid structure, or systematic
distortions associated with the map projections).
The total vertical transport rate  at a particular layer interface is defined as the combination of
resolved vertical velocity and the local time‐rate of change of the layer interface height:





h
t



w

The total vertical transport rate is used for subsequent vertical advection calculations for all
pollutants. In simple tests in which a uniform pollutant field of unity mixing ratio is transported
throughout a single regional grid over several days using actual episodic meteorological inputs,
this approach has been shown to provide nearly exact (to within 10‐3‐10‐4%) consistency
between the density and pollutant fields.
Vertical advection is solved using a specific implicit backward‐Euler integration scheme
designed specifically for CAMx (Emery et al, 2011). Since implicit schemes are absolutely stable,
only one solution step is necessary per driving time step. Explicit approaches require
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS
www.camx.com
74

March 2016

CAMx User’s Guide Version 6.3
4. Core Model Formulation

potentially many sub‐steps (on the order of 10‐100) to maintain a stable solution, which
introduces the potential for excessive numerical diffusion.
4.4.2 Sub‐Grid Turbulent Transport: Diffusion
CAMx employs a first‐order eddy viscosity (or “K‐theory”) approach by default to represent sub‐
grid turbulent diffusion (or mixing). As a “local” closure technique, K‐theory only treats mass
transfer cell‐by‐cell (horizontal) or layer‐by‐layer (vertical), analogously to the diffusion of heat
through a solid medium. Whereas K‐theory adequately characterizes horizontal diffusion and
weak vertical mixing during neutral and stable conditions, the shortcomings of K‐theory are
related to its inadequate treatment of deep vertical convective boundary layer mixing. In
convective situations, buoyant plumes deriving energy in the surface layer are quickly and
efficiently mixed deep into the atmosphere within eddies that are usually much larger than the
individual model layers. Therefore, K‐theory may mix the convective boundary layer much less
efficiently than commonly observed. This has been shown to have very important ramifications
for chemistry, especially during transition periods between stable/neutral and convective
conditions. CAMx includes the option to use K‐theory vertical mixing (default) or the non‐local
Asymmetric Convective Model (ACM2) from Pleim (2007).
4.4.2.1 Horizontal Diffusion
As discussed by Yamartino (2000) advection solvers such as Bott and PPM reduce numerical
diffusion to the point where modelers need to be concerned about including appropriate levels
of explicit horizontal diffusion. Currently, there is very little information on the appropriate
level of horizontal diffusion for Eulerian grid models. This issue is not limited to CAMx.
Explicit horizontal diffusion coefficients are determined within CAMx using a deformation
approach based on the methods of Smagorinsky (1963):

KX /Y



2
2
 u v  
xy  u v 
Ko 
   
  

4 2  y x 
 x y  

1/ 2

Separate diffusivity components are generated for fluxes in the x‐ and y‐directions since KX and
KY are calculated for separate cell faces in the Arakawa C grid arrangement. The value of K0 is
specified according to the approach in MM5 (Anthes and Warner, 1978):
K0

 3  10 3

 x y
t

A maximum value of KX/Y is set to maintain numerical stability for the given grid‐specific
timestep. A minimum value is set to 1 m2/s. Horizontal diffusion is applied using an explicit
simultaneous two‐dimensional flux‐divergence calculation.
4.4.2.2 K‐theory Vertical Diffusion
The default vertical diffusion solver (K‐theory) uses a standard implicit backward‐Euler
integration scheme. Gridded vertical diffusion coefficients (Kv) must be supplied to the model
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

75

www.camx.com

March 201
16

CAMx User’s Guide Versio
on 6.3
4. Core Model Formulation

for the master
m
grid via input file;; the user maay optionallyy supply verrtical diffusio
on coefficien
nts
for any or
o all nested grids. Wherreas the verttical advecti on step is so
olved in a sin
ngle step, K‐‐
theory ve
ertical diffussion is solved
d over poten
ntially severaal sub‐steps,, depending upon the
magnitud
de of the verrtical diffusivvity relative to the verticcal layer dep
pths (CAMx uses the smaller
of the current grid time step or a maximum 5 minute su b‐step). Thiis ensures no
on‐local diffusive
coupling among all laayers that exxhibit high diffusivity, esspecially for long drivingg time steps on
g
that wo
ould otherwise only expe
erience mixiing between
n adjacent laayers during a
coarser grids
single ste
ep.
4.4.2.3 ACM2
A
Vertical Diffusion
Pleim (20
007) has devveloped the ACM2 for th
he Weather Research an
nd Forecastin
ng (WRF)
meteorological mode
el and the Co
ommunity Multiscale
M
Aiir Quality (CM
MAQ) modeel. The ACM2
a
adjaccent layers using
u
K‐theory; (2) rapid upward non
n‐local mixin
ng
includes:: (1) mixing among
from the surface laye
er to all laye
ers through the
t boundarry layer (the convective aaspect); and
d (3)
slower co
ompensatingg downward
d mixing laye
er‐by‐layer frrom the top
p of the boun
ndary layer tto
the surfaace (the asym
mmetric aspect). Figure 4‐3 shows tthis approacch schematiccally. Duringg
non‐convvective cond
ditions, ACM2 reverts baack to the loccal K‐theoryy componentt. Thus, ACM
M2
includes the basic feaatures of bo
oth local and the most im
mportant com
mponent of non‐local
exchange
e.

e turbulent exchange am
mong layerss within a
Figure 4‐‐3. Schemattic representtation of the
vertical grid
g column during convvective adjustment in th
he ACM2 (taaken from Pleim [2007])).
The ACM
M2 paramete
erization is an alternative
e option to tthe default KK‐theory app
proach. All
variabless needed to calculate
c
the
e transfer raates are avai lable from the existing input files, o
or are
already calculated
c
within CAMx for
f other pu
urposes. Thee CAMx ACM
M2 option do
oes not requ
uire
that ACM
M2 be run wiithin the me
eteorologicall model usedd to derive in
nputs for CA
AMx. Howevver,
using ACM2 in both meteorologi
m
ical and chem
mistry modeels does lead
d to consisteent boundaryy
m
cal interfacee programs (e.g., WRFCA
AMx, MM5CA
AMx)
layer chaaracterization. Certain meteorologic
include an
a option to generate ve
ertical diffusiivities in thee same mann
ner as the ap
pproach used
within th
he CMAQ AC
CM2 algorithm, which pe
erforms its ddiffusivity callculations internally. Th
he
ACM2 difffusivity calcculation is a hybrid of tw
wo methodollogies: (1) bo
oundary layeer scaling baased
on Hostlaag and Boville (1993); an
nd (2) local scaling
s
accorrding to Rich
hardson num
mber and verrtical
shear sim
milar to Liu and Carroll (1
1996). The former
f
is useed within the stable bou
undary layerr or if
COMPREHEN
NSIVE AIR QUALITY
Y MODEL WITH EX
XTENSIONS

76

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
4. Core Model Formulation

it is larger than the latter. The CAMx implementation of ACM2 supports OMP and MPI parallel
processing.
Note that ACM2 increases CAMx runtime considerably. Whereas the original CAMx K‐theory
solver is implicit and does not need to use small time steps, ACM2 and its K‐theory counterpart
both use explicit solvers, which must divide the model time step into many very small sub‐steps
to generate a stable solution. The slow‐down can be exacerbated by using the much larger
ACM2 vertical diffusivity values instead of the standard diffusivity options available in the CAMx
meteorological preprocessors.
NOTE: ACM2 does not work with the Decoupled Direct Method (DDM) or the Integrated
Process Rate (IPR) component of the Process Analysis (PA) tool.

4.5 Wet Deposition
Wet deposition is the predominant removal process for fine particles. Particles act as cloud
condensation nuclei and resulting cloud droplets effieciently grow and accrete into
precipitation. Particles can also be directly scavenged by precipitation via impaction. The rates
of nucleation and impaction depend upon cloud type (e.g., prolonged widespred stratiform vs.
brief localized convection), precipitation rate, and particle and cloud water size distribution.
Wet deposition can also be an important removal process for relatively soluble gases through
the following series of steps:



Diffusion/absorption of gas molecules into cloud droplets;



Scavenging of cloud droplets by precipitation;



Diffusion of ambient gases into falling precipitation;



Possible aqueous‐phase reactions within cloud and rain water.

Each of the steps above may be reversible, so that the overall net removal of gases depends on
the results of forward and backward processes at each step. The rate at which these processes
occur depends on cloud type and the extent to which each pollutant dissolves in water and its
overall reaction rate once in solution. Cloud water droplets can absorb gases from the air up
the limit of their solubility in water. For many pollutants this solubility far exceeds the amount
of pollutant present in the air as determined by the Henry’s Law constant, which is defined as
the equilibrium ratio of pollutant concentrations in the liquid‐phase to the gas‐phase. High
values for the Henry’s law constant (>10,000 M/atm) indicate a strong tendency to dissolve into
water droplets, whereas low values (<100 M/atm) indicate a tendency to remain in the air
(Seinfeld and Pandis, 1998). Equilibrium between air and water concentration is usually
established on time scales of minutes, so equilibrium conditions can generally be assumed to
exist in the atmosphere.
The basic model implemented in CAMx is a scavenging approach in which the local rate of
concentration change c/t within or below a precipitating cloud depends on a scavenging
coefficient :
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

77

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
4. Core Model Formulation

c
t

 Λc

The scavenging coefficient is determined differently for gases and particles, based upon
relationships described by Seinfeld and Pandis (1998). Two components are calculated for
gases: (1) accretion of cloud droplets that contain dissolved gases, and (2) direct diffusive
uptake of ambient gases into falling precipitation. Two components are also determined for
particles: (1) accretion of cloud droplets that contain particle mass from the nucleation process,
and (2) impaction of ambient particles into falling precipitation with an efficiency that is
dependent upon particle and raindrop size. Each of these processes is described below.
The external environmental inputs to the CAMx wet deposition algorithm include the three‐
dimensional gridded distribution of cloud and precipitation water contents, with the
precipitation contents broken down into liquid, snow, and ice (“graupel”). Scavenging rate
equations were derived in terms of equivalent liquid precipitation rates, so the input
precipitation water contents are internally translated into this metric.
The following general assumptions are made in the CAMx scavenging model:
1) Rain drops, snowflakes, and graupel particles are each separately represented by a single
mean size, mass, and fall speed, which are determined from equivalent liquid
precipitation rate;
2) There is no mixed‐phased precipitation within a given grid cell – the dividing line between
liquid rainfall and the two frozen forms is 273 K;
3) Snow is only associated with stratiform precipitation, and graupel only with convective
precipitation;
4) Liquid cloud water is allowed to exist below 273 K – a linear ramp function is applied to
apportion total cloud water into liquid form between 233‐273 K (all cloud water is
assumed to be in ice crystal form below 233 K);
5) All gasses can directly diffuse into or from liquid rainfall (only strong acids can diffuse into
frozen precipitation) at a rate according to the precipitation’s state of saturation,
pollutant diffusivity, and aerodynamic considerations;
6) All gases can dissolve into liquid cloud water, which can be scavenged by all precipitation
forms – dissolved gasses are in equilibrium with ambient concentrations according to
Henry’s Law;
7) PM is irreversibly scavenged directly by all precipitation forms via impaction, and by
uptake into cloud water (liquid and ice) as condensation nuclei that is itself scavenged by
all precipitation forms;
8) All in‐cloud PM mass exists in cloud water (i.e., no “dry” aerosols exist in the interstitial air
between cloud droplets) – all PM species and sizes are hygroscopic and internally mixed.
4.5.1 Precipitation Parameters
The mean raindrop diameter dd (m) and fall speed vd (m/s) are taken from the empirical
estimates of Scott (1978). The drop diameter is related to rainfall rate P (mm/hr), and the fall
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

78

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
4. Core Model Formulation

speed relationship has been modified to better agree with data provided by Seinfeld and Pandis
(1998):
 9.0  10 4 P 0.21

dd

 3100 d d

vd

The precipitation water content Lp (g/m3) is related to rainfall rate by

Lp



P w
3.6  106 vd

where w is liquid water density (g/m3). This equation can be simplified to a dependency on
just rainfall rate when the above relations for drop velocity and size are substituted. The
resulting expression for rainfall rate as a function of precipitation water content is
1.27

 1  10 7 L p 

P

 
w



Locatelli and Hobbs (1974) developed power law equations relating average diameters, fall
speeds, and masses for 15 precipitating ice “habits”. We consolidated these relationships into
two forms, termed “snow” and “graupel”, by fitting new power law curves to the Locatelli and
Hobbs data. For snow,

vd 

0.83 d d 0.20

m d  0.035 d d 1.8
And for graupel,

vd 

1.1 d d 0.61

md  0.059 d d 2.6
Note that in these equations for ice, dd is in mm and crystal mass md is in mg. By assuming that
the number density and mass of snow/graupel crystals are equal to those for rain drops given
equal precipitation water contents, we can relate the magnitudes of ice size and fall speed to
equivalent liquid precipitation rate.
4.5.2 Gas Scavenging
Wet scavenging of gases by precipitation occurs within and below precipitating clouds. Below
the cloud, the total gas concentration in a given grid cell is available for scavenging. Within a
cloudy cell the total gas concentration must first be partitioned into an aqueous fraction caq
within cloud water and the remaining gaseous fraction cg within the interstitial air. Both
aqueous and interstitial gasses within a cloudy cell are available for scavenging, but are
removed at different rates as described below.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

79

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
4. Core Model Formulation

4.5.2.1 Scavenging of Gases in Cloud Water
The aqueous concentration in liquid cloud water (or cloud ice in the case of strong acids) is
determined by Henry’s Law thereby assuming the solution exists in steady‐state equilibrium.
The Henry’s Law constant for a given gas species k0 (M/atm) specifies the ratio of pollutant
mass in aqueous solution (M or mol/lh2o) to its equilibrium partial pressure (atm) at standard
pressure and temperature. This constant is adjusted for temperature T and converted to a
dimensionless molar ratio:

H

  1
1 
 k0 R T exp  A 
 
  298 T 

where R is the ideal gas constant (8.20610‐2 l atm/mol K), and A is the temperature
dependence factor. Dissociation of ammonia, nitric acid, and sulfur dioxide as a function of
cloud and rainwater pH is also considered in the solubility calculation. The effective Henry’s
Law constant H thus expresses the equilibrium ratio of the aqueous concentration caq (mass per
volume of water) to the gas concentration cg (mass per volume of air),


H

caq
cg

which are related to total concentration c by
c 

c g  caq

Lc

w

where Lc is cloud liquid water content (g/m3).
The fraction of gases present in cloud water can be removed by precipitation via accretion of
cloud water onto the falling hydrometeor (liquid or ice). As the hydrometeor falls, it sweeps a
cylindrical volume per unit time equal to
V




4

d d  d c 2 vd

where dc is the size of cloud droplets. This implies that the ambient motion of cloud droplets is
insignificant compared to the hydrometeor fall speed. Due to aerodynamic perturbations of air
flow around the falling hydrometeor, a collection efficiency is applied, i.e., the fraction of cloud
droplets within the collection volume that are scavenged by precipitation. For large
hydrometeors with sizes greater than 0.5 mm and droplets 10‐20 µm, we take this efficiency to
be 0.9 (Seinfeld and Pandis, 1998). Also, we further assert that (dd + dc)2 ~ dd2. Assuming then
that a mono‐disperse distribution of hydrometeors are falling through a mono‐disperse
distribution of cloud water droplets, the scavenging coefficient for precipitation collecting cloud
water is
 2
Λc 
d d vd E N d
4
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

80

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
4. Core Model Formulation

where E is the collection efficiency (0.9) and Nd is the hydrometeor number density. The
number density can be expressed in terms of equivalent rainfall rate, fall velocity, and drop size:

Nd

2.8  10 7 P
 d d3 vd / 6



Substituting the relationship between Nd and rainfall parameters, then
 4.2  10  7

Λc

EP
dd

The cloud water scavenging coefficient is scaled by the ratio of aqueous pollutant concentration
to total grid cell concentration to achieve the aqueous‐phase scavenging coefficient:

Λa



Λc

caq Lc
c w

4.5.2.2 Scavenging of Ambient Gases
Given the relatively short residence times of falling precipitation through a given grid cell,
aqueous equilibrium between ambient gas and precipitation cannot be assumed and so the
transfer of ambient gas into liquid rainfall (or ice in the case of strong acids) is explicitly
calculated. The maximum rate of transfer W of a gas to a falling hydrometeor containing no
pre‐existing pollutant mass is
W



K c H cg

The mass transfer coefficient Kc can be determined for a falling hydrometeor with speed vd and
diameter dd by
1/ 3
1/ 2
Dg 
v d d d     


 2  0 . 6
Kc 

dd 
    D g  


where Dg and  are the molecular diffusivity of the gas species and air, respectively.
Following the methodology of Seinfeld and Pandis (1998), the rate of caq increase can be
represented by a mass balance with the rate of transport to the hydrometeor:

dc
1
 d d3 aq   d d2 W
dt
6
The expression for W is substituted into the equation above, rearranged, and expressed in
terms of rainfall velocity. We then assume that through a given model layer the ambient gas
concentration and hydrometeor pH and size is constant. Multiplying by the number density of
falling hydrometeors Nd described above yields the gas‐phase concentration scavenged by all
drops falling through the layer:
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS
www.camx.com
81

March 2016

CAMx User’s Guide Version 6.3
4. Core Model Formulation

 1.67  10  6

Λg

P Kc
d d vd

Within cloudy layers, the gas‐phase and aqueous‐phase scavenging coefficients are added to
provide the total in‐cloud scavenging rate for gases,  = g+a. For below‐cloud scavenging,
 = g.
4.5.2.3 Solubility Limits on Gas Scavenging
There is a chance for rainfall to become super‐saturated for sparingly soluble gasses as it falls
through a grid column. The net scavenging coefficient for gases described above provides for
the maximum potential uptake rate into clean rain water, so careful consideration must be
given to appropriately determine the sign and magnitude of ambient grid cell concentration
change according to the degree of rainfall saturation. The change in gas concentration is
relaxed toward the difference between the maximum possible gas in solution for the given
conditions ceq, and the amount of pre‐existing gas in solution from layers above c0,

c  (ceq  c0 ) 1  exp  t 
Here ceq is determined from the total liquid water in the cell (rain plus cloud water) and from
Henry’s law equilibrium according to the total gas concentration in the cell (ambient grid
concentrations plus total gas in pre‐existing solution). If the concentration change is positive,
mass is added to the rain water (c0 is augmented) and removed from the grid cell; if negative,
mass is removed from rain water (c0 is decremented) and added to the grid cell.
4.5.3 Aerosol Scavenging
4.5.3.1 Scavenging of Aqueous Aerosols
All aerosols within cloudy layers are assumed to exist within cloud water. Therefore, the
scavenging coefficient for aqueous aerosols is exactly the same as for the scavenging of cloud
droplets:  = c.
4.5.3.2 Scavenging of Dry Particles
Wet scavenging of dry particles only occurs below precipitating clouds. We use the same
scavenging coefficient as derived for the collection of cloud droplets:
Λc

 4.2  10  7

EP
dd

For rain or graupel, the collection efficiency E is a function of particle size dp, and is given by
Seinfeld and Pandis (1998) as:

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

82

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
4. Core Model Formulation

E d p  



4
1  0.4 Re1 / 2 S c1/ 3  0.16 Re1 / 2 S c1 / 2
Re S c






4 
  1  2 Re1 / 2 
 w








St  S *

 
*
 St  S  2 / 3 

3/ 2

where µ and µw are the kinematic viscosity of air (1.8×10‐5 kg/m/s) and water (10‐3 kg/m/s),
respectively,  = dp/dd is the ratio of particle size to hydrometeor size, Re is the Reynolds
number for the hydrometeor, Sc is the Schmidt number for the collected particle, and St is the
Stokes number of the collected particle. The Reynolds number is given by

Re

Dd vd
2



while the Schmidt number is





Sc

Dp

where Dp is the particle Brownian diffusivity:



Dp

kT C
3 d p

Here, k is the Boltzman constant (1.38×10‐23 J/K) and C is the Cunningham correction factor for
small particles:
C

 1

 0.55 d p  
2 

1.257  0.4 exp 
  
dp 


and where  is the mean free path of air (6.5×10‐8 m). The Stokes number is given by

St



vd d p2  p C
9 d d

where p is the particle density. The S* parameter is given by

S*



COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

1.2  ln1  Re  / 12
1  ln1  Re 

83

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
4. Core Model Formulation

Note that particle size dp and density p are affected by aerosol water content, which is
determined from local humidity and PM deliquescence properties according to the ISORROPIA
model (see Section 5).
Snow presents a complication for the efficiency calculation, since it is not a single solid mass but
rather a collection of crystals arranged in a myriad of possible shapes that can lead to significant
aerodynamic drag. This drag affects particle collection efficiency. To simplify the calculation,
we use the value for E determined for rain and graupel, but set a lower limit for E of 110‐3
based on the work of Sauter and Wang (1989).

4.6 Dry Deposition
Dry deposition can be an important removal process for many compounds. Due to the difficulty
of making direct measurements of dry deposition and the need for a suitable model
parameterization, dry deposition is often treated as a first‐order removal mechanism, where
the flux of a pollutant to the surface is the product of a characteristic deposition velocity and its
concentration in the “surface layer” (i.e., the lowest model layer). Deposition velocities are
derived from models that account for the reactivity, solubility, and diffusivity of gases, the sizes
of particles, local meteorological conditions, and season‐dependent surface characteristics. The
factors affecting deposition are discussed in more detail below.
For a given species, particle size, and grid cell, CAMx determines a deposition velocity for each
landuse type present in that cell and then linearly combines them according to the fractional
distribution of landuse. The deposition flux is used as the lower boundary condition in the
vertical diffusion algorithm. Aerosol size spectra and species‐dependent properties needed for
the deposition velocity calculations are externally supplied to CAMx for all pollutant species via
the chemistry parameters file; gridded landuse is supplied to the master grid and optionally any
nested fine grids; the season is determined by the simulation date and location on the globe.
Movement of material along a path from the atmosphere, through any plant canopy, and onto
the various plant and ground surfaces within and below the canopy is typically modeled by
analogy to an electrical circuit. Resistances in serial and parallel arrangements are used to
represent the relative ease with which material moves through different portions of the
deposition pathway. Each branch of the circuit represents a different path by which material
may be deposited. For example, gaseous pollutants may transfer through the lowest layers of
the atmosphere partially into a plant canopy, through the stomatal openings on plant leaves
and into the plant mesophyll tissue. Alternatively, the material may travel all the way through
the plant canopy and deposit on the ground surface.
CAMx offers two dry deposition options: the original approach based on the work of Wesely
(1989) and Slinn and Slinn (1980); and an updated approach based on the work of Zhang et al.
(2001; 2003). Both of these options are briefly described below.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

84

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
4. Core Model Formulation

4.6.1 The Wesely/Slinn Model
4.6.1.1 Dry Deposition of Gases
Wesely (1989) developed a resistance model that incorporates the major elements described
above. Deposition velocity vd is calculated from three primary resistances r (s/m) in series:

1
ra  rb  rs



Vd

The aerodynamic resistance ra represents bulk transport through the lowest model layer by
turbulent diffusion, and operates equivalently for all gases and small particles. The magnitude
of this resistance depends on the intensity of turbulent motion, which in turn depends on solar
insolation, wind speed, surface roughness, and near‐surface temperature lapse rate. In CAMx it
is calculated from
ra

1
k u*



 1

ln     h 
  zo 


where u* is friction velocity (m/s), k is von Karman’s constant, z is the lowest model layer
midpoint height (m), z0 is the surface roughness length (m), and h is a stability correction term.
The surface layer parameterization of Louis (1979) is used to supply friction velocity and
stability correction as a function of input surface meteorology and roughness length.
Roughness length is internally assigned according to season and the input gridded distribution
of 11 landuse types as described in Section 3. In general, aerodynamic resistance is at a
minimum on warm, sunny days with strong mixing due to surface heating and mechanical
turbulence, and at a maximum on cool, calm nights when turbulent mixing is suppressed.
The quasi‐laminar sublayer (or boundary) resistance rb represents molecular diffusion through
the thin layer of air directly in contact with the particular surface to which material is being
deposited. It is usually assumed to depend only on the molecular diffusivity of each pollutant
species, and is given by

rb



2 S c2 / 3
k u*

where Sc is the Schmidt number, or the ratio of air viscosity to species molecular diffusivity.
Over land, surface resistance rs is expressed as several more serial and parallel resistances that
depend upon the physical and chemical properties of the surface in question:

rs



1
1
1
1
1



rst  rm ruc rdc  rcl rac  rgs

where the first set of parallel resistances represents the pathway into the stomatal (rst) and
mesophyll (rm) portions of active plants, the second is the pathway into the upper canopy (ruc),
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

85

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
4. Core Model Formulation

the third is the pathway into the lower canopy (rdc and rcl), and the fourth is the pathway to the
ground surface (rac and rgs). Many of these resistances are season‐ and landuse‐dependent, and
are built into Wesely’s model; some in turn are adjusted within CAMx for solar insolation and
surface wetness (vegetation is assumed to be unstressed). A few other resistances have been
developed by Wesely for SO2 and ozone, and so are scaled for each gaseous species based on
the following pollutant properties:



Molecular diffusivity (determined from molecular weight,




Henry’s law solubility (H);
Chemical reactivity toward oxidation of biological substances (f).

M g / M H 2 O );

This allows the resistance approach to be used to estimate deposition velocities for a wide
range of gaseous pollutants.
The surface resistances for strong acids (e.g., nitric, sulfuric, and hydrochloric acid, peroxides)
are set to zero given their strong rates of update by biota and other surfaces (Huebert and
Robert, 1985; Wesely and Hicks, 2000). The species for which surface resistance is set to zero
are defined in the CAMx chemistry parameters file.
Over water, the surface resistance for all gas species other than ozone is based on some
improvements adopted by Kumar et al. (1996) following Sehmel (1980):

rs



1
3.9  10

5

H Ts u*

where Ts is surface temperature (K). For ozone, this equation has been updated to
parametrically match the tendencies of measured ozone fluxes reported by Helmig et al. (2012)
from ship‐borne measurements:

rsO3 

1
110  5 106 H Ts3 u *
4

where Ts is in C rather than K. The cubic temperature dependence fits the deposition velocity
response to the range of sea surface temperatures reported in the Helmig et al. data. The
additional 1×10‐4 term sets an upper limit on rs and a lower limit on deposition velocity so that
the latter does not fall much below 0.01 cm/s. A lower limit of 1500 s/m is placed on rs such
that ozone deposition over water does not exceed 6.5 cm/s, which is the upper limit in the
measured data.
4.6.1.2 Dry Deposition of Aerosols
Surface deposition of particles occurs via diffusion, impaction, and/or gravitational settling.
Particle size is the dominant variable controlling these processes. The resistance approach of
Slinn and Slinn (1980), as implemented by Kumar et al. (1996), has been adopted in CAMx.
Particle deposition velocity for a given aerosol size is calculated using the following resistance
equation:
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS
www.camx.com
86

March 2016

CAMx User’s Guide Version 6.3
4. Core Model Formulation

vd



v sed 

1
ra  rb  ra rb v sed

where vsed is the gravitational settling (or sedimentation) velocity. This velocity is dependent on
aerosol size and density:

vsed

D 2 g C p



18

where D is the log‐mean particle diameter (m) of a given size section, ρp is particle density
(g/m3), g is gravitational acceleration, and  is the viscosity of air. The factor C is the
Cunningham correction for small particles, as described earlier for wet scavenging of particles.
Note that particle size and density are affected by aerosol water content, which is determined
from local humidity and PM deliquescence properties according to the ISORROPIA model (see
Section 5).
Aerodynamic resistance ra is identical to the value used for gaseous dry deposition. Resistance
to diffusion through the quasi‐laminar sub‐layer layer depends on aerosol Brownian diffusion
and inertial impaction. Particles are assumed to remain on a surface once they impact, so
resuspension effects are ignored. Boundary resistance rb is given by
rb





u* S

2 / 3
c

1
 10  3 / S t



The stokes number St is calculated from

St



vsed u*2
g

4.6.1.3 Specification of Season
The Wesely (1989) deposition algorithm specifies the various surface resistances by land cover
type for five seasons: Spring, Summer, Fall, Winter, and Winter with snow cover. CAMx
internally defines a season map to determine four of these five seasons by month and latitude
(Table 4‐2). Five latitude bands exist in each hemisphere:







Tropical
Sub‐tropical
Temperate
Cool
Polar

< 20
20 to 35
35 to 50
50 to 75
>75

The seasons in the Northern and Southern hemispheres are offset by six months. This offset
does not cause any discontinuity at the equator because all 12 months are defined as summer
in the tropical band at the equator. This season map is generalized and may not be ideal for all
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

87

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
4. Core Model Formulation

locations. The season map is coded into data statements in the “CAMx/chmdat.f”
subroutine and could be changed to better suit a specific region.
The season map shown in Table 4‐2 does not consider snow cover because it is quite variable in
space and time. Gridded snow cover data are specified in the time‐variant 2D surface input file
(see Section 3 and Section 4.7 below). Snow covered grid cells are assigned the Wesely (1989)
surface resistances for the category “winter with snow cover”, regardless of the season.

Table 4‐2. Relationships between season and month/latitude used in the CAMx Wesely/Slinn
dry deposition model. Exception: seasons for the area within 50N‐75N and 15W‐15E are
internally set to those of latitude band 35‐50 to account for regions of Europe in which the
climate is influenced by the Gulf Stream.
Month
Northern
Hemisphere
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec

Southern
Hemisphere
Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun

< 20
Tropical
Summer
Summer
Summer
Summer
Summer
Summer
Summer
Summer
Summer
Summer
Summer
Summer

Latitude Band
20 ‐ 35
35 ‐ 50
Sub‐tropical
Temperate
Winter
Winter
Spring
Winter
Spring
Spring
Spring
Spring
Summer
Spring
Summer
Summer
Summer
Summer
Summer
Summer
Summer
Fall
Fall
Fall
Fall
Fall
Fall
Winter

50 ‐ 75
Cool
Winter
Winter
Winter
Spring
Spring
Summer
Summer
Summer
Fall
Fall
Winter
Winter

> 75
Polar
Winter
Winter
Winter
Winter
Winter
Spring
Summer
Fall
Winter
Winter
Winter
Winter

4.6.2 The Zhang Model
Environment Canada’s AURAMS air quality model uses a state‐of‐the‐science deposition
scheme that possesses an updated representation of non‐stomatal deposition pathways (Zhang
et al. 2003; Zhang et al. 2008). The approach incorporates the “leaf area index” (LAI), which is
an important aspect of newer dry deposition schemes that allows for scaling of pollutant
uptake into biota of varying densities. LAI is defined as the ratio of the one‐sided green leaf
area to a unit area of the ground. It is measured by satellite instruments at fairly high spatial
resolution. The Zhang model has been tested extensively through its use in daily air quality
forecasting in Canada, and has been shown to reproduce observed fluxes of ozone and SO2 with
reasonable accuracy. In CAMx, the Zhang model has tended to generate lower ozone
deposition rates relative to the Wesely model, which leads to higher ozone predictions overall.
This effect is seasonally dependent and will vary with the definition of LAI. Ozone is less
sensitive to the source of LAI (whether Zhang defaults or satellite‐enhanced) than to the choice
of deposition model.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

88

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
4. Core Model Formulation

4.6.2.1 Dry Deposition of Gases
The gas resistance model of Zhang et al. (2003) invokes the same 3‐resistance equation for
deposition velocity as the Wesely (1989) model. The equations for aerodynamic resistance (ra)
and boundary resistance (rb) are very similar to the Wesely (1989) formulations. However,
Zhang et al. (2003) replace the surface resistance (rs) equation with a new relationship for the
overall canopy resistance:
rc



1
1  Wst
1
1


rst  rm rcut
rac  rgs

where Wst is the fraction of stomatal blocking under wet conditions, rcut is the cuticle resistance,
and all other resistances have similar meaning to the Wesely model. Stomatal resistance (rst) is
calculated using a sunlit/shade (so‐called “two‐big‐leaf”) stomatal resistance sub‐model.
Following Wesely (1989), values for rg and rcut are calculated for SO2 and O3 and then scaled for
other gaseous species. Cuticle resistance is slightly different from that defined in traditional
big‐leaf models in that it also considers the aerodynamic and quasi‐laminar resistances of
individual leaves. This is done by parameterizing rcut as a function of friction velocity, similar to
the concept of overall cuticle uptake considered in a multi‐layer model framework.
LAI is used in functions for rac, and rcut, where the LAI for any given day is linearly interpolated
from monthly default LAI as a function of landuse type. To account for LAI effects on surface
roughness (z0), a similar daily LAI interpolation is applied to that parameter. Hence, the Zhang
model does not require the specification of season, as all resistance equations are continuous
over each month (note that CAMx automatically applies the 6‐month offset for applications in
the southern hemisphere).
For snow on the ground and leaves, both rgs and rcut are adjusted by a snow cover fraction,
which is calculated from snow depth, snow age, and landuse type as described in Section 4.7.
Snow cover is defined through the input 2D surface file, as described in Section 3. For surfaces
without canopies, rgs is defined as the resistance to any surface (e.g. soil, ice, snow and water),
rac is set to zero, and very large values are used for rst, rm and rcut.
Over water, the updated temperature‐dependent ozone surface resistance equation described
above for the Wesely scheme is also used for the Zhang scheme.
The Zhang model includes a set of embedded annual surface roughness ranges and monthly LAI
specific to each of the 26 landuse categories. The capabilities of the scheme were extended by
adding the option to use episode‐specific (i.e., satellite‐derived) LAI data. Satellite‐based LAI
data from MODIS (MODerate‐resolution Imaging Spectroradiometer)2 can be processed into
gridded LAI fields that are passed to CAMx as an optional record in the time‐invariant 2D
surface input file (see Section 3). The optional gridded LAI fields are used to scale the default
landuse‐specific LAI values. For each grid cell, a landuse‐weighted default LAI is determined
2 MODIS provides LAI at 250 meter spatial resolution and 16 day temporal resolution.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

89

www.camx.com

March 201
16

CAMx User’s Guide Versio
on 6.3
4. Core Model Formulation

accordingg to the land
duse fraction
ns present. An LAI scalinng factor is tthen determ
mined as the ratio
of the inp
put LAI to th
he landuse‐w
weighted deffault LAI. Thhis factor is u
used to scalee the individual
default LAI
L values for each landu
use type in th
he grid cell, subject to th
he annual m
maximum to
minimum
m range amo
ong the monthly default values. Figuure 4‐4 illusttrates how tthe use of
satellite LAI data intrroduces additional episo
ode‐specific vvariation intto the LAI fieeld.

d in the Zhang dry depo
osition schem
me
Figure 4‐‐4. Comparison of montthly LAI dataa embedded
against episode‐spec
e
cific LAI for June
J
2005.
4.6.2.2 Dry
D Depositio
on of Aeroso
ols
Theoreticcally, particles in the size
e range 0.1‐1.0 m m ddeposit at raates much leess than 0.01
1
cm/s, but such value
es are compaarable only to
t laboratoryy (wind tunn
nel) studies. According tto
Zhang et al. (2001), much
m
higherr values have
e been obtaiined in manyy field studiees, includingg for
sub‐micrron sulfate in
n which depo
osition veloccities of one to two ordeers of magniitude higherr
have bee
en measured
d. For examp
ple, Gallaghe
er et al. (19997) state thaat much high
her depositio
on
velocity values,
v
typiccally 1 cm/s or
o more for sub‐micron aerosol dep
position to a forest, are
consisten
nt across the
e aerosol size
e spectrum. The Zhang et al. (2001) study deveeloped a simple
paramete
erization of particle dry deposition as
a a functionn of aerosol size and landuse that
predicts higher depo
osition velociities for sub‐‐micron aeroosols, especially over rough vegetatted
surfaces..
COMPREHEN
NSIVE AIR QUALITY
Y MODEL WITH EX
XTENSIONS

90

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
4. Core Model Formulation

The deposition of aerosols is based on the model of Slinn (1982), but using simplified empirical
parameterizations to replace detailed canopy information. The aerosol deposition velocity is
defined as:

vd  vsed 

1
ra  rb

where vsed and ra have identical meanings as the Slinn and Slinn (1980) definition described
earlier for the Wesely/Slinn deposition model. Note that in this case, the virtual serial
resistance rarbvsed has been removed, which results in higher values of deposition velocity. The
quasi‐laminar boundary resistance is given by

1
 0u* EB  EIM  EIN R1

rb 

where the variable E includes the collection efficiencies for Brownian diffusion, impaction, and
interception, respectively, R1 is a factor representing the fraction of particles that stick to the
surface, and 0 is an empirical constant that is set to a value of 3. The Brownian collection
efficiency depends on the Schmidt number, while the impaction efficiency and R1 depend on
the Stokes number.
The collection efficiency by interception also exists if the particle passes an obstacle at a
distance shorter than its physical dimensions (e.g., large particles passing near hairy leaves).
Zhang et al. (2001) adopted a simple equation for this term that is a function of particle
diameter and a characteristic radius, for which default values are given for different landuse
and seasonal categories.
Figure 4‐5 compares estimated particle deposition velocities from the Zhang model, the Slinn
and Slinn (1980) model, and the AERMOD model (EPA, 1998). Calculations were made for
daytime, neutrally stable conditions for a range of wind speeds and landuse categories. Figure
4‐5 shows that the Zhang model increases deposition velocities over forest by roughly an order
of magnitude for the 0.1‐1 m range, yet reduces deposition velocities above 1 m.

4.7 Snow Cover and Surface Albedo
Surface albedo for snow‐covered grid cells is calculated according to snow cover, snow age, and
land cover type. The approach is based on literature describing the evolution of snow albedo in
the WRF/NOAH land surface model (LSM) over the past decade (Ek et al., 2003; Wang and Zeng,
2010; Livneh et al., 2010; and Barlage et al., 2010). Fractional snow cover (fs) is accounts for the
effects of surface roughness elements (shrubs, trees, rocks and other structures) extending
above thin/patchy snow:
1

exp

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

exp

91

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
4. Core Model Formulation

Forest
100

wind = 2
wind = 5
wind = 10

V d (cm s-1)

10

wind = 15
wind = 2
wind = 5

1

wind = 10
wind = 15
wind = 2

0.1

wind = 5
wind = 10

0.01
0.001

wind = 15
0.01

0.1

1

10

100

Particle diameter ( m)

Figure 4‐5. Comparison of particle dry deposition velocities as a function of size and wind
speeds (m/s) for three models: black – Zhang et al. (2001); blue – Slinn and Slinn (1980);
orange – AERMOD (EPA, 1998). Results are shown for a forest landuse category during
daytime neutral stability. Particle density was set at 1.5 g/cm3.
where  = 2.6, W is snow water equivalent depth (SWE), and Wc is the threshold SWE above
which fs = 100%. Following Wang and Zeng (2010) and Linveh et al. (2010), Wc is set to 0.01 m
for barren or low vegetation (grasslands) and to 0.2 m for tall vegetation (forest), except an
intermediate value of 0.02 m is assigned to range, mixed agriculture/range, and shrub lands
where vegetation is typically higher than grasses (Table 4‐4). Throughout CAMx we apply a
common approximation that actual snow depth is 10SWE.
Snow albedo (as) is allowed to evolve to account for the optical effects of melting and
accumulation of dirt/soot, following the approach of Linveh et al. (2010):

where amax is the maximum fresh snow albedo (0.85; Barlage et al., 2010), t is the number of
days since the last snowfall, A = 0.94 (0.82) and B = 0.58 (0.46) during the accumulation
(ablation) phase. Accumulation occurs during cold periods when surface temperature is below
273 K, whereas ablation occurs during melting periods when surface temperature is at 273K.
Snow albedo is constrained to a lower bound of 0.4. Snow age is refreshed to zero (and snow
albedo to 0.85) when SWE accumulates by more than 0.001 m/hr (accumulating snow depth > 1
cm/hr).
The resultant grid‐cell average surface albedo (a) is a linear combination of snow albedo (as)
and terrestrial (non‐snow) albedo (at):
1

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

92

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
4. Core Model Formulation

where at is defined according to landuse type (Tables 3‐4, 3‐5). In case a distribution of
multiple landuse types exists within a given grid cell, a linear weighting scheme is employed to
account for variable snow cover fractions for each individual landuse type:
1
where the sum is over all landuse types, fn is the fractional coverage of landuse n, at(n) is the
default terrestrial albedo for landuse n, fs(n) is the fractional snow cover for landuse n, and as(n)
is the calculated snow albedo for landuse n. Figure 4‐6 shows an example of grid‐cell albedo
evolution for a hypothetical 20‐day springtime snow event (assuming ablation conditions) for
low and tall vegetation grid cells with a terrestrial (non‐snow) albedo of 0.05. Several snow
accumulation events are specified to occur over the first 12 days, followed by rapid melting to
zero depth by day 20. While total albedo increases to peak values of 0.85 quite rapidly for low‐
vegetation, the value for tall vegetation lags and peaks just above 0.5 at maximum snow depth.
Both cases indicate effects from snow depth (fractional snow cover) and snow age.

1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0

Low Vegetation
Tall Vegetation
Depth

1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0

Depth (m)

Albedo

Snow Depth and Albedo with Age

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

Days of Example Event

Figure 4‐6. Example of grid‐cell albedo evolution for a hypothetical 20‐day springtime snow
event (assuming ablation conditions) for low and tall vegetation grid cells with a terrestrial
(non‐snow) albedo of 0.05.

4.8 Surface Model for Chemistry and Re‐Emission
The CAMx surface model is an optional capability that treats: (1) chemical degradation and/or
transformation of deposited pollutant mass on soil, vegetation and an overlying snowpack; (2)
volatilization of chemical products back into the air (re‐emission); and (3) loss from leaching
into soil, penetration into plant tissue, and uptake into snow melt water. The surface model
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

93

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
4. Core Model Formulation

treats any subset of species listed in the core model’s chemical mechanism. Limitations of the
current implementation include:



The surface model cannot be used with the Plume‐in‐Grid treatment;



Deposition to water surfaces is assumed to be irreversible and thus is not tracked by the
surface model;



Wet deposition does not contribute to surface mass, as compounds in aqueous solution
are assumed to be immediately lost to surface water processes (absorption, runoff, etc.).

4.8.1 Surface Model Algorithms
Figure 4‐7 displays the surface model processes schematically and Table 4‐3 defines parameters
that are referred to in Figure 4‐7. While core model algorithms are used to deposit compounds
to the surface and re‐emit them to the atmosphere, the surface model tracks the accumulation
of mass on terrestrial surface media (soil, vegetation and snow), subsequent chemical
transformation (both heterogeneous and photolysis), re‐emission to the atmosphere, and
physical removal.

Figure 4‐7. Schematic of the CAMx surface model.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

94

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
4. Core Model Formulation

Table 4‐3. Description of CAMx surface model variables shown in Figure 4‐7.
Variable

Definition

Units

As

Areic mass of compound on soil or snow

mol ha‐1

Ap

Areic mass of compound on vegetation

mol ha‐1

Ssoil

Soil‐air partitioning coefficient

unitless

Ssnow

Snow‐air partitioning coefficient

unitless

Sveg

Vegetation‐air partitioning coefficient

unitless

kleach

Soil leaching rate coefficient

min‐1

Kmelt

Snow melt loss rate coefficient

min‐1

kpen

Leaf penetration rate coefficient

min‐1

J

Photolysis chemistry rate coefficient

min‐1

K

Heterogeneous chemistry rate coefficient

min‐1

Rleach

Leaching or snow melt loss rate

mol ha‐1 min‐1

Rpen

Leaf penetration rate

mol ha‐1 min‐1

Rchem

Chemistry rate

mol ha‐1 min‐1

After deposition to a snow‐free surface grid cell is calculated each time step, the newly
deposited mass increments are divided among soil and vegetation compartments and added to
total surface mass in each compartment accumulated during the run. The net soil/vegetation
split for a given grid cell is determined by the combination of the fractional coverage of each
landuse type in that cell and landuse‐specific split factors. The fractional coverage of 11
(Wesely) or 26 (Zhang) landuse categories in each grid cell is an external input to CAMx (Section
3). The soil/vegetation splits assigned to each landuse category are internally defined within
CAMx and assumed to be seasonally constant. Values for soil/vegetation splits are estimated
based on simple conceptual considerations of the amount of annual‐averaged vegetation (i.e.,
leaf area index) typical of each category (Table 4‐4).
Snow is activated in the surface model when snow depth is sufficiently deep to cover exposed
soil. The lower limit on snow depth is 10 cm to be consistent with the approach described in
Section 4.7 in which a 10 cm depth completely covers low‐vegetation landuse. In such cases,
the soil/vegetation split is replaced by the snow cover fraction such that the soil fraction is
entirely snow‐covered and the vegetation fraction is progressively covered with deeper snow
depth. The soil compartment transitions to a snow compartment; sorption coefficients and
rates for chemistry and loss covert to the values set for snow (as described below). With very
deep snow exceeding 200 cm, high vegetation is completely covered and the surface model
reduces to a single compartment for snow.
The surface model uses partitioning (equilibrium) coefficients to calculate the amount of
accumulated material sorbed to soil/snow and vegetation. The sorbed fraction is subject to
chemical reactions and physical removal associated with soil leaching, plant penetration, and
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

95

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
4. Core Model Formulation

Table 4‐4(a). Wesely landuse categories and associated annual‐averaged soil/vegetation split
factors, UV albedo, and SWE Wc.
Category
Number

Land Cover Category

Soil
Fraction

Surface Parameters
UV
Snow Wc
Albedo
(m SWE)

1

Urban

0.7

0.08

0.2

2

Agricultural

0.2

0.05

0.01

3

Rangeland

0.5

0.05

0.02

4

Deciduous forest

0.1

0.05

0.2

5

Coniferous forest, wetland

0.1

0.05

0.2

6

Mixed forest

0.1

0.05

0.2

7

Water

n/a

0.04

n/a

8

Barren land

1.0

0.08

0.01

9

Non‐forested wetlands

0.2

0.05

0.01

10

Mixed agricultural/range

0.3

0.05

0.02

11

Rocky (with low shrubs)

0.5

0.05

0.01

Table 4‐4(b). Zhang landuse categories and associated annual‐averaged soil/vegetation split
factors, UV albedo, and SWE Wc.
Category
Number

Land Cover Category

Soil
Fraction

Surface Parameters
UV
Snow Wc
Albedo
(m SWE)

1

Water

n/a

0.04

n/a

2

Ice

n/a

0.5

0.01

3

Inland lake

n/a

0.04

n/a

4

Evergreen needleleaf trees

0.1

0.05

0.2

5

Evergreen broadleaf trees

0.1

0.05

0.2

6

Deciduous needleleaf trees

0.1

0.05

0.2

7

Deciduous broadleaf trees

0.1

0.05

0.2

8

Tropical broadleaf trees

0.1

0.05

0.2

9

Drought deciduous trees

0.1

0.05

0.2

10

Evergreen broadleaf shrubs

0.5

0.05

0.03

11

Deciduous shrubs

0.5

0.05

0.02

12

Thorn shrubs

0.5

0.05

0.03

13

Short grass and forbs

0.5

0.05

0.01

14

Long grass

0.3

0.05

0.02

15

Crops

0.2

0.05

0.01

16

Rice

0.2

0.05

0.01

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

96

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
4. Core Model Formulation

17

Sugar

0.2

0.05

0.01

18

Maize

0.2

0.05

0.01

19

Cotton

0.2

0.05

0.01

20

Irrigated crops

0.2

0.05

0.01

21

Urban

0.7

0.08

0.2

22

Tundra

0.2

0.05

0.01

23

Swamp

0.2

0.05

0.01

24

Desert

1.0

0.08

0.01

25

Mixed wood forest

0.1

0.05

0.2

26

Transitional forest

0.1

0.05

0.2

snow melt. The un‐sorbed fraction is available for re‐emission. Separate chemical‐specific soil‐
air, vegetation‐air, and snow‐air partitioning coefficients are set in the CAMx chemistry
parameters file. They represent the equilibrium ratio of chemical on a surface to chemical in air
at the air‐surface interface. For example, a compound with a partitioning coefficient of 10,000
(unitless) has an equilibrium concentration on the surface that is 10,000 times more than that
in air.
Chemistry can simply decay deposited material as a removal process, or it can generate
products that can subsequently re‐emit depending on the products’ partitioning coefficient. All
surface removal processes are assumed to be irreversible and result in a permanent removal of
mass. Chemistry, soil leaching, plant penetration, and snow melt loss are dependent on
chemical properties of the compounds and also on numerous site‐specific factors such as soil,
vegetation, and snow properties, highly transient meteorological conditions, etc. Often these
factors are unknown or fall within a range. The rates of these processes are defined as the
process rate coefficient (k) times the mass on the surface area, or areic mass (A):
Rprocess = kprocess  Asurface
When the actual rate coefficients (or inversely, the half‐lives, t½) are unknown for the
substance, they can be generalized by 5 classes:
1.
2.
3.
4.
5.

Very fast:
Fast:
Moderate:
Slow:
Very slow:

t½ = 0.04 d
t½ = 0.21 d
t½ = 1.0 d
t½ = 5.0 d
t½ = 25 d

k = 17 d‐1
k = 3.3 d‐1
k = 0.69 d‐1
k = 0.14 d‐1
k = 0.03 d‐1

= 1.2×10‐2 min‐1
= 2.3×10‐3 min‐1
= 4.8×10‐4 min‐1
= 9.7×10‐5 min‐1
= 2.1×10‐5 min‐1

A 6th class can be added by setting the k‐value to zero or a de minimis value to effectively
remove the process from consideration. In this manner chemicals can be modeled with an
estimated half‐life that is unique for each process.
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

97

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
4. Core Model Formulation

Note that all partitioning coefficients and rates other than photolysis are fixed and ignore
dependence on various environmental conditions (e.g., temperature, pressure, surface type,
surface moisture, etc.). Photolysis rates are specified by the user to represent peak direct‐
exposure clear‐sky values at zero zenith (solar noon) and are internally adjusted for solar angle,
cloud attenuation (as calculated for atmospheric photolysis), and shade effects using
multiplicative factors. A multiplicative “shade factor” is defined to reduce photolysis rates
within/below vegetation. Snow cover reduces shading effects to account for enhanced
reflection and internal UV scattering within the snowpack.
Losses by soil leaching, plant penetration, and snow melt are arbitrarily accelerated during rain
events, such that a 1 mm/hr rainfall rate results in an e‐folding loss of surface mass in 1 hour.
Mass loss within the snowpack by melting alone occurs only when surface temperature is at
273 K. Snowpack loss also occurs during snowfall such that a 1 cm/hr accumulation results in
an e‐folding loss of surface mass in 24 hours by successively burying pollutant mass and limiting
its ability to diffuse through the snowpack. The model assumes that no surface mass is re‐
introduced as snow depth/fraction decrease during sublimation or melting (i.e., irreversible loss
of surface mass as implemented for soil and vegetation).
The approach for re‐emission of volatilized (un‐sorbed) mass is consistent with the CAMx dry
deposition algorithm. Since water surfaces are not considered by the surface model, re‐
emission fluxes from water are ignored in this implementation. Dry deposition of material from
the lowest model layer to the surface is treated as an irreversible (fully sorbed) first‐order flux
through the use of a dry deposition velocity. Re‐emission of volatilized (un‐sorbed) mass is also
treated as a first‐order 1‐way flux using an “effective” velocity that is similar in form to
deposition:

ve 

1
ra  rb

where ra is the aerodynamic resistance to turbulent transfer through the lowest model layer,
and rb is the resistance to molecular diffusion through the laminar sub‐layer in contact with
surface elements. The deposition surface resistance term rs is missing since only the pre‐
determined un‐sorbed fraction of surface mass is considered for surface‐to‐air transfer. The ra
and rb terms are calculated by the surface model in exactly the same manner as the values used
for dry deposition to ensure consistency.
4.8.2 Running CAMx With the Surface Model
The CAMx surface model parameters that need to be specified for each compound or surface
reaction to be tracked are as follows:

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

98

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
4. Core Model Formulation

Ssoil

Soil‐air partitioning coefficient

unitless

Sveg

Vegetation‐air partitioning coefficient

unitless

Ssnow

Snow‐air partitioning coefficient

unitless

kleach

Soil leaching rate coefficient

min‐1

kpen

Leaf penetration rate coefficient

min‐1

kmelt

Snow melt loss rate coefficient

min‐1

Jsoil

Soil photolysis rate coefficient

min‐1

Ksoil

Soil heterogeneous chemistry rate coefficient

min‐1

Jveg

Vegetation photolysis rate coefficient

min‐1

Kveg

Vegetation heterogeneous chemistry rate coefficient

min‐1

Jsnow

Snow photolysis rate coefficient

min‐1

Ksnow

Snow heterogeneous chemistry rate coefficient

min‐1

These values are set at the end of the CAMx chemistry parameters file; an example of the
chemistry parameters file format is shown in Figure 4‐8. A control record is also needed at the
top of the chemistry parameters file to define the number of species and reactions to track.
A CAMx namelist control file variable called “SURFACE_MODEL” must be set to “true” in order
to invoke the surface model. When the surface model is invoked, the surface model section of
the chemistry parameters file is read and the respective equilibrium and rate variables are set
accordingly.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

99

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
4. Core Model Formulation

CAMx Version
|VERSION6.3
Mechanism ID
|2
Aerosol Option
|NONE
Description
|CB6r2 (r98/30/13 version) + ECH4
No of gas species |75
No of aero species |0
No of reactions
|216
Prim photo rxns
|23 1 8 9 21 27 28 38 43 47 50 56 88 92 97 98 108 112 114 117
119 128 129 161
No of sec photo rxn|6
ID, prim ID, scale |64 56 1.0
|90 88 1.0
|163 1 0.07
|196 1 0.015
|197 1 0.08
|201 1 0.08
SrfMod #spc, #rxns |3 2
.
.
.
Surface Model
Species
SoilSorb SoilLeach
VegSorb
VegPen
SnoSorb
SnoMlt
1 HNO3
1.00E+10 1.00E-10 1.00E+10 1.00E-10 1.00E+10 9.70E-05
2 PNA
1.00E+10 1.00E-10 1.00E+10 1.00E-10 1.00E+10 9.70E-05
3 HONO
1.00E+00 1.00E-10 1.00E+00 1.00E-10 1.00E+00 9.70E-05
Rxn Precursor Product
Soil K
Soil J
Veg K
Veg J
Snow K
Snow J
1 HNO3
HONO
0.00E+00 0.00E+00 0.00E+00 0.00E+00 0.00E+00 1.00E-03
2 PNA
HONO
0.00E+00 0.00E+00 0.00E+00 0.00E+00 1.00E-01 0.00E+00

Figure 4‐8. The portions of the CAMx chemistry parameters file (highlighted) to support the
surface model. In this example, 3 gases are treated, where nitric acid (HNO3) and
peroxynitric acid (PNA) react to form nitrous acid (HONO). All three are subject to decay by
soil leaching, plant penetration, and snow melt loss. The values shown here are for
illustrative purposes only and do not represent any known surface chemistry mechanism.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

100

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
5. Chemistry Mechanisms

5. CHEMISTRY MECHANISMS
The photochemical mechanisms currently supported in CAMx are listed in Table 5‐1. All are
balanced for nitrogen conservation so that predicted NOy can be calculated as the sum of
nitrogen containing species. Mechanisms 2 through 6 can be linked to optional modal (CF) and
size‐segregated (CMU) primary and secondary particulate matter (PM) treatments. CAMx
includes algorithms for inorganic aqueous chemistry (RADM‐AQ), inorganic gas‐aerosol
partitioning (ISORROPIA), and organic gas‐aerosol partitioning and oxidation (VBS or SOAP).
The PM treatments require additional gas species as PM precursors and use products from the
gas‐phase photochemistry for the production of sulfate, nitrate, and condensable organic
gases. The CF PM treatment also supports several optional mercury species. Additionally,
there is an interface that allows a simpler user‐defined chemical mechanism to be employed in
the model (Mechanism 10). A listing of all reactions and rate expressions for supported
photochemical mechanisms are provided in the appendices.

Table 5‐1. Gas‐phase chemical mechanisms currently implemented in CAMx v6.3.
Mechanism ID
6
2
3
4

5

10

Description
CB05 (Yarwood et al., 2005b). 156 reactions among 51 species (38 state gases, 13 radicals).
CB6 “Revision 2” (CB6r2; Yarwood et al., 2010; Yarwood et al., 2012a; Hildebrandt Ruiz and
Yarwood, 2013). 216 reactions among 75 species (55 state gases, 20 radicals).
CB6r2 with updates to include reactions involving oceanic halogen compounds (CB6r2h;
Yarwood et al., 2014). 304 reactions among 115 species (88 state gases, 27 radicals).
CB6 “Revision 3” (CB6r3) that includes updates to improve NO2‐organic nitrate branching
under winter conditions (Emery et al., 2015). 220 reactions among 77 species (55 state
gases, 22 radicals).
A version of SAPRC07 that includes updates to support toxics and numerical expressions of
rate constants to support the current chemistry mechanism compiler (SAPRC07TC; Carter,
2010; Hutzell et al., 2012). 565 reactions among 117 species (72 state gases, 45 radicals).
A user‐defined simple chemistry mechanism can be developed for any gas and/or
particulate species, which is defined by a “Mechanism 10” parameters file and solved within
a user‐supplied subroutine called “chem10.f.”

The selection of which mechanism to employ in a given CAMx application is determined by the
“chemistry parameter” input file. This file defines the mechanism number, the number of gas
and aerosol species, and the number of reactions for the mechanism, lists the species by name
with associated physical‐chemical properties, lists the reaction rate constants and temperature
dependencies for each reaction, and defines which reactions are photolytic. Chemistry
parameter input files for the available mechanisms are provided with CAMx and should not be
modified by users. See Section 3 for additional information on the format and usage of these
files. Chemistry parameters files are specific to versions of CAMx. Always use chemistry
parameters files with the right CAMx version number, do not attempt to use files for another
CAMx version.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

101

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
5. Chemistry Mechanisms

5.1 Gas‐Phase Chemistry
5.1.1 Carbon Bond
The Carbon Bond IV (CB4) mechanism was first developed by Gery et al. (1989), and was
subsequently updated in the 1990’s to include revised PAN chemistry, additional radical‐radical
termination reactions and updated isoprene chemistry (Carter 1996; Whitten et al., 1996).
Additional CB4 updates were then made to expand ozone modeling from urban to
regional/rural environments and to support secondary aerosol chemistry, mercury and toxics
(Yarwood et al., 2005a).
Several newer Carbon Bond versions are available in CAMx as described below. Table 5‐2 lists
chemical species names and properties common to all CAMx Carbon Bond mechanisms.
5.1.1.1 Carbon Bond 2005
Mechanism 6 is the 2005 version of Carbon Bond (CB05) developed for EPA atmospheric
modeling studies (Yarwood et al., 2005b). Updates in CB05 include:












Updated rate constants based on 2003 – 2005 IUPAC and NASA evaluations.
An extended inorganic reaction set for urban to remote tropospheric conditions.
NOx recycling reactions to represent the fate of NOx over multiple days.
Explicit organic chemistry for methane and ethane.
Explicit methylperoxy radical, methyl hydroperoxide and formic acid.
Lumped higher organic peroxides, organic acids and peracids.
Internal olefin (R‐HC=CH‐R) species called IOLE.
Higher aldehyde species ALDX making ALD2 explicitly acetaldehyde.
Higher peroxyacyl nitrate species from ALDX called PANX.
Lumped terpene species called TERP.

CB05 was evaluated against smog chamber data from the Universities of North Carolina and
California at Riverside. The new higher aldehyde and internal olefin species improve
mechanism performance for these species and produce oxidants more rapidly at low VOC/NOx
ratios. The new terpene species improves simulation of oxidants and PM from biogenic
emissions. Several new organic peroxide species improve the simulation of oxidants that are
involved in PM sulfate formation. The addition of explicit methylperoxy radical improves the
simulation of hydrogen peroxide under low NOx conditions.
5.1.1.2 Carbon Bond Version 6
Carbon Bond version 6 (CB6) was developed by Yarwood et al. (2010). Since then, CB6 has
undergone 2 major updates, as described below. Mechanism 2 is CB6 revision 2 (CB6r2;
Hildebrandt Ruiz and Yarwood, 2013).
Several organic compounds that are long‐lived and relatively abundant, namely propane,
acetone, benzene and ethyne (acetylene), were added explicitly in CB6 to improve oxidant
formation from these compounds as they are oxidized slowly at the regional scale. Alpha‐
dicarbonyl compounds (glyoxal and analogs), which can from secondary organic aerosol (SOA)
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

102

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
5. Chemistry Mechanisms

Table 5‐2. Species names and descriptions common to all Carbon Bond Mechanisms in CAMx.
Species

Description

Carbon #

C

H

O

Mol. Wt.

BZO2

Peroxy radical from OH addition to benzene

6

6

7

5

159.1

C2O3

Acetylperoxy radical

2

2

3

3

75.0

CRO

Alkoxy radical from cresol

7

7

7

1

107.1

CXO3

C3 and higher acylperoxy radicals

3

3

5

3

89.0

EPX2

Peroxy radical from EPOX reaction with OH

5

5

9

5

149.1

HCO3

Adduct from HO2 plus formaldehyde

1

1

3

3

63.0

HO2

Hydroperoxy radical

1

1

1

28.0

ISO2

Peroxy radical from OH addition to isoprene

5

5

9

3

117.1

MEO2

Methylperoxy radical

1

1

3

2

47.0

NO3

Nitrate radical

O

3

62.0

3

1

16.0

1

1

16.0

Oxygen atom in the O (P) electronic state

O1D

Oxygen atom in the O (D) electronic state

OH

Hydroxyl radical

OPO3

Peroxyacyl radical from OPEN

RO2

Operator to approximate total peroxy radical concentration

ROR

Secondary alkoxy radical

TO2

1

1

17.0

4

3

4

115.0

4

7

2

87.1

1

4

7

1

71.1

Peroxy radical from OH addition to TOL

7

7

9

5

173.1

XLO2

Peroxy radical from OH addition to XYL

8

8

11

5

187.1

XO2

NO to NO2 conversion from alkylperoxy (RO2) radical

4

7

2

87.1

XO2H

NO to NO2 conversion (XO2) accompanied by HO2 production

4

7

2

87.1

XO2N

NO to organic nitrate conversion from alkylperoxy (RO2) radical

4

7

2

87.1

AACD

Acetic acid

2

2

4

2

60.0

ACET

Acetone

3

3

6

1

58.1

ALD2

Acetaldehyde

2

2

4

1

44.0

ALDX

Propionaldehyde and higher aldehydes

3

3

6

1

58.1

BENZ

Benzene

6

6

6

CAT1

Methyl‐catechols

7

7

8

CO

Carbon monoxide

1

1

CH4

Methane

1

1

4

CRES

Cresols

7

7

CRON

Nitro‐cresols

7

EPOX

Epoxide formed from ISPX reaction with OH

ETH

4

78.1
2

124.1

1

28.0

8

1

108.1

7

7

3

153.1

5

5

10

3

118.1

Ethene

2

2

4

28.0

ETHA

Ethane

2

2

6

30.1

ETHY

Ethyne

2

2

2

26.0

ETOH

Ethanol

2

2

6

1

46.1

FACD

Formic acid

1

1

2

2

46.0

FORM

Formaldehyde

1

1

2

1

30.0

GLY

Glyoxal

2

2

2

2

58.0

GLYD

Glycolaldehyde

2

2

4

2

60.0

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

103

16.0

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
5. Chemistry Mechanisms

Species

Description

H2O2

H

O

Mol. Wt.

Hydrogen peroxide

2

2

34.0

HNO3

Nitric acid

1

3

63.0

HONO

Nitrous acid

1

2

47.0

HPLD

hydroperoxyaldehyde

5

5

8

3

116.1

INTR

Organic nitrates from ISO2 reaction with NO

5

5

9

4

147.1

IOLE

Internal olefin carbon bond (R‐C=C‐R)

4

4

8

56.1

ISOP

5

5

8

68.1

ISPD

Isoprene
Isoprene product (lumped methacrolein, methyl vinyl ketone,
etc.)

4

4

6

1

70.1

ISPX

Hydroperoxides from ISO2 reaction with HO2

5

5

10

3

118.1

KET

Ketone carbon bond (C=O)

4

4

8

1

72.1

MEOH

Methanol

1

1

4

1

32.0

MEPX

Methylhydroperoxide

1

1

4

2

48.0

MGLY

Methylglyoxal

3

3

4

2

72.0

N2O5

Dinitrogen pentoxide

5

108.0

NO

Nitric oxide

1

30.0

NO2

Nitrogen dioxide

2

46.0

NTR

Organic nitrates

3

119.1

O3

Ozone

3

48.0

OLE

Terminal olefin carbon bond (R‐C=C)

3

3

6

OPAN

Peroxyacyl nitrate (PAN compound) from OPO3

4

4

3

6

161.0

OPEN

Aromatic ring opening product (unsaturated dicarbonyl)

4

4

4

2

84.0

PACD

Peroxyacetic and higher peroxycarboxylic acids

2

2

4

3

76.0

PAN

Peroxyacetyl Nitrate

2

2

3

5

121.0

PANX

C3 and higher peroxyacyl nitrate

3

3

5

5

135.0

PAR

Paraffin carbon bond (C‐C)

1

5

12

PNA

Peroxynitric acid

PRPA

Propane

ROOH

Higher organic peroxide

SO2

Sulfur dioxide

SULF

Sulfuric acid (gaseous)

TERP

Monoterpenes

10

10

16

136.2

TOL

Toluene and other monoalkyl aromatics

7

7

8

92.1

XOPN

Aromatic ring opening product (unsaturated dicarbonyl)

5

5

6

XYL

Xylene and other polyalkyl aromatics

8

8

10

NTR1

Simple organic nitrates

4

9

3

119.1

NTR2

Multi‐functional organic nitrates

4

9

4

135.1

ECH4

Emitted methane (to enable tracking seperate from CH4)

1

1

4

XPRP

Operator for organic nitrates from PRPA

3

3

7

2

89.1

XPAR

Operator for organic nitrates from PAR

1

5

11

2

117.1

CRNO

Nitro‐cresol oxy radical

7

7

6

3

152.1

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

Carbon #

4

C

4

9

1
3

3

8

4

10
2

104

42.1

72.1
4

79.0
44.1

2

90.1

2

64.0

4

2

98.0

98.1
106.2

16.0

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
5. Chemistry Mechanisms

Species

Description

Carbon #

C

H

O

Mol. Wt.

CRN2

Nitro‐cresol peroxy radical

7

7

6

4

168.1

CRPX

Nitro‐cresol hydroperoxide

7

7

7

4

169.1

CAO2

Ring‐opening product from methyl catechol

7

7

9

5

173.1

via aqueous‐phase reactions (Carlton et al., 2007), were added in CB6 to improve support for
SOA modeling. Precursors to alpha‐dicarbonyls in CB6 are aromatics, alkenes and ethyne. CB6
included several updates to peroxy radical chemistry that improved formation of hydrogen
peroxide (H2O2) and therefore sulfate aerosol formation. The gas‐phase reaction of dinitrogen
pentoxide (N2O5) with water vapor is slower in CB6 than CB05, which reduced nighttime
formation of nitric acid, although heterogeneous reactions on aerosol surfaces may dominate
nitric acid formation at night (Brown et al, 2006). CB6 included the calculation of the
heterogeneous N2O5 hydrolysis rate as a function of nitrate, chloride, and water concentrations
in particles (Bertram and Thornton, 2009) when PM is explicitly simulated; if no PM chemistry is
included, CAMx sets the heterogeneous rate to the IUPAC (2015) N2O5 hydrolysis rate.
The core inorganic chemistry mechanism for CB6 was based on evaluated data from the IUPAC
tropospheric chemistry panel as of January, 2010 (Atkinson et al., 2010). IUPAC also was the
primary source for photolysis data in CB6 with some data from the 2006 NASA/JPL data
evaluation (Sander et al., 2006) or other sources for photolysis of some organic compounds.
There were changes to the organic chemistry for alkanes, alkenes, aromatics and oxygenates.
The most extensive changes were for aromatics and isoprene. Chemistry updates for aromatics
were based on the updated toluene mechanism (CB05‐TU) developed by Whitten et al. (2010)
extended to benzene and xylenes. The isoprene mechanism was revised based on several
recently published studies (Paulot et al., 2009a,b; Peeters et al., 2009).
CB6 was evaluated using 339 experiments from several chambers at the University of California
at Riverside and the Tennessee Valley Authority. The performance of CB6 and CB05 in simulating
chamber studies was comparable for alkanes, alkenes, alcohols and aldehydes with both CB6 and
CB05 performing well and exhibiting 20% or less bias for maximum ozone. For species that were
explicitly added in CB6 (ethyne, benzene and ketones), CB6 performed much better than CB05. For
aromatics, CB6 improved upon CB05 by reducing under prediction bias in maximum ozone to about
10% for benzene, toluene and xylene. For isoprene, both CB05 and CB6 show little bias for
maximum ozone (less than 5%) but CB6 tended to form ozone too slowly. CB6 improved upon
CB05 for simulating mixtures of VOCs. For mixtures without aromatics, both CB05 and CB6 showed
minimal bias for maximum ozone. For mixtures including aromatics, both CB05 and CB6 under
predicted maximum ozone but bias was reduced from about 30% for CB05 to about 20% for CB6.
CB6 revision 1 (CB6r1) included revised chemistry for isoprene and aromatic hydrocarbons and
more NOx‐recycling from the degradation of organic nitrates (Yarwood et al., 2012a). Revision
2 (CB6r2) increased detail in the formation and fate of organic nitrates (ON), including organic
nitrate destruction by reactions in aerosols (Hildebrandt Ruiz and Yarwood, 2013). ONs are
formed when VOCs degrade in the presence of NOx and are important in the atmosphere
because they sequester NOx and can contribute to organic aerosol (OA). NO2 is released when
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

105

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
5. Chemistry Mechanisms

ONs degrade by photolysis in the gas‐phase, returning NOx to the atmosphere where it may
contribute to ozone production. CB6r2 differentiates organic nitrates between simple alkyl
nitrates that remain in the gas‐phase and multi‐functional ONs that can partition into OA
(Hildebrandt Ruiz and Yarwood, 2013). ONs present in aerosols are then assumed to undergo
hydrolysis to nitric acid with a lifetime of approximately 6 hours based on laboratory
experiments and ambient data. These changes tend to reduce regional concentrations of ozone
and ONs, and increase nitric acid. Regional modeling simulations using CAMx with CB6r2 show
that accounting for ON hydrolysis in aerosols improve performance for ozone and in simulating
the partitioning of NOy between ONs and nitric acid.
5.1.1.3 Carbon Bond Version 6 with Halogen Chemistry
Mechanism 3 is an extension of CB6r2 chemistry that adds reactions involving ocean‐borne halogen
compounds (CB6r2h; Yarwood et al., 2014). Bromine reactions were integrated with previously
developed reactions for iodine (Yarwood et al., 2012b) and chlorine (Tanaka et al., 2003; Koo et al.,
2012) with rate constants updated to currently accepted values (IUPAC, 2014a and b) and
mechanism revisions to promote consistency. The additoinal halogen compounds and reactions
added to CB6r2 are listed in Appendix B.
The chlorine (Cl) reaction mechanism is based on Koo et al. (2012) with the following updates:







Reaction rate constants updated to IUPAC (2014a and b) as necessary;
Cl‐atoms with organic compounds are limited to alkanes and isoprene
Added ClO radical reactions with BrO and IO
Added ClNO3 hydrolysis to HOCl on aerosols
Cl‐atom reactions with organic compounds limited to alkanes and isoprene

Reactions of Cl‐atoms with organic compounds are limited to alkanes and isoprene. Cl‐atom
production from the photolysis of chloromethanes is included only for those halomethanes that
are included as sources of Br from seawater. Degradation of anthropogenic chlorocarbons (e.g.,
HCFCs) is not included in the mechanism. The dominant source of atmospheric Cl is expected
to be sea salt emissions. Hydrochloric acid (HCl) is displaced into the gas phase when sea salt
aerosols are acidified by nitric and sulfuric acids. The HCl formed from sea salt can react with
dinitrogen pentoxide (N2O5) on aerosol surfaces to produce nitryl chloride (ClNO2) which
photolyzes to produces Cl‐atoms. When PM is explicitly modeled, the heterogeneous reaction
rate for N2O5 + HCl is calculated using the parameterization developed by Bertram and
Thornton (2009).
The bromine (Br) reaction mechanism is similar to the mechanisms of Yang et al. (2005),
Smoydzin and von Glasow (2009) and Parrella et al. (2012) and is more compact than the
mechanisms of Vogt et al. (1999), Whitten and Yarwood (2008) and Ordóñez et al. (2012).
Reaction rate constants for the Br mechanism are from IUPAC (2014a and b). Hydrolysis of
BrNO3 is included as pseudo gas‐phase reaction with a rate constant comparable to hydrolysis
of N2O5. The largest source of atmospheric Br is sea salt aerosol (Yang et al., 2005) although the
mechanism by which sea salt Br enters the gas‐phase differs from that for Cl depletion under
acid conditions (discussed above) and this can also occur at neutral pH. Other sources of
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

106

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
5. Chemistry Mechanisms

atmospheric Br included in the mechanism are decomposition of the halomethanes CHBr3,
CH2Br2, CH2BrCl, CHBr2Cl and CHBrCl2.
The iodine (I) reaction mechanism is based on Yarwood et al. (2012b) with the following
updates:






Reaction rate constants updated to IUPAC (2014a and b) as necessary
Added IO radical reactions with ClO and BrO
Added INO3 hydrolysis to HOI on aerosols
Removed INO and related reactions

Reactions of INO were removed because INO concentrations were found to be small (Yarwood
et al., 2012b). Hydrolysis of INO3 is included as pseudo gas‐phase reaction with a rate constant
comparable to hydrolysis of N2O5.
Emissions from oceans are the major source of atmospheric iodine (Carpenter, 2003), including
methyl iodide (CH3I), other iodo‐methanes (CH2I2, CH2ICl, CH2IBr), larger alky iodides, and molecular
iodine (I2). Iodine emissions result both from biological and photochemical processes in ocean
water (Moore and Tokarczyk, 1993; Moore and Zafirou, 1994). Photochemical processes that cause
iodine emissions are linked to reactions of dissolved ozone and thereby to enhanced ozone
deposition to oceanic waters (Ganzeveld et al., 2009; Helmig et al., 2012).
Reactions among the radicals ClO, BrO, and IO are included to interconnect the mechanisms for
different halogens. Atmospheric reactions of Cl‐atoms, Br‐atoms and I‐atoms can produce or
destroy tropospheric ozone through a series of catalytic cycles, where each halogen atom is
regenerated in the reactions and therefore one atom can potentially destroy many O3
molecules. Catalytic destruction of O3 by Cl and Br is terminated only when deposition removes
reservoir species, e.g., by dry or wet deposition of HCl and HBr. The atmospheric reactions of I‐
atoms differ from Br and Cl in several ways:





I‐atoms do not abstract H from organic compounds in contrast to Br and Cl‐atoms;
Formation of oxides is more extensive for I (IO, OIO, I2O2, IxOy) than for Br (BrO) or Cl
(ClO);
Larger iodine oxides (IxOy) form aerosols whereas Cl and Br oxides remain in the gas
phase.

Aerosol formation by larger iodine oxides is a sink for reactive I that can terminate O3
destruction by reactive I.
5.1.1.4 Carbon Bond Version 6, Revision 3
Mechanism 4 is CB6 revision 3 (CB6r3), which includes updates to improve NO2‐alkyl nitrate
branching in cold conditions (Emery et al., 2015). Alkyl nitrate formation can influence ozone
production because both NO and radicals are terminated by alkyl nitrate formation. However,
temperature dependence of NO2‐alkyl nitrate branching is omitted from current photochemical
model mechanisms, i.e., CB05‐TU (Whitten et al., 2010), CB6 (Hildebrandt Ruiz and Yarwood,
2013), SAPRC11 (Carter and Heo, 2013) and RACM2 (Goliff et al., 2013) and also from the
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS
www.camx.com
107

March 2016

CAMx User’s Guide Version 6.3
5. Chemistry Mechanisms

explicit Master Chemical Mechanism (http://mcm.leeds.ac.uk/MCM/). Lee et al. (2014)
considered how cold winter conditions affect alkyl nitrate branching and concluded that
omitting the temperature dependence may cause a 15% high bias in ozone formation.
We have developed the CB6r3 chemical mechanism from CB6r2 to extend applicability to
winter and summer conditions. The temperature and pressure dependent formation of alkyl
nitrates includes reactions involving propane (CB6 species “PRPA”) and other alkanes (CB6
species “PAR”). CB6r3 was designed to produce the same alkyl nitrate yields as CB6r2 at room
temperature and pressure (298 K and 1 atm). See Appendix D for a complete listing of
reactions, rate expressions, and VOC properties.
Alkyl nitrates (RONO2) are formed when alkanes are oxidized in the atmosphere in the presence
of nitric oxide (NO). Alkanes are compounds of hydrogen and carbon with only single bonds
connecting the atoms, e.g., methane (CH4), ethane (C2H6), propane (C4H8), etc. Analyses of air
samples collected in western US oil and gas development basins during wintertime ozone
events show that alkanes dominate the organic gases present in the air. The formation of alkyl
nitrates from alkanes can be described by the following reactions in which an alkane (RH) reacts
with hydroxyl radical (HO•)3 and oxygen (O2) to form an alkyl peroxy radical (RO2•) that has two
potential reaction pathways with NO•:
1)
2)
3a)
3b)

HO• + RH
R• + O2
RO2• + NO•

 R• + H2O
 RO2•
 RONO2
 RO• + NO2•

Perring et al. (2013) have reviewed the atmospheric impacts of alkyl nitrate formation. The
yield of alkyl nitrate is determined by the branching ratio among reactions 3a and 3b, which
depends on both temperature and pressure (Atkinson et al., 1983). The association reaction of
RO2 with NO in reaction 3a is favored over reaction 3b at lower temperatures and higher
pressures.
Emery et al. (2015) confirmed the directionality of the ozone effect hypothesized by Lee et al.
(2014) when the most recent temperature/pressure equations of Arey et al. (2001) for alkyl
nitrate branching were incorporated into CB6r3, representing the alkane mix of a high winter
ozone episode in the Uintah Basin of Utah. Recent experimental data of Yeh and Ziemann
(2014) confirm the expression of Arey et al. (2001) for n‐alkanes containing 3 to 14 carbon
atoms. Weighting measured organic gas concentrations by their OH reactivity indicates which
species are most likely to participate in ozone formation. Considering the alkanes represented
by CB species “PAR” measured in the Uintah Basin, those with 4 to 7 carbon atoms dominate
OH reactivity and indicate that the temperature/pressure dependence for pentane (with 5
carbons) may be considered representative for alkanes in the Uinta Basin. While CB6r3 is
suitable for representing the alkane mixture reacting in the Uinta Basin, the derivation of CB6r3
does not rely upon this particular mixture of alkanes.
3

The dot signifies that hydroxyl (HO) is a radical, i.e., has one unpaired electron. Note that NO and NO2 also are
radicals. O2 is a di‐radical.
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS
www.camx.com
108

March 2016

CAMx User’s Guide Version 6.3
5. Chemistry Mechanisms

5.1.2 SAPRC 2007
The 2007 update to the SAPRC chemistry mechanism, called SAPRC07 (Carter, 2010), replaced
the dated SAPRC99 mechanism. The version implemented in CAMx is SAPRC07TC, which
includes additional model species to explicitly represent selected toxics and reactive organic
compounds and uses numerical expressions of rate constants that are compatible with the
current chemistry mechanism solver (Hutzell et al., 2012). Chlorine chemistry is not included in
the CAMx implementation. See Appendix E for a complete listing of reactions, rate expressions,
species definitions, and VOC properties.
5.1.3 Implicit Gas‐Phase Species
All photochemical mechanisms in CAMx employ fixed concentrations for molecular oxygen (O2),
molecular hydrogen (H2), and methane (CH4). Concentrations for these compounds are set to
the following constant mixing ratios (i.e., they are not impacted by the chemical solution):
[O2] = 2.095×105 ppm
[H2] = 0.60 ppm
[CH4] = 1.75 ppm
Mechanisms 2 and 3 (CB6r2) includes a species named ECH4 to represent emitted methane
over and above the global background of 1.75 ppm.
5.1.4 Photolysis Rates
The rates of atmospheric photolysis reactions depend upon solar irradiance and therefore are
sensitive to the amount of solar radiation transmitted through the atmosphere as well as
reflected from the earth’s surface (albedo). Photolysis rates are externally derived assuming
clear‐sky conditions as a function of five parameters: solar zenith angle, altitude above ground,
total ozone column, surface albedo, and terrain height. The rates are provided to CAMx as a
large lookup table that spans the range of conditions for each of the five dimensions. The
lookup table is developed using a CAMx pre‐processor that incorporates the Tropospheric
Ultraviolet and Visible (TUV) radiative transfer model (NCAR, 2011). TUV employs a standard
atmosphere density profile for Rayleigh scattering and other absorbers such as oxygen. User‐
specified ozone column values are used to scale a typical vertical ozone profile within TUV. A
default aerosol profile from Elterman (1968) is combined with typical aerosol optical properties
within TUV to account for haze.
The CAMx version of TUV is modified to output photolysis rate information in a format directly
compatible with all CAMx photochemical mechanisms. See Sections 2 and 3 for more
information on developing photolysis inputs.
As CAMx runs, the lookup rates are interpolated to the specific conditions in each grid cell.
They are then adjusted for any local cloud cover and local aerosol attenuation (if PM is
simulated). Additionally, solar angle‐dependent temperature and pressure adjustments are
applied to five key photolysis reactions (NO2, O3, acetaldehyde, and two formaldehyde
reactions).
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

109

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
5. Chemistry Mechanisms

5.1.4.1 Cloud and Aerosol Adjustments
Photochemistry is strongly influenced by the presence of clouds, which can both attenuate and
enhance the actinic flux of ultraviolet (UV) and visible radiation responsible for photolysis.
Their specific radiative impacts depend on many factors, including height, depth and fractional
sky cover; water content; and water phase (i.e., liquid droplets or ice crystals). Aerosols also
influence photochemistry according to their optical properties and mass loading as a function
of altitude and depth. CAMx includes a fast in‐line version of TUV (Emery et al., 2010) to
calculate photolysis adjustment profiles through each cloudy, aerosol‐laden grid column.
The in‐line TUV is run twice for each grid column: first for non‐cloudy conditions with the same
Elterman (1968) aerosol profile used in the full‐science TUV pre‐processor, and second with
clouds and simulated aerosols to derive a vertical profile of the cloudy:clear actinic flux ratio (in
the case that aerosols are not run in CAMx, the Elterman profile is used consistently). This ratio
is then applied as a multiplicative factor to the clear‐sky value in each grid cell. This approach
maintains accuracy in the calculation of clear‐sky photolysis rates, while allowing clouds and
aerosols to be directly involved in radiative transfer calculations through each grid column.
TUV includes a calculation of integrated atmospheric density above the CAMx domain, based
on the U.S. standard atmosphere, so that atmospheric attenuation of the UV stream is properly
calculated entering the model top. Other aspects of the in‐line TUV model were substantially
streamlined to minimize runtimes. First, radiative calculations are performed for only a single
representative wavelength (350 nm). Second, since absorption by gases occurs in rather
narrow UV bands relative to the broad‐band influence of clouds, the absorption from oxygen,
ozone, nitrogen dioxide and sulfur dioxide were removed. Third, the extraterrestrial flux was
not needed as it cancels out in the calculation of the cloudy: clear ratio. Finally, the plane‐
parallel version of the delta‐Eddington approach was used in lieu of the more complex and
expensive pseudo‐spherical geometry. Preliminary tests against the full‐science TUV showed
that the streamlined version resulted in less than 1% differences in actinic flux ratio for a range
of cloudy conditions (Emery et al., 2010).
Optical depth  expresses the reduction of incident light I0 through a light attenuating medium
of depth z according to
I



I 0 e 

The in‐line TUV adjustment scheme utilizes cloud optical depth fields provided by the CAMx
cloud/rain file, and aerosol optical depths calculated from the PM mass concentrations
simulated by CAMx.
The CAMx meteorological interface pre‐processors generate cloud water and optical depth
fields from the variable fields present in the raw meteorological output files. Cloud optical
depth is calculated in each model grid cell according to the approach of Del Genio et al. (1996)
and Voulgarakis et al. (2009), which satisfactorily approximates the effects of random cloud
overlap according to

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

110

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
5. Chemistry Mechanisms





3 L z
Fc 3 / 2
2  w rd

where L is cloud liquid water content (g/m3), z is the cell depth containing cloud water, w is
the density of liquid water (106 g/m3), and Fc is fractional cloud cover. The mean cloud droplet
radius rd is not dependent on water phase, and is assumed to be a typical tropospheric value of
1.510‐5 m for liquid cloud water. TUV assumes constant Mie scattering parameters for clouds:
a single scattering albedo of 0.99, and an asymmetry factor of 0.85.
When CAMx is run with PM, vertical haze opacity profiles are calculated from simulated aerosol
concentration fields. When CAMx is run with only gas‐phase chemistry without aerosols,
photolysis rates are only adjusted for clouds. Aerosol optical parameters are best determined
from Mie theory, but in the interest of minimizing impacts to model speed and considering the
degree of uncertainty in the simulated aerosol concentrations themselves, a simpler method
was adopted. Aerosols are assumed to exist as an external mixture of their component
chemical species. Aerosol light extinction (scattering and absorption) is a function of each
species’ concentration, extinction efficiency, and affinity for hygroscopic growth. Total aerosol
optical depth is determined by summing extinction over all species and multiplying by layer
depth.
Dry extinction efficiencies and single‐scattering albedos for each aerosol species, valid at 350
nm, are externally defined in the CAMx chemistry parameters file. While these can be altered
by the user, the chemistry parameters files that are provided with the CAMx distribution
include default values according to Takemura et al. (2002), as shown in Table 5‐3.

Table 5‐3. Default dry extinction efficiency and single‐scattering albedo at 350 nm (Takemura
et al., 2002) in the distributed CAMx chemistry parameters file.
Species
Sulfate
Nitrate
Ammonium
Organics
Elemental Carbon
Crustal (Fine+Coarse)
Sea salt (Na+Cl)

Dry Extinction Efficiency
(m2/g)
710‐6
710‐6
710‐6
710‐6
1810‐6
0.410‐6
1.510‐6

Single‐Scattering
Albedo
0.99
0.99
0.99
0.80
0.25
0.70
0.99

Takemura et al. (2002) provide extinction efficiencies and single‐scattering albedos for sulfate,
organics, soot, total dust, and sea salt; we have extended the sulfate values to nitrate and
ammonium. The asymmetry factor is internally set to a default value of 0.61 regardless of the
composition of the aerosols.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

111

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
5. Chemistry Mechanisms

Hygroscopic aerosols are also flagged in the chemistry parameters file. For each flagged
species, an internal growth factor is applied to the dry extinction efficiencies according to the
relative humidity conditions in each grid cell. The growth curve is taken from the Phase I report
of the Federal Land Managers’ Air Quality Related Values Workgroup (FLAG, 2000). By default,
the relative humidity growth factor is flagged for sulfate, nitrate, ammonium and sea salt; a
single growth factor is applied for all hygroscopic species (Figure 5‐1). Minimum and maximum
limits on relative humidity are set at 1% and 95%, respectively.

Extinction Adjustment, f(RH)
25

F(RH)

20

15

10

5

0
0

5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95

Relative Humidity (%)

Figure 5‐1. Relative humidity adjustment factor applied to the dry extinction efficiency for
hygroscopic aerosols (FLAG, 2000).
5.1.4.2 Effects of Surface Albedo and Snow Cover
Photolysis rates depend on the amount of solar radiation reflected from the Earth’s surface
(albedo). UV albedo is assigned within CAMx according to the distribution of gridded landuse
provided by the time‐invariant 2D surface file (Tables 3‐4 and 3‐5). Snow‐free UV albedos fall in
the range 0.04 to 0.08 and are constant in time. Analyses of reflected UV radiation recorded in
satellite data (Herman and Celarier, 1997) report similar UV albedo values in the range 0.02‐
0.08 for typical terrestrial and water surfaces. Snow is much more reflective than other types
of surfaces and so it is important to characterize the effect of snow cover on photolysis rates.
The CAMx photolysis rate input file is generated for five surface albedos, two of which
represent the non‐snow range (0.04 – 0.10) and four that represent the snow range (0.1 – 0.2 –
0.5 – 0.9). CAMx determines the landuse‐ and snow‐weighted average surface albedo in each
grid cell (Section 4.7) and interpolates photolysis rates between the five albedos.
5.1.5 Gas‐Phase Chemistry Solvers
Solving the time evolution of gas‐phase chemistry requires numerically integrating a set of
ordinary differential equations (ODEs) and is among the most computationally expensive
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

112

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
5. Chemistry Mechanisms

operations performed in a photochemical grid model. One reason for this is that the ODEs
describing tropospheric chemistry are “stiff” – meaning that the chemical species involved have
widely varying production and/or decay times. The computational efficiency of the gas‐phase
chemistry solver strongly influences the overall efficiency of a grid model. CAMx includes two
chemistry solvers that offer trade‐offs between accuracy and efficiency.
The most accurate solution methods available for stiff ODEs are “Gear” type implicit solvers
(Gear, 1971) such as LSODE (Hindmarsh, 1983). Gear solvers are stable when applied to “stiff”
problems, such as gas‐phase chemistry, but are generally too slow for routine use in grid model
applications. Hertel et al. (1993) developed an implementation of the Euler Backward Iterative
(EBI) method that is very efficient and also accurate because it utilizes explicit algebraic
formulae to solve several important groups of species (HOx, NOx, etc.).
5.1.5.1 LSODE
CAMx includes the double precision version of the Livermore Solver for Ordinary Differential
Equations (LSODE; Hindmarsh, 1983) distributed by the Netlib repository of numerical
algorithms (http://www.netlib.org/). LSODE is too slow for everyday use but valuable as a
reference method within CAMx. LSODE is based on Gear’s method with numerical refinements
to improve efficiency and ease of use (Radhakrishnan and Hindmarsh, 1993). Gear methods
(Gear, 1971) are implicit and employ backwards‐differentiation formulae to step forward in
time by taking multiple steps. The converged solutions at each step are saved in a history
matrix and used to predict the next solution. Thus, LSODE must initially take short time steps to
build the history matrix and may then take progressively longer steps. LSODE is most efficient
for long integration times (and inefficient for short integration times) and therefore least
burdensome for coarse grid model applications that have relatively long coupling times
between gas‐phase chemistry and other processes, e.g., advection.
User‐supplied information required by LSODE is essentially the error control parameters and
the functions defining the system of ODEs, f(y,t), where y is the vector of species concentrations
and t is time. Supplying a subroutine to evaluate the time derivatives of species concentrations
(f = dy/dt) is mandatory. Supplying a function to evaluate the Jacobian matrix (J = df/dy) is
optional since, if not supplied, LSODE can derive a numerical Jacobian by finite difference
between repeated evaluations of f. Supplying an algebraic Jacobian ensures accuracy, although
a numerical Jacobian may be equally accurate if adequate precision (e.g., double precision) is
employed. Supplying an algebraic Jacobian is more efficient when J is sparse, but for
condensed mechanisms such as CB05 J is not sparse and the numerical Jacobian method is
faster. CAMx uses the numerical Jacobian method with a relative error tolerance of 10‐7 and an
absolute error tolerance of 10‐10.
5.1.5.2 EBI Solver
The backward Euler method solves concentrations (y) as y(t+h) = y(t) + hf, where f is the time
derivative of species concentrations (f= dy/dt) evaluated at t+h. The method must be iterated
to convergence in y(t+h) because species concentrations are interdependent. The basic EBI
method is not efficient for stiff problems such as tropospheric chemistry because convergence
is slow and the step size (h) must be short. Hertel et al. (1993) greatly improved the efficiency
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

113

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
5. Chemistry Mechanisms

and accuracy of the method by developing analytical solutions to the EBI equation for groups of
species that are strongly coupled (e.g. HOx and NOx/O3). Time steps of up to 3 minutes are
taken by the Hertel EBI solver in CAMx.

5.2 Aerosol Chemistry
The gas‐phase chemistry is run alone (no aerosols) by supplying CAMx with a chemistry
parameters file with the aerosol option keyword set to “NONE”, the number of aerosol species
set to zero, and the entire list of aerosol species parameters omitted (see Section 3). Aerosols
are treated by supplying CAMx with a chemistry parameters file with the aerosol keyword set
to “INERT”, “CF”, “CF_VBS” or “CMU”. In all such cases, the number of aerosol species, the
number of size sections and their size ranges, and various aerosol parameters are specified.
The aerosol keyword “INERT” allows the user to define any number of arbitrarily‐named inert
particulate species to be carried by the model during a photochemical simulation (e.g.,
modeling the dispersion of only wind‐blown dust).
Aerosol chemistry processes can be run together with gas‐phase chemistry using two options
for treating aerosol size distributions: the CF scheme and CMU scheme. The CF scheme divides
the size distribution into two static modes (coarse and fine). Primary species can be modeled
as fine and/or coarse particles, while all secondary (chemically‐formed) species are modeled as
fine particles only. The CMU scheme employs a sectional approach that dynamically models
the size evolution of each primary and secondary aerosol constituent across a number of fixed
size sections. The CF and CMU options require a minimum set of specific aerosol species with
associated chemistry. Aerosol water is explicitly treated in both CF and CMU options, which
affects aerosol size and density.
5.2.1 Additional Gas‐Phase Species
When either the CF or CMU aerosol option is selected, the following gas‐phase species are
added to model gas‐aerosol interactions:
1)
2)
3)
4)

Ammonia (NH3) as a precursor for inorganic aerosol.
Gaseous sulfuric acid (SULF) as a precursor to sulfate aerosol.
Sodium (Na) and hydrogen chloride (HCL) as products of acidified sea salt aerosol.
Separately‐tracked emitted (“primary”) VOCs that form intermediate organic condensable
gas (CG) species via oxidation reactions: toluene, xylene, monoterpenes, sesquiterpenes,
and isoprene.
5) Several intermediate CG species that may condense to secondary organic aerosol (SOA) or
are products of SOA volatilization.

5.2.2 Aerosol Processes
Aerosol chemical and thermodynamic processes include the following:
1) Aqueous sulfate and nitrate formation in resolved cloud water using the RADM aqueous
chemistry algorithm (Chang et al., 1987).
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

114

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
5. Chemistry Mechanisms

2) Partitioning of inorganic aerosol constituents (sulfate, nitrate, ammonium, and natural
minerals) between the gas and aerosol phases using the ISORROPIA thermodynamic
module (Nenes et al., 1998, 1999); uptake of nitric acid by calcium in soil dust particles is
calculated external to ISORROPIA.
3) Organic aerosol‐gas partitioning and oxidation chemistry using two options:
a. A semi‐volatile equilibrium scheme called SOAP (Strader et al., 1999) that forms a
condensed “organic solution phase”;
b. A hybrid 1.5‐dimension volatility basis set (1.5‐D VBS) approach (Koo et al., 2014)
providing a unified framework for gas‐aerosol partitioning and chemical aging of both
primary and secondary organic aerosols.
Organic aerosol treatments are described in more detail in separate sub‐sections below.
Aqueous chemistry is not explicitly applied to sub‐grid clouds; clouds are assumed to either
occupy the entirety of a grid cell volume, or be completely absent from it. The cell‐averaged
effect of sub‐grid clouds is treated in the CAMx meteorological preprocessors that generate
three‐dimensional gridded cloud input fields. Cloudy grid cells are determined by cloud liquid
water contents above a threshold of 0.05 g/m3. Aqueous chemistry is calculated for each cloud
grid cell at each model time step.
In cloudy grid cells undergoing aqueous chemistry, the ISORROPIA equilibrium algorithm is
called every time step to ensure that rapidly evolving sulfate, nitrate and neutralizing cations
are in balance with the local environment. In cloud‐free grid cells, ISORROPIA is called on a
unique aerosol “coupling” time step that is defined within the chemistry parameters file. By
default, the aerosol coupling time step is 15 minutes, and this is used for all master and nested
grids in a simulation regardless of the grid‐specific driving time step.
Uptake of nitric acid on mineral dust particles is one of the pathways of particle nitrate
formation. For example, calcium in soil dust particles reacts with nitric acid to form calcium
nitrate. Based on Saharan dust study (Astitha et al., 2009), we estimate about 6% mass fraction
of calcium carbonate (CaCO3) in fine dust particles (FCRS), and half of it is assumed to be
replaced by calcium nitrate. Since the current ISORROPIA implementation in CAMx does not
consider mineral cations other than sodium, nitrate uptake by calcium in soil dust is calculated
external to ISORROPIA. CAMx outputs total particulate nitrate, i.e., the sum of particle nitrate
determined by ISORROPIA and calcium nitrate.
Table 5‐4 shows the inorganic aerosol species that can be included with the CF scheme. Some
species must be present for this scheme (“Mandatory Species”) to establish linkages between
gas and aerosol phase chemistry. Other species are optional (meaning that they can be
removed from the chemistry parameters input file), except that sodium and chloride must
always be present or absent together (i.e., one cannot be present without the other). If sodium
and chloride are not modeled then default background values are used within CAMx.
In the CMU scheme, CRST is used to identify all primary inert material, which replaces the CF
species of FPRM, FCRS, CPRM, and CCRS in Table 5‐4. Individual aerosol species names specify
both the constituent and the size section using a set naming convention, e.g., PSO4_1 refers to
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

115

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
5. Chemistry Mechanisms

Table 5‐4. List of inorganic PM species for the CAMx CF aerosol option.
Internal Label
PSO4
PNO3
PNH4
PH2O
NA
PCL
PEC
FPRM
FCRS
CPRM
CCRS

Name
Sulfate
Particulate Nitrate
Particulate Ammonium
Aerosol Water Content
Sodium
Particulate Chloride
Primary Elemental Carbon
Fine Other Primary (diameter2.5 µm)
Fine Crustal (diameter2.5 µm)
Coarse Other Primary
Coarse Crustal

Mandatory Species
X
X
X
X

particle sulfate in size section 1. The CMU scheme requires that the complete list of all aerosol
species be present in the chemistry parameters file (i.e., no aerosol species are optional).
5.2.2.1 SOAP
SOAP is the default SOA chemistry/partitioning module when the aerosol keyword is set to “CF”
or “CMU” in the chemistry parameters input file. Directly emitted (primary) organic aerosol is
treated by SOAP as a single non‐volatile species called POA that does not chemical evolve.
However, POA does influence the evolution of SOA. SOA species exist in equilibrium with
condensable gasses (CG) that can be produced by VOC oxidation:
VOC + oxidant → CG ↔ SOA
CG formation from VOC oxidation reactions (Table 5‐5) is handled within the SOAP module
rather than the main gas‐phase chemistry, as described below. This approach has the following
advantages: (1) separates the VOC precursors and lumping schemes for oxidant chemistry and
SOA formation (e.g., for aromatics, different lumping schemes may be appropriate for oxidant
and SOA formation); (2) allows the same SOA mechanism to be used with different oxidant
mechanisms; (3) allows inclusion of SOA precursors without explicitly defining oxidant reactions
(e.g., sesquiterpenes are explicit in the SOA module but their oxidant formation may be
represented by surrogate species).
Emissions of SOA precursors must be provided separately from the emissions of oxidant
precursors, e.g., isoprene emissions must be speciated both as ISOP for oxidant chemistry and
ISP for SOA chemistry. This “double counting” of emissions is correct because the species (e.g.,
ISOP and ISP) serve different purposes. Ideally, emissions processors will develop VOC
speciation schemes that suit both oxidant and SOA modeling. In the absence of more refined
information, emissions of SOA precursors may be set equal to emissions of oxidant precursors
as follows: ISP = ISOP; TRP = TERP; BNZA = BENZ; TOLA = TOL or ARO1; XYLA = XYL or ARO2.
Some emitted VOCs are semi‐volatile (SVOCs) and can condense directly to SOA, e.g., biogenic
emissions of oxygenated VOCs. SOA formation from SVOC emissions may be accounted for by
including the SVOC in the CAMx emissions as one of the CGs listed in Table 5‐6. Choose a CG
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

116

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
5. Chemistry Mechanisms

Table 5‐5. SOA precursor reactions included in the CAMx SOAP module.
Precursor

Reaction

Toluenes

TOLA + OH  TO2
TO2 + NO
TO2 + HO2
XYLA + OH  XLO2
XLO2 + NO
XLO2 + HO2
BNZA + OH  BZO2
BZO2 + NO
BZO2 + HO2

CG Products1

k2982

Anthropogenic

Xylenes

Benzene

Isoprene

ISP + OH
ISP + O3
ISP + NO3
TRP + O
TRP + OH
TRP + O3
TRP + NO3
SQT + OH
SQT + O3
SQT + NO3

Terpenes

Sesquiterpenes

Notes:
1
2
3

5.63E‐12
9.04E‐12
1.49E‐11
1.85E‐11
9.04E‐12
1.49E‐11
1.22E‐12
9.04E‐12
1.49E‐11

0.036 CG1 + 0.069 CG2
0.22 SOAH3
0.022 CG1 + 0.064 CG2
0.21 SOAH3
0.037 CG1 + 0.46 CG2
0.19 SOAH3
Biogenic
0.015 CG3 + 0.12 CG4
none
none
0.065 CG5 + 0.29 CG6
0.065 CG5 + 0.29 CG6
0.065 CG5 + 0.29 CG6
0.065 CG5 + 0.29 CG6
0.85 CG7
0.85 CG7
0.85 CG7

9.99E‐11
1.27E‐17
6.74E‐13
3.60E‐11
6.77E‐11
7.63E‐17
6.66E‐12
1.97E‐10
1.16E‐14
1.90E‐11

Yield values are in ppm/ppm.
Rate constants are shown for 298 K and 1 atmosphere in molecules/cm‐3 and 1/s.
SOAH represents non‐volatile oxidation products.

Table 5‐6. Properties of CG/SOA pairs in the CAMx SOAP module.
Species
CG1/SOA1
CG2/SOA2
CG3/SOA3
CG4/SOA4
CG5/SOA5
CG6/SOA6
CG7/SOA7
SOAH
SOPA
SOPB

Molecular
Weight
(g mole‐1)
150
150
130
130
180
180
210
150
220
220

Saturation
Concentration
(µg m‐3 at 298 K)
1.15
81.6
0.726
136
3.92
55.8
0
0
0
0

Heat of
vaporization
(kJ mole‐1)
19.9
18.0
42.0
42.0
75.5
75.5
–
–
–
–

that has appropriate volatility properties, and account for any molecular weight difference
between the SVOC and surrogate CG (moles CG emitted = moles SVOC x MWSVOC/MWCG).
The SOAP module consists of two parts: gas‐phase oxidation chemistry that forms CG, and
equilibrium partitioning between gas and aerosol phases for each CG/SOA pair. The physical
properties of CG/SOA pairs are shown in Table 5‐6. The CG yields are expressed as ppm of CG
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

117

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
5. Chemistry Mechanisms

formed per ppm of VOC reacted so that CG concentrations follow the CAMx convention for
gases and are in ppm. The SOAs formed from the CGs are in units of µg/m3 as are all other
aerosol species.
Polymerization reactions in organic aerosol phases will increase the molecular weight of the
condensed aerosol and reduce the volatility. Detailed descriptions of polymerization depend
upon the chemical composition of the organic and inorganic aerosol phases (e.g., aerosol
acidity). SOAP assumes that SOA is polymerized to a non‐volatile form (i.e., moved
permanently to the aerosol phase) with a lifetime of about 1 day (Kalberer et al., 2004).
Polymerization slowly forms organic aerosol polymers called SOPA (anthropogenic) and SOPB
(biogenic).
Total SOA is the sum of SOA1‐7 plus SOAH, SOPA and SOPB. Total organic aerosol is the sum of
total SOA and the single POA species.
5.2.2.2 1.5‐D VBS
The VBS organic aerosol (OA) chemistry/partitioning module is selected when the aerosol
keyword is set to “CF_VBS” in the chemistry parameters input file. VBS works with the 2‐mode
CF size option but is not currently compatible with the CMU sectional size option.
The VBS approach (Donahue et al., 2006; Robinson et al., 2007) provides a unified framework
for gas‐aerosol partitioning and chemical aging of both POA and SOA. It uses a set of semi‐
volatile OA species with volatility equally spaced in a logarithmic scale (the basis set). VBS
member species are allowed to react further in the atmosphere (chemical aging) to describe
volatility changes (i.e., shifting between volatility bins). First generation VBS models use one‐
dimensional basis sets (1‐D VBS) wherein organic compounds are grouped only by volatility and
thus are unable to describe varying degrees of oxidation observed in atmospheric OA of similar
volatility. To overcome this shortcoming, a two dimensional VBS (2‐D VBS) was developed
where organic compounds are grouped by oxidation state as well as volatility (Donahue et al.,
2011, 2012). However, use of 2‐D VBS in a 3‐D PGM has been limited due to high
computational cost.
A hybrid VBS approach is implemented in CAMx, called 1.5‐D VBS, which combines the
simplicity of the 1‐D VBS with the ability to describe evolution of OA in the 2‐D space of
oxidation state and volatility (Koo et al., 2014). Figure 5‐2 shows a schematic diagram of the
1.5‐D VBS scheme currently implemented in CAMx. This scheme uses five basis sets to describe
varying degrees of oxidation in ambient OA: two basis sets for chemically aged oxygenated OA
(OOA; anthropogenic and biogenic) and three for freshly emitted OA (hydrocarbon‐like OA
[HOA] from meat‐cooking and other anthropogenic sources and biomass burning OA [BBOA]).
Each basis set has five volatility bins ranging from 10‐1 to 103 µg m‐3 in saturation concentration
(C*), which roughly covers the volatility range of semi‐volatile organic compounds (SVOCs). An
effective heat of vaporization (H) value of 35 kJ mole‐1 is used for all SOA species. For POA, H
is estimated using the following empirical formulas:

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

118

www.camx.com

March 201
16

CAMx User’s Guide Versio
on 6.3
5. Chemistry Mechan
nisms

Figure 5‐‐2. Schemattic diagram of
o the CAMxx VBS module. The mod
del VBS speccies name
consists of 4 charactters that indicate the ph
hase (P – parrticle; V – vaapor), the so
ource (A –
anthropo
ogenic; B – biogenic;
b
C – cooking; F – fire), the fformation (P
P – primary;; S – secondary),
and the volatility
v
bin
n number. The
T solid and dashed arrrows repressent gas‐aerrosol
partition
ning and che
emical aging,, respectively. The thic k colored arrrows repressent POA
emission
ns or oxidation of SOA precursors.
p

COMPREHEN
NSIVE AIR QUALITY
Y MODEL WITH EX
XTENSIONS

119

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
5. Chemistry Mechanisms

∆H

4

∆H

11

C*298K

85 kJ mole‐1

C*298K

(For biomass burning; May et al., 2013c)

85 kJ mole‐1

(For other primary; Ranjan et al., 2012)

Table 5‐7 lists the model OA compounds assigned to the volatility bins. Their molecular
structures were determined by placing them on the 2‐D volatility space defined by Donahue et
al. (2011, 2012). The 1.5‐D VBS scheme adjusts oxidation state as well as volatility in response
to chemical aging by simplifying the 2‐D VBS model. Chemical aging of SOA and oxygenated
POA is modeled by shifting OA mass along a pre‐defined pathway of the OOA basis set, which
reduces volatility while increasing oxidation state. POA aging, which would require different
pathways from the HOA (or BBOA) basis set to the OOA basis set, is simplified in this 1.5‐D VBS
scheme where oxidation products of POA are represented as a mixture of POA and OPOA in the
next lower volatility bins. The gas‐phase OH reaction rates for POA and anthropogenic SOA are
assumed to be 4x10‐11 and 2x10‐11 cm3 molecule‐1 s‐1, respectively. Aging of biogenic SOA is
disabled in our implementation based on previous modeling studies that found aging biogenic
SOA led to a significant over‐prediction of OA in rural areas (Lane et al., 2008; Murphy and
Pandis, 2009). Additional details on the 1.5‐D VBS model can be found elsewhere (Koo et al.,
2014; Hildebrandt Ruiz et al., 2015). Total OA is the sum of all OA in the five volatility bins from
primary formation (PAP + PCP + PFP) and from secondary formation (PAS + PBS).

Table 5‐7. Molecular properties of the 1.5‐D VBS species.
Basis Set

OOA

HOA

BBOA

Model Species
Namea
PAS0 & PBS0
PAS1 & PBS1
PAS2 & PBS2
PAS3 & PBS3
PAS4 & PBS4
PAP0 & PCP0
PAP1 & PCP1
PAP2 & PCP2
PAP3 & PCP3
PAP4 & PCP4
PFP0
PFP1
PFP2
PFP3
PFP4

C*b

OSC c

C#

O#

‐3

(µg m )
0d
1
10
100
1000
0d
1
10
100
1000
0d
1
10
100
1000

MW

OA/OC
‐1

0.102
‐0.188
‐0.463
‐0.724
‐0.973
‐1.52
‐1.65
‐1.78
‐1.90
‐2.00
‐0.704
‐1.02
‐1.29
‐1.52
‐1.73

7
7.25
7.5
7.75
8
17
17.5
18
18.5
19
10
11
12
13
14

4.90
4.38
3.84
3.30
2.74
2.69
2.02
1.34
0.632
0.0
4.32
3.60
2.85
2.08
1.27

(g mole )
172
167
163
158
153
278
275
272
268
266
205
208
211
213
215

2.05
1.92
1.81
1.70
1.59
1.36
1.31
1.26
1.21
1.17
1.71
1.58
1.47
1.37
1.28

a: See Figure 5‐2 for the model species naming convention.
b: Effective saturation concentration.
c: Average oxidation state of carbon.
‐3
d: Properties of the lowest volatility bins were estimated assuming C* = 0.1 µg m , but they actually represent all OA with C* ≤
‐3
0.1 µg m , and are treated as non‐volatile in the model.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

120

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
5. Chemistry Mechanisms

Table 5‐8 lists input emission species users need to prepare for the 1.5‐D VBS OA scheme.
Anthropogenic and biogenic VOC precursors are simply copied from the respective emissions
for the CB gas‐phase mechanisms (TOL, XYL, BENZ, ISOP, TERP) to their “A” and “B”
counterparts. Sesquiterene (SQT) is entirely biogenic and identical to that used by SOAP. The
CAMx VBS scheme allocates POA emissions from five source types to the PAP, PCP, and PFP
species based on emission factors (Table 5‐9) determined from laboratory experiments. VBS
uses source‐specific volatility distribution factors for gasoline vehicles (POA_GV), diesel vehicles
(POA_DV), meat cooking (POA_MC), and biomass burning (POA_BB) based on recent chamber
studies (May et al., 2013a,b,c; Woody et al., 2015). For other POA emissions (POA_OP), VBS
applies distribution factors estimated by Robinson et al. (2007).

Table 5‐8. Input species for 1.5‐D VBS scheme.
Species
TOLA
XYLA
BNZA
ISPA
TRPA
TOLB
XYLB
BNZB
ISPB
TRPB
SQT
IVOG
IVOD
IVOA
IVOB
POA_GV
POA_DV
POA_MC
POA_OP
POA_BB

Description
Toluene (anthropogenic)
Xylene (anthropogenic)
Benzene (anthropogenic)
Isoprene (anthropogenic)
Monoterpenes (anthropogenic)
Toluene (biogenic)
Xylene (biogenic)
Benzene (biogenic)
Isoprene (biogenic)
Monoterpenes (biogenic)
Sesquiterpenes (biogenic)
IVOC from gasoline engines
IVOC from diesel engines
IVOC from other anthropogenic sources
IVOC from biomass burning
POA from gasoline vehicles
POA from diesel vehicles
POA from meat cooking
POA from other anthropogenic sources
POA from biomass burning

Notes

Anthropogenic VOC precursors

Biogenic VOC precursors

IVOC precursors

POA precursor emissions assigned to PAP and
PFP modeled species

Table 5‐9. Volatility distribution factors used to allocate POA emissions from five different
source types to the five PAP, PCP, and PFP volatility bins.
Emission Fraction for volatility bin with C* of
POA species
POA_GV
POA_DV
POA_MC
POA_OP
POA_BB

0
0.27
0.03
0.35
0.09
0.2

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

1
0.15
0.25
0.35
0.09
0.1

10
0.26
0.37
0.1
0.14
0.1

121

100
0.15
0.24
0.1
0.18
0.2

1000
0.17
0.11
0.1
0.5
0.4

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
5. Chemistry Mechanisms

Emissions of intermediate‐volatility organic compounds (IVOCs; 104 ≤ C* ≤ 106) make important
contributions to OA in the atmosphere but generally are missing from emission inventories
because neither VOC nor POA emission factors account for IVOCs. A pre‐processor (PREPVBS)
can be used to map source‐specific POA emissions to the five distinct POA emission species for
VBS, and to scale IVOC emissions from total non‐methane organic compound (NMOC)
emissions based on smog chamber data (Jathar et al., 2014).
5.2.3 Aerosol Sectional Approach
Unlike the CF scheme, where each species is represented by a single particle size, the CMU
scheme institutes an additional step to distribute the bulk aerosol concentrations from the
aqueous/aerosol chemistry modules into each size bin. For inorganic aerosol species,
ISORROPIA yields the bulk aerosol composition at equilibrium. The aerosol size distribution is
then determined by distributing the change in aerosol mass during the time step into each size
bin using a weighting factor (Pandis et al., 1993). The fraction fi,k of total flux of species i
between gas and aerosol phases that condenses onto or evaporates from an aerosol size
section k is given by,

f i ,k 





2N k d k Di ci  cieq  k  1
,
 2N k d k Di ci  cieq  k  1





k

where Nk and dk are the number and mean diameter of particles in the section k, respectively,
Di, ci, and cieq are the diffusivity, bulk gas‐phase concentration, and equilibrium concentration at
the particle surface of species i, respectively, k=2/adk,  is the mean free path of air, and a is
the accommodation coefficient (Pandis et al., 1993). Assuming that cieq is independent of
particle size, the fraction is reduced to,
f i ,k 

N k d k  k  1
 fk .
 N k d k  k  1
k

The above weighting factor then depends on the surface area only.
For organic aerosols, SOAP calculates the bulk equilibrium composition. Using the pseudo‐ideal
solution assumption (Strader et al., 1999), the effect of chemical composition of the particle can
be incorporated into the weighting factor:
f i ,k 



   1 .
c    1

N k d k c i  xi ,k c i*

N

k



d k ci  xi ,k

k

*
i

k

k

where xi,k is the mole fraction of species i in the section k and ci* is the effective saturation
concentration of species i. Since the fraction determines the composition of each size section,
the above equation should be solved iteratively at each time step. Assuming that the chemical
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

122

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
5. Chemistry Mechanisms

composition changes slowly during a time step, however, the mole fractions can be
approximated with those from the previous time step (Koo et al., 2003).
For cloud/fog droplets, RADM is used to calculate sulfate and nitrate formation in the bulk
aqueous phase. The added mass is then distributed into each size bin by a weighting factor
which is based on the size‐resolved aqueous chemistry model simulation results (Fahey and
Pandis, 2001).
In addition, mathematical descriptions of nucleation and coagulation have been added to the
CMU scheme. The nucleation model employs the nucleation rate parameterization proposed
by Russell et al. (1994). The model assumes a linear sulfuric acid vapor concentration variation
for the given time step of the aerosol module based on the initially available sulfuric acid and
assigns all the nucleated mass to the first section of the distribution. The coagulation rate of
the aerosol particles is modeled according to Seinfeld and Pandis (1998). A high‐resolution
distribution is used for the coagulation calculations by subdividing each section of the original
distribution into 3 additional sections.

5.3 Mercury Chemistry
Mercury exists in the atmosphere as elemental mercury, Hg(0), and oxidized mercury, Hg(II)
(Schroeder and Munthe, 1998). Hg(II) can be inorganic (e.g., mercuric chloride, HgCl2) or
organic (e.g., methyl mercury, MeHg). It can also be present as particulate mercury (e.g.,
mercuric oxide, HgO, or mercury sulfide, HgS). In the global atmosphere, Hg(0) is the dominant
form. Hg(II) typically constitutes a few percent of total mercury and is predominantly in the gas
phase. MeHg concentrations in the atmosphere are negligible, about a factor of 10 to 30 lower
than Hg(II) concentrations, based on analysis of precipitation samples conducted by Frontier
Geosciences, Inc. (e.g., Seigneur et al., 1998). However, Hg(II) becomes methylated in water
bodies, where it can bioaccumulate in the food chain. Hg(0) is sparingly soluble and is not
removed significantly by wet deposition; its dry deposition velocity is also believed to be low.
As a result, Hg(0) has a long atmospheric lifetime, on the order of several months, that is
governed by its oxidation to Hg(II). On the other hand, Hg(II) is quite soluble; it is consequently
removed rapidly by wet and dry deposition processes. Particulate mercury, Hg(p), is mostly
present in the fine fraction of particulate matter (PM2.5), although some Hg(p) may be present
in coarse PM (e.g., Landis and Keeler, 2002).
Known transformations among inorganic mercury species include the gas‐phase oxidation of
Hg(0) to Hg(II), the aqueous‐phase oxidation of Hg(0) to Hg(II), the aqueous‐phase reduction of
Hg(II) to Hg(0), various aqueous‐phase equilibria of Hg(II) species, and the adsorption of Hg(II)
to PM in both the gas‐phase and aqueous‐phase. The inorganic mercury chemistry modules
implemented in CAMx are based on our current knowledge of these transformations. However,
it should be noted that our knowledge of mercury chemistry continues to evolve as new
laboratory data become available, and the Hg chemical kinetic mechanisms in CAMx and other
models that treat the atmospheric fate of mercury will need to be revised accordingly.
Below, we provide additional details on the gas‐ and aqueous‐phase mercury chemistry
mechanisms implemented in CAMx, and the implementation approach.
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

123

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
5. Chemistry Mechanisms

5.3.1 Gas‐Phase Chemistry
The gas‐phase transformations include the oxidation of Hg(0) to Hg(II) by ozone (O3) (Hall,
1995), hydrogen peroxide (H2O2) (Tokos et al., 1998) hydroxyl radicals (OH) (Sommar et al.,
2001; Pal and Ariya, 2003; 2004), bromine (Br) (Ariya et al., 2002), and hypobromite (BrO)
(Raofie and Ariya, 2003). The oxidation of Hg(0) by O3, H2O2, and OH are given by the following
three reactions:

k = 3  10-20 cm3 molec-1s -1

Hg(0) (g) + O3 (g)  Hg(II)(g),

Hg(0) (g) + H 2 O 2 (g)  Hg(OH)2 (g),
Hg(0) (g) + OH (g)  Hg(OH)2 (g),

k = 8.5 10-19 cm3 molec-1s -1
k = 8  10-14 cm3 molec-1s -1 ,

while oxidation by Bromine is based on a sequence of 5 reactions (Seigneur and Lohman, 2008):

Hg(0) (g) + Br (g)  HgBr (g),

k1 = 3.6  10

-13

 T 
P

 298 

  8537
HgBr(g)  Hg(0)(g), k2 = 3.9109 exp

 T 
HgBr (g) + Br (g)  HgBr2 (g),

k3 = 2.5 10

-1.86

cm 3 molec-1s -1

s-1

-10  T 

-0.57



 298 

cm3 molec-1s -1

-0.57

HgBr (g) + OH (g)  HgBrOH (g),

 T 
k 4 = 2.5 10-10 

 298 

Hg(0) (g) + BrO (g)  Hg(II) (g),

k 5= 1 10 -15 cm 3 molec -1s -1

cm3 molec-1s -1

The reaction rate constants provided above are for temperatures in the range of 20 to 25oC; no
temperature dependence information is available. For the bromine reactions, T is the
temperature in degrees Kelvin, and P is the pressure in atmospheres. The five reactions are
treated as a single reaction, with an effective Hg(0) first‐order rate constant that is a function of
the individual reaction rates and the concentrations of Br, BrO and OH based on the assumption
that Br, BrO and OH concentrations don’t change by their reactions with mercury. This
treatment is similar to that of Holmes et al. (2006), who considered the oxidation of Hg(0) by
bromine atoms with a set of three reactions. The effective first‐order rate constant is calculated
by the following expression:
k Br k3 Br   k 4 OH 
k eff  1
 k5 BrO 
k 2  k3 Br   k 4 OH 

s -1

5.3.2 Adsorption of Hg(II) on PM
In the first implementation of mercury in CAMx, Hg(II) adsorption on PM was considered only in
the aqueous phase (see below), using an adsorption coefficient derived from available
experimental data (Seigneur et al., 1998; Ryaboshapko et al., 2002). It is essential to also
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS
www.camx.com
124

March 2016

CAMx User’s Guide Version 6.3
5. Chemistry Mechanisms

consider the adsorption of gaseous Hg species to PM because gas/particle conversion also
affects Hg deposition (Lindberg et al., 2007). Rutter and Schauer (2007a) reported results of
laboratory work measuring the adsorption of reactive gaseous mercury (RGM) to atmospheric
and synthetic particles as a function of temperature. Their experimental results suggest that
surface area rather than PM mass controls the partitioning process. They reported three
surface‐area adsorption coefficients for urban PM (collected in Milwaukee, WI, and Riverside,
CA), ammonium sulfate and adipic acid, respectively. They expressed the surface‐area
adsorption coefficient (Ksa) as follows:



K sa  Hg p, ads RGM  Asp  PM



where Ksa is in m3m‐2, Hgp,ads and RGM are in pg m‐3, Asp is the specific surface area of ambient
PM in m2μg‐1 and PM is the ambient urban PM concentration in μg m‐3. Here, Hgp,ads refers only
to the adsorbed RGM, i.e., it does not include non‐volatile primary particulate mercury. Rutter
and Schauer (2007a) also found that the Ksa obtained for urban PM falls between that of
ammonium sulfate (more RGM adsorption) and adipic acid (less RGM adsorption). Their
laboratory experiments lead to the following value for Ksa as a function of temperature (in K) for
adsorption to urban PM:
K sa  10 4250 / T 10 

(1)

Rutter and Schauer’s (2007b) experimental results also show a ten‐fold increase in adsorption
of RGM to sodium chloride compared to ammonium sulfate and organic particulate compounds
(a larger increase was observed for sodium nitrate). Thus, the adsorption coefficient for RGM
adsorption to sea‐salt is about 10 times that for urban PM:
K sa  10 4250 / T  9 

(2)

Following the approach used in Vijayaraghavan et al. (2008), we treat all non‐sea‐salt PM as
urban PM for simplicity and use Equation (1) to simulate RGM adsorption to all non‐sea‐salt
PM. The adsorption to sea‐salt PM is calculated using Equation (2). Thus, the effective
adsorption coefficient for each aerosol size section is calculated as:
K sa , eff  10 4250 / T  9  Fss  10 4250 / T 10  1  Fss 

where Fss is the fraction of sea‐salt in that size section. In the CAMx implementation, we
assume that RGM is adsorbed on primary fine and coarse PM.
5.3.3 Aqueous‐Phase Chemistry
The aqueous‐phase chemistry includes the reduction of Hg(II) to Hg(0) via reaction with
hydroperoxy radicals (HO2) and by the formation of the sulfite complexes (at low HCl
concentrations), HgSO3 and Hg(SO3)22‐, as well as the oxidation of Hg(0) to Hg(II) by dissolved
O3, OH, and Cl2. Adsorption of Hg(II) species on atmospheric particulate matter (PM) is
simulated using an adsorption coefficient (K = 34 L g‐1) recommended by Seigneur et al. (1998).
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

125

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
5. Chemistry Mechanisms

The relevant reactions are listed below. Note that the gas‐liquid equilibria and ionic equilibria
of the non‐mercury species (e.g., SO2, O3) involved in the mercury aqueous‐phase chemistry are
not shown here, since they are identical to those in the other CAMx mechanisms.
5.3.2.1 Gas‐liquid Equilibria

Hg(0) (g)  Hg(0) (aq), K = 0.11 M atm-1 (Sanemasa, 1975; Clever et al., 1985)
HgCl2 (g)  HgCl2 (aq), K = 1.4  106 M atm -1 (Lindqvist and Rodhe, 1985)
Hg(OH) 2 (g)  Hg(OH) 2 (aq), K = 1.2  10 4 M atm -1 (Lindqvist and Rodhe, 1985)
The Henry's Law constants listed above are for temperatures in the range of 20 to 25oC.
Temperature dependence information is available for the Hg(0) Henry's Law constant but the
validity of this information for temperatures below 0oC is not established.
5.3.2.2 Aqueous‐phase Equilibria

HgCl 2 (aq)  Hg 2 + + 2Cl - ,

K = 10 -14 M 2 (Sillen and Martel, 1964)

Hg(OH)2 (aq)  Hg 2 + + 2OH - ,
Hg 2 + + SO 32 -  HgSO 3 ,

K = 10-22 M 2 (Sillen and Martel, 1964)

K = 2.1  1013 M -1 (van Loon et al., 2001)

HgSO 3 + SO 32 -  Hg(SO 3 ) 22 - ,

K = 1  1010 M -1 (van Loon et al., 2001)

5.3.2.3 Adsorption of Hg(II) on PM in the Aqueous Phase

Hg(II) (aq)  H(II) (p), K = 34 L g -1 (Seigneur et al., 1998)
5.3.2.4 Aqueous‐phase Kinetics
Hg(0) (aq) + O 3 (aq)  Hg 2 + ,

k = 4.7  10 7 M -1s -1 (Munthe, 1992)

Hg(0) (aq) + OH (aq)  Hg 2 + , k = 2  109 M -1s -1 (Lin and Pehkonen, 1997)
HgSO 3 (aq)  Hg(0) (aq),

k = 0.0106 s -1 (van Loon et al., 2000)

Hg(II) (aq) + HO 2 (aq)  Hg(0) (aq), k = 1.7 10 4 M -1s -1 (Pehkonen and Lin, 1998)
Hg(0) (aq) + HOCl (aq)  Hg 2 + , k = 2.09  106 M -1s -1 (Lin and Pehkonen, 1998)
Hg(0) (aq) + OCl -  Hg 2 + , k = 1.99 10 6 M -1s -1 (Lin and Pehkonen, 1998)
In the last two reactions listed above, HOCl and OCl‐ come from the dissolution and subsequent
dissociation of molecular chlorine (Cl2). Note that Hg(II) (aq) refers to all divalent Hg species in
solution (i.e., Hg2+ + HgCl2(aq) + Hg(OH)2(aq) + HgSO3 + Hg(SO3)22‐). The rate constants listed for
the aqueous‐phase kinetics are for temperatures in the range of 20 to 25oC. Temperature
dependence information is available for the HgSO3 reduction reaction.
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

126

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
5. Chemistry Mechanisms

As mentioned previously, the gas‐ and aqueous‐phase Hg transformations presented above
represent the state of the science from a decade ago (Ryaboshapko et al., 2002; Seigneur et al.,
2001a, 2004) and our knowledge of mercury chemistry continues to evolve. For example,
Gardfeldt and Johnson (2003) challenged the aqueous‐phase reduction of Hg(II) to Hg(0) by
dissolved HO2, suggesting that this pathway is unimportant. There also seems to be some
circumstantial evidence of reduction of Hg(II) to Hg(0) in power plant plumes from various
experimental studies that is not accounted for in current treatments of Hg chemistry (e.g.,
Edgerton et al., 2001; Seigneur et al., 2001b). Additional details are provided in a scoping study
for mercury deposition conducted for the Midwest Regional Planning Organization by Seigneur
et al. (2003).
5.3.4 Implementation Approach
The approach used to implement the mercury transformation pathways, discussed above, into
CAMx is based on the assumption that the mercury species concentrations are much smaller
than those of the species with which they react. Thus, the concentrations of the non‐mercury
species can be assumed to be constant during the mercury chemistry calculations and analytical
solutions are available for both the gas‐phase and aqueous‐phase conversions.
The mercury chemistry discussed in the previous sections requires the concentrations of the
following non‐mercury species: O3, H2O2, OH, SO2, HO2, Cl2, HCl, Br, BrO and atmospheric
particulate matter (PM). The concentrations of most of these species are available from CAMx.
However, the halogen compounds Cl2, Br, and BrO are only included for one specific gas‐phase
mechanism (CB6r2h, Mechanism 3) and otherwise not explicitly simulated. Since the mercury
chemistry is currently not linked to any halogens that might be available from the gas‐phase
chemistry, we specify typical vertical profiles of Cl2, Br, and BrO concentrations. The Cl2
concentrations are prescribed to be non‐zero over oceans and zero elsewhere. Also, daytime
Cl2 concentrations are lower than nighttime values to account for the fact that Cl2 is photolyzed
during the day. The zenith angle is used for the determination of night/day. A 2‐D array of
integer values (1 if ocean, 0 if not) is used to determine if the grid column is predominantly over
ocean. This array is initialized at the beginning of the simulation from an input file and is
specific for the modeling domain and grid. For Br and BrO, vertical profiles over land and ocean
are prescribed, with higher values over ocean than over land. During the night, the
concentrations of these species are assumed to be zero, since the photolysis of Br2 is the
primary source for these radicals.
The mercury aqueous‐phase chemistry module also requires the specification of cloud liquid
water content (LWC) and cloudwater pH. Both these variables are available from CAMx – the
mercury aqueous‐phase chemistry module is invoked after the CAMx PM aqueous‐phase
chemistry calculations are performed, so the cloudwater pH has already been calculated. Note
that the PM aqueous‐phase module (based on the RADM aqueous‐phase chemistry module
that is also used in Models‐3/CMAQ) does not explicitly simulate the cloud chemistry of OH and
HO2 radicals. The concentrations of these radicals can be reduced by their heterogeneous
chemistry within clouds (e.g., Jacob, 2000; Jaegle et al., 2001). In the CAMx implementation,
we account for this by reducing the concentrations of OH and HO2 radicals by factors of 2 and
10, respectively.
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

127

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
5. Chemistry Mechanisms

5.3.5 Chemistry Parameters for Mercury
The mercury chemistry module requires total PM concentrations, so mercury can only be
modeled in conjunction with PM chemistry. Mercury chemistry is selected by including
mercury species among the list of modeled species. The CAMx mercury species names are:
HG0 – elemental gaseous mercury, or Hg(0)
HG2 – reactive gaseous mercury, or Hg(II)
HGP – primary particulate mercury, or Hg(P)
HGIIP – reactive gaseous mercury, or Hg(II) adsorbed onto fine PM
HGIIPC – reactive gaseous mercury, or Hg(II) adsorbed onto coarse PM
CAMx requires that all five or none of these species be included in a simulation. Therefore,
mercury chemistry is not required for PM modeling, but if mercury chemistry is selected then
all five mercury species must be modeled. All of the rate constants and equilibrium constants
for the mercury chemistry module are hard‐coded and so no mercury reaction rate data are
included in the chemistry parameters input file (see Section 3). This is similar to the RADM
aqueous chemistry and ISORROPIA inorganic aerosol equilibrium modules.
Several physical properties of the mercury species must be specified on the chemistry
parameters file (see Figure 3‐3a). The physical properties specified for the gas species (Henry’s
Law, molecular weight, surface reactivity) influence the deposition characteristics. The Henry
constant for HG2 is assumed to be similar to that of HNO3 because these two gases have similar
solubility. The HG2 species represents HgCl2 and Hg(OH)2; the Henry constant for the former is
1.4106 M atm‐1 and for the latter it is 1.2104 M atm‐1. The surface resistance factor is set to
zero for strong acids, such as HNO3, that have a strong tendency to stick to surfaces – this
forces the surface resistance calculated in the dry deposition algorithms to zero. The reactivity
parameter for HG2 is set to 0, as for HNO3.
The dry deposition of HG0 is set to zero by choosing a very low Henry constant (similar to CO).
This is based on the assumption that natural emissions and dry deposition of HG0 balance each
other over the modeling domain. This assumption is justified by the fact that the atmospheric
lifetime of HG0 (about 1 year) greatly exceeds its residence time (days to weeks) within a
regional modeling domain. If natural emissions of HG0 are not included in the mercury
emissions inventory, the dry and wet deposition of HG0 should be zero by setting a Henry
constant of smaller than 110‐8 M atm‐1. However, if natural emissions of Hg(0) are used in the
CAMx simulation, the Henry constant should be set to 0.111 with a temperature factor of ‐4970
K (Clever et al., 1985).

5.4 Simple Chemistry Via Mechanism 10
The chemical mechanisms in CAMx require significant effort to prepare emissions data and can
result in extensive run times. There are many cases when air pollution problems could be
investigated with a much simpler chemistry scheme. An example of this would include
modeling SO2 from a few specific sources over a relatively small region, and treating conversion
to sulfate by assuming a representative decay rate. CAMx provides an option to configure a
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

128

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
5. Chemistry Mechanisms

simple and flexible scheme that can be used for modeling chemical reactions other than the
ozone or secondary PM reactions treated by the full‐science mechanisms.
The simple chemistry scheme is selected by specifying 10 for the mechanism ID in the chemistry
parameters input file. The user must develop specific chemical reactions and code them into
the subroutine CAMx/chem10.f; an example subroutine is available in the source code
directory. Follow the guidelines in that subroutine to implement your specific set of reactions.
This approach requires some knowledge from the user, but also provides complete flexibility.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

129

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
6. Plume‐in‐Grid Submodel

6. PLUME‐IN‐GRID SUBMODEL
Photochemistry is a highly non‐linear problem because chemical reaction rates among most
compounds depend upon their ambient concentrations. In Eulerian air quality models, ambient
concentrations depend on how well the modeling grid resolves emissions, transport, and
chemical history. Thus, grid resolution plays a vital role in the ability of the model to properly
characterize photochemical conditions. Increasing resolution should in theory lead to a better
model as the time/space discretization tends toward a continuum. However, practical and
theoretical considerations suggest that the lower limit on horizontal grid spacing is about 1000
meters for Eulerian air quality models such as CAMx. Nevertheless, even higher resolution is
often necessary to adequately simulate chemistry within concentrated point source plumes.
Plume‐in‐grid treatments have been developed to track individual plume segments (or puffs) in
a Lagrangian sense, accounting for plume‐scale dispersion and chemical evolution until such
time as puff mass can be adequately represented within the larger grid model framework.
Then the puff mass is added to the grid system as a virtual source, and that mass is
subsequently carried by the grid model processes. It is important to understand that the
inclusion of a Lagrangian puff model within an Eulerian grid model is a forced construct. The
formulations among the various modeling approaches are fundamentally different and there is
no theoretically “correct” methodology.
The CAMx Plume‐in‐Grid (PiG) sub‐model addresses the size and chemical evolution of point
source plumes. Two PiG options are available that vary in their chemical complexity. Both
approaches share common design features for puff initialization, puff structure, transport, and
growth. They deviate in how they treat chemistry and when they transfer mass from puffs to
grid cells. This section details the structure and functionality of both options.
1) GREASD PiG: The Greatly Reduced Execution and Simplified Dynamics PiG option is
aimed at treating the early chemical evolution of large NOx plumes when mostly
inorganic gas‐phase reactions are operative. GREASD PiG works with OSAT/PSAT
because of the simplified approach employed and because compatibility with source
apportionment was an explicit design objective.
2) IRON PiG: The the Incremental Reactions for Organics and NOx PiG option treats the
full suite of gas‐phase photochemistry for all types of point sources. Gas‐phase
chemistry is simulated within each plume segment using an “incremental chemistry”
approach (EPRI, 2000), whereby puffs carry the incremental contributions of the puff
relative to the grid concentrations. IRON PiG supports the Reactive Tracer (RTRAC)
Probing Tool, but it does not work with other Probing Tools and it does not treat PM.

6.1 CAMx PiG Formulation
6.1.1 Basic Puff Structure and Diffusive Growth
Both GREASD and IRON PiG sub‐models share a common physical structure and growth
algorithm. A stream of plume segments (puffs) is released from a designated point source.
Each puff possesses a longitudinal length and directional orientation defined by the separation
of a leading and a trailing point. Initial separation of these points is determined by the wind
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

130

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
6. Plume‐in‐Grid Submodel

vector at final plume rise. Each point is then subsequently and independently transported
through the gridded wind fields, which directly accounts for puff stretching and changes to
centerline orientation due to deforming shears. The "position" of each puff is defined at the
center point of each puff between the endpoints, and is used to identify the grid cell in which
the puff resides for the calculation of diffusive growth and chemistry.
Like other puff models, the shape of each puff is characterized by a spread tensor, which is
defined from a set of Gaussian standard deviations () along the three spatial axes (x, y, z).
Diffusive growth is described by the evolution of these values. The total cross‐sectional width
extends ±1.5y from puff centerline. Similarly, the total cross‐sectional depth extends ±1.5z
from puff centerline (with limits placed on depth by the ground and by capping stable layers
aloft). The total longitudinal length is the distance between the puff endpoints with an
additional ±1.5x added in each direction. Horizontal area and total volume are calculated
using the formulae for an ellipse. Figure 6‐1 presents a schematic representation of each puff
in horizontal cross‐section.

1.5

P (t

y

1.5

x

)
ead
P (l

)
rail

Figure 6‐1. Schematic representation of CAMx PiG puff shape in the horizontal plane.
Directional orientation of the puff is arbitrary, and evolves according to wind direction,
shears and diffusive growth along its trajectory.

PiG puff growth is based on SCICHEM theory and concepts (EPRI, 2000), but includes some
simplifications. SCICHEM solves predictive spatial moment equations with second‐order
closure that relate the evolution of the puff spread tensor (ij=ij) to resolved mean shears
and turbulent velocity statistics. The Reynolds‐averaged second‐moment transport equation is
given as
d ij
dt

  ik

u j
x k

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

  jk

xi u j c 
x j u ic 
u i


Q
Q
x k

131

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
6. Plume‐in‐Grid Submodel

where u is the mean wind vector component, the primed values represent turbulent
fluctuations from the mean, and the angle brackets denote integrals over space. The Reynolds
averaging process always introduces higher‐order fluctuation correlations, and this is given by
the turbulent flux moments x u c  , where u  c  represents the turbulent flux of
concentration. It is these last two diffusion terms that SCICHEM solves in its second‐order
closure scheme.
For sub‐puff scale turbulence, SCICHEM employs the restriction that the only active off‐diagonal
component of x u c  is the symmetric horizontal term (i=x, j=y), but it is applied only for the
large‐scale (meso to synoptic) contribution to puff deformation when puff scales reach such
dimensions. In CAMx, we ignore this off‐diagonal flux moment term altogether since puffs are
ultimately terminated when puff scales approach much smaller grid scales (typically < 50 km).
SCICHEM also makes the assumption that the horizontal turbulence is isotropic,
x u c   y v c  . This results in a single diffusivity equation for both x and y directions, and a
single diffusivity for the z direction:
y vc

Kx  K y 

Kz 

Q

zwc
Q

The SCICHEM second‐order tendency equations are used to model the time‐evolution of PiG
puff turbulent flux moments (represented by diffusivities Kx=Ky and Kz) and their contribution to
the evolution of puff spread (represented by the diagonal components of the puff spread
tensor, x2, y2 and z2). Puff spread is defined for puff depth (z), puff width (y), and puff
length (x). We account for the effects of grid‐resolved shears of horizontal wind in the
evolution of lateral and longitudinal spread. But we assume that the evolution of vertical
spread is solely the result of turbulent fluxes, which are orders of magnitude larger than grid‐
resolved shears of vertical wind.
The resulting Reynolds‐averaged second‐moment transport equations for CAMx PiG are:

d z2
dt
d y2
dt
d x2
dt

 2K z

 2 2y D  2 y z S y

 2K x

 2 x2 D  2 x z S x

 2K x

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

132

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
6. Plume‐in‐Grid Submodel

where D is total explicit (grid‐resolved) horizontal shear of horizontal wind (i.e., “deformation”,
see Section 4.4). S is the explicit vertical shear of horizontal wind, which is broken down into
puff‐relative lateral (y) and longitudinal (x) components:
1/ 2

Sx



 du 2 dv 2 

cos    

dz 
 dz



 du 2 dv 2 

sin     

dz 
 dz

1/ 2

Sy

The difference between  and  represents the relative angle between the puff’s longitudinal
orientation and the direction of vertical shear, respectively; a cross‐puff shear results in more
lateral spread while along‐puff shear results in more longitudinal spread. The explicit shear
terms containing D and S may be toggled by the user: (1) shear effects are always applied to
puff growth rates; (2) shear effects are applied only within the boundary layer but never above;
or (3) shear effects are never applied.
The SCICHEM tendency equation for the horizontal turbulent flux moment is

d
y v c 
dt

 Qq 2  A

q
y v c 


where A = 0.75, q2 = vv , and  is the horizontal turbulent length scale. Separate equations
are given for two different boundary layer turbulence scales (shear‐ and buoyancy‐produced),
such that
y v c 

y v c 



y v c 



shear

buoyancy

Within the surface‐based boundary layer, the horizontal velocity variance is given by
2
q buoyancy

 0.13 w*2 1  1.5 exp z / z i 

 2.5 u*2 1  z / zi 

2
qshear

where u* is the friction velocity, w* is the convective velocity scale, z is height above the
surface, and zi is the height of the surface‐based boundary layer. The horizontal turbulent
length scale is given by
1





2
shear

0.3 z i 

 buoyancy

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

1

2



1

0.65 z 2

 0.3 z i

133

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
6. Plume‐in‐Grid Submodel

In the stable boundary layer, only the shear components of q2 and  are applied. Above the
boundary layer, SCICHEM applies rough approximations for the velocity variance and turbulent
length scale. In CAMx we have set these values to q2 = 0.1 m2/s2, and  = 300 m.
The SCICHEM tendency equation for the vertical turbulent flux moment is
d
z  w c 
dt



A



qv
QK zeq  z wc 
v



where qv2 = ww , v is the vertical turbulent length scale, and Kzeq is the equilibrium diffusivity.
Whereas a specific equation for Kzeq is formulated for SCICHEM, we have chosen to specify the
value of this parameter from the gridded fields of vertical diffusivity in CAMx. Again SCICHEM
gives separate equations for shear‐ and buoyancy‐produced turbulence scales.
Within the surface‐based boundary layer, the vertical velocity variance is given by
q v2

q v2

buoyancy

shear

 1.5 u *2 1  z / z i 

 1.1 w*2  z / z i 

2/3

1.05  z / z i 

The vertical turbulent length scale for both shear and buoyancy is equal to shear given above
for horizontal length scale. Above the boundary layer, SCICHEM applies rough approximations
for the velocity variance and turbulent length scale and we have adopted these for CAMx: qv2 =
0.01 m2/s2, and v = 10 m.
The external variables needed by PiG to complete the dispersion calculations include zi, u* and
w*. All of these are available from an internal module in CAMx that calculates these boundary
layer similarity theory parameters. Thus, no additional parameters are needed to be input to
the model.
6.1.2 Puff Transport
A fresh set of new puffs are released from all PiG point sources within the modeling domain for
the duration of the smallest time step among the master and all nested grids. The length of
each puff is determined by the combination of the mean total wind speed at the height of final
plume rise and time step. Limits are placed on maximum extruded length based on half the
finest resolution in the given simulation. If winds and time‐steps are such that the maximum
allowed length is violated, then several puffs are automatically emitted from a given stack per
time step. The user can also set a maximum time interval of release if more puffs (better plume
resolution) are desired over the default automated release interval. The orientation of the puff
length is along the total wind vector. Total puff volume is determined by stack volumetric flow
rate in conjunction with growth due to turbulent entrainment following the SCICHEM approach.
Initially, x = y and z values are explicitly calculated from this entrainment calculation.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

134

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
6. Plume‐in‐Grid Submodel

The effects of resolved wind shear on plume deformation (i.e., at plume scales larger than
individual puff scales) are treated using a “chained puff” approach (Figure 6‐2). Points at the
leading and trailing edges of the puff centerline are individually transported through the
gridded wind fields, which directly accounts for puff stretching and changes to centerline
orientation due to deforming shears. Since puffs can extend over multiple layers, layer density‐
weighted average wind components are determined for each endpoint based on the vertical
coverage of the puff. The “chain” aspect means that at least initially (as puffs are emitted from
the stack) the trailing point of a puff emitted at time t will be the leading point of a puff emitted
at time t+dt. However, as the puffs are advected downstream, the leading point of one puff will
deviate from the trailing point the puff behind it due to evolving puff depth and wind fields.
The “position” of each puff is defined by its center point between the endpoints. This position
defines the grid domain and grid cell in which the puff resides for the calculation of diffusive
growth and chemistry. This definition holds even if the puff is sufficiently long that the
endpoints are in different grid cells (or even different nested grids if near a nest boundary).
With our definition for termination when horizontal area approaches grid cell area, the puff
length should not extend across more than two grid cells.

Stack

Figure 6‐2. Plan‐view schematic representation of a chain of PiG puffs emitted from a point
source into an evolving gridded wind field. The red line connected by dots represents puff
centerlines, with dots representing leading and trailing points of each puff. The CAMx
computational grid is denoted by the blue lines.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

135

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
6. Plume‐in‐Grid Submodel

6.2 GREASD PiG
The GREASD PiG is designed for large NOx point sources, where only inorganic chemistry is
operative during early plume evolution. The intention of GREASD PiG is to more properly age
emitted NOx within the confined plume volume to mitigate the artificially rapid chemical
processing of fresh NO to NO2 to ozone that would otherwise occur if immediately released into
a large grid volume. A condensed inorganic chemical mechanism is employed in GREASD PiG
that includes 23 reactions involving oxidation of NOx and SOx. Therefore, GREASD PiG should
not be used for VOC sources. Chemical limitations applied within GREASD PiG cause puffs to
transfer their mass to the grid before oxidant production from VOCs is no longer suppressed;
this can occur before puffs reach a size threshold determined by the grid spacing.
The chemical evolution of large NOx point source plumes can be characterized in three stages
(EPRI, 2000), as described below:



Stage 1: Early plume conditions where NOx concentrations are very high and radical
concentrations are negligible – simple NO/NO2/O3 photostationary state applies along
with the NO‐NO self reaction;



Stage 2: Mid‐range plume conditions where radical concentrations are sufficiently large
to generate secondary inorganic acids like nitrate and sulfate;



Stage 3: Long‐range plume conditions where sufficient mixing with the ambient air leads
to the full range of gas‐phase reactions involving VOC oxidation and ozone formation.

The objective for GREASD PiG is to transfer mass to the grid at about the time when radical
production via organic chemistry starts to become important, so GREASD PiG treats plume
chemistry during Stages 1 and 2. We define the onset of Stage 3 chemistry when the following
criterion is met:

kOH CO   kOH SO2 
 1
kOH NO2 
At this point GREASD puffs transfer all of their mass to the grid before the onset of Stage 3.
This specific design constraint is also compatible with the requirements of the source
apportionment Probing Tools.
Kumar and Russell (1996) and Karamchandani et al. (1998) found that PiG models with
simplified inorganic chemistry produced results that were very similar to full chemistry. The
chemical mechanism for GREASD PiG includes 23 reactions listed in Table 6‐1 that were
selected as follows:







Reactions for the NO‐NO2‐O3 photo‐stationary state established in sunlight (1‐3)
Self‐reaction of NO important only at very high NO concentrations (4)
Production of OH by photolysis of O3 and HONO in sunlight (5‐9)
Production of nitric acid in sunlight (10)
Formation of NO3 and N2O5 at night (11‐17)

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

136

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
6. Plume‐in‐Grid Submodel

Table 6‐1. List of 23 reactions for GREASD PiG including correspondence to CAMx reaction
numbers in the CB05, CB6 and SAPRC07TC mechanisms.
Chemical Mechanism for GREASD PIG
Number

Reaction

1

Corresponding Reaction
Number for Grid Chemistry
CB05

CB6

S07

NO2 = NO + O

1

1

1

2

O + O2 + M = O3 + M

2

2

2

3

O3 + NO = NO2

3

3

7

4

NO + NO + O2 = 2 NO2

22

24

10

51

NO + NO2 + H2O = 2 HONO

23

41

N/A

6

O3 = O1D

9

9

18

7

O1D + M = O + M

10

10

21

8

O1D + H2O = 2 OH

11

11

20

9

HONO = NO + OH

25

43

23

10

NO2 + OH = HNO3

28

45

25

11

NO2 + O3 = NO3

7

26

8

12

NO3 = NO2 + O

14

27

17

13

NO3 = NO

15

28

16

14

NO3 + NO = 2 NO2

16

29

9

15

NO3 + NO2 = NO + NO2

17

30

15

16

NO3 + NO2 = N2O5

18

36

11

N2O5 = NO3 + NO2

21

37

12

N2O5 + H2O = 2 HNO3

19

39

13

19

SO2 + OH = SULF + HO2

63

52

44

20

OH + CO = HO2

66

123

29

21

FORM = 2 HO2 + CO

75

97

204

22

FORM = CO

76

98

205

23

HO2 + NO = OH + NO2

30

25

31

17
18

2

1. Rate for GREASD PiG reaction 5 is set to zero when used with SAPRC07TC.
2. Rate for GREASD PiG reaction 18 may be enhanced by reaction on aerosol.







Production of nitric acid at night (18)
Production of sulfuric acid in sunlight (19)
Removal of OH by CO (20)
Production of OH by photolysis of formaldehyde (21‐22)
Conversion to OH of any HO2 formed in 20‐22 (23)

These reactions dominate gas‐phase chemistry in plumes from major NOx emitters during
stages 1 and 2. Table 6‐1 also shows the correspondence between GREASD PiG reactions and
the complete gas‐phase chemical mechanisms implemented in CAMx. This mapping is used in
CAMx to set the rate constants and photolysis rates for GREASD PiG reactions from
corresponding reactions in the grid chemical mechanisms. This implementation ensures
consistency with these chemical mechanisms. The GREASD PiG performs gas‐phase chemistry
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

137

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
6. Plume‐in‐Grid Submodel

for only those chemical species emitted directly into the plume, and does not include the
influence from any background compounds on the grid except for ozone, carbon monoxide, and
formaldehyde. Assuming zero background for gas species is reasonable for the early stages of
NOx plumes because puff concentrations are orders of magnitude larger than ambient
concentrations. On the other hand, background ozone, carbon monoxide, and formaldehyde
are the primary sources of oxidants in the condensed mechanism that drive inorganic
processing of plume NOx to other forms of oxidized nitrogen (NOz), so these are handled as
“incremental” species as described below for IRON PiG. The Livermore Solver for Ordinary
Differential Equations (LSODE) is used to solve the condensed mechanism in double‐precision.

6.3 Particulate Matter In PiG
Non‐linear effects and errors in gas‐phase chemistry are transmitted to (and potentially
amplified by) the PM aqueous chemistry and partitioning algorithms. The implementation of
PM and PSAT in PiG prompted the need to limit the impact of such issues similarly to the
constraints imposed by OSAT for gas‐phase chemistry. Therefore, PM and PSAT can only be run
using the GREASD PiG option. As for gas‐phase chemistry, only inorganic PM chemistry is
treated by GREASD PiG. Note also that the PiG PM treatment was designed specifically for the
static 2‐mode PM chemistry only (CF), and will not operate for the multi‐section PM chemistry
(CMU).
The GREASD PiG gas‐phase chemistry oxidizes NOx and SOx emissions to nitric and sulfuric
acids, which are PM precursors. To maintain consistency with the grid chemistry, aqueous PM
chemistry (RADM‐AQ) is then performed at every time step if the puff resides in a cloudy grid
cell. However, inorganic gas/PM partitioning (performed by ISORROPIA) among sulfate, nitrate
and ammonia is not performed by GREASD PiG, but is delayed until the masses of these
compounds are dumped to the grid. Additionally, chemistry and partitioning for secondary
organic aerosols is not performed.

6.4 IRON PiG
The IRON PiG model incorporates a complete treatment of gas‐phase chemistry in point source
pollutant plumes, while secondarily adding additional features central for treating toxic
pollutants not normally carried by the standard CAMx chemical mechanisms. Therefore the
IRON PiG can treat a wide variety of point source emissions, including VOC sources.
IRON PiG adopts the “incremental chemistry” concept from the SCICHEM model (EPRI, 2000),
whereby each puff carries concentrations relative to ambient background. This results in the
possibility of both positive and negative puff concentrations depending on how the chemistry
evolves. The full gas‐phase chemistry mechanism chosen for a given run is solved twice for
each puff at each time step: first for the vertically‐averaged background concentrations from
the grid cells vertically spanned by the puff; and second for the sum of puff and background
concentrations. The LSODE solver is used to solve both chemistry steps. After both steps are
completed, the updated background concentrations are subtracted from the updated
puff+background concentrations, yielding the new puff incremental concentrations. Note that
the updated background concentrations are used for reference only in the puff incremental
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

138

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
6. Plume‐in‐Grid Submodel

chemistry algorithm; the actual grid concentrations are not affected and are separately solved
by the grid chemistry routine.

6.5 PiG Features
This section describes specific features of the PiG submodel; certain features are always active
while others can be optionally invoked for a particular CAMx run. The IRON PiG sub‐model
includes two constructs designed specifically to facilitate the incremental chemistry approach:



The concept of “virtual dumping” to handle the chemical impacts of large puffs that can
overlap other puffs within a given grid column;



The concept of multiple puff “reactor” cells to account for the chemical effects of
concentration distributions within each puff.

Each of these is described below.
6.5.1 Puff Layer Spanning (IRON and GREASD)
The PiG is designed to chemically process point source plume mass within streams of puffs until
such time that each puff can be adequately resolved on the horizontal grid. Puffs are allowed
to vertically span multiple grid model layers before they reach horizontal grid scales. This
introduces technical implications for defining “background” concentrations and ambient
conditions for puff chemistry, as well as for transferring plume incremental mass to the grid.
The solution employed is to:
1) Assume that the vertical distribution of puff concentration is always uniform;
2) Distribute puff mass transfer (via “leaking” and “dumping”) to the grid according to the
puff fractional coverage across each model layer and by density‐weighting; and
3) Determine mean background concentrations and other ambient conditions (e.g.,
temperature, humidity, etc.) over the puff vertical span via similar fractional layer‐density
weighting.
Horizontally, the mean background concentration and ambient conditions are taken from the
single host grid column containing each puff center point, even if the puff is large and/or spans
a horizontal cell interface. As described earlier, puffs are considered to be elliptical in the
horizontal, with the minor axis spanning the cross‐wind puff width (defined as 1.5y), and the
major axis spanning the along‐wind puff length (defined as length 1.5y on each end).
However, given the complications associated with multiple layers spanning and mass‐weighting
of ambient inputs and dumped mass, puffs are rectangular and uniform in the vertical, with
total puff depth defined as 1.5z.
6.5.2 Puff Overlap and the Idea of Virtual Dumping (IRON only)
The chemical effects of puff overlap were considered to be an important attribute of IRON PiG.
However, we wished to maintain a relatively simple approach, while appropriately accounting
for the largest probable effects. We assume that the largest effect is the condition in which
older expansive puffs span significant fractions of their host grid cell, thereby potentially
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

139

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
6. Plume‐in‐Grid Submodel

overlapping all other puffs contained within the same cell. Instead of geometrically
determining fractional overlap puff‐by‐puff, we instead introduce a process that we refer to as
“virtual dumping.”
For a given grid column, the mass from all puffs determined to be “sufficiently” large are
temporarily dumped to the column, distributed according to each puff’s vertical layer span, and
added together along with the pre‐existing grid concentrations. This process is referred to as a
“virtual dump” of puff mass to the grid column. The criteria to determine a “sufficiently” large
puff is the same that initiates puff mass leaking to the grid (as described below). In this
approach, all large puffs contribute to the background chemistry step for other puffs occupying
the same grid column. Double‐counting is avoided by not including a puff’s contribution to the
background while its specific background and incremental chemical calculations are performed.
6.5.3 Multiple Puff Reactors (IRON only)
Accounting for full chemistry potentially introduces significant non‐linear effects that are highly
dependent upon the distribution of pollutant mass within each puff. Especially for ozone,
aircraft flights through power plant plumes have often recorded wide concentration variations
relative to ambient conditions: within the plume core where NOx remains concentrated, ozone
is often depressed and follows NO‐NO2‐ozone equilibrium, whereas on plume fringes where
NOx is dilute and mixes with ambient VOC, ozone concentrations can exhibit concentration
maxima. Past models have accounted for cross‐plume chemistry variations through the use of
reactors, with approaches ranging from multiple rectangular slabs to concentric shells.
The user may select multiple reactors as well to sub‐divide the puff. Any number of reactors
may be chosen (the default is 1). Multiple reactors simply divide the total puff volume evenly,
and the initial mass assignments for newly emitted puffs are made using the standard error
function that results in an initial Gaussian‐like mass/concentration distribution among the
reactors. This provides a mechanism for simulating the differing chemical processing that takes
place in various concentration regimes. As the purpose of the reactors is merely to represent
the range of photochemical conditions that are likely to occur at various locations within the
puff as it undergoes differential shearing and mixing, there is no particular physical orientation
assigned to these reactors with respect to each other or to the puff as a whole. Thus, there is
no communication (i.e., diffusional mass exchange) between the reactors. The same
background concentration chemistry applies to all reactors of a given puff. When puff mass is
leaked or dumped, all reactors shed the same relative fraction of mass.
In summary, chemistry is solved for each puff “reactor” in three steps:
1) The layer‐mean background (grid + overlapping puff) concentrations and environmental
conditions over the volume occupied by the puff are stored and then chemically updated
via the LSODE gas‐phase chemistry mechanism;
2) The pre‐updated mean background concentrations are added to the puff increments and
the total concentrations are chemically updated; and
3) The updated results from step 1 are subtracted from the updated results of step 2 to
provide the updated incremental concentrations.
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

140

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
6. Plume‐in‐Grid Submodel

6.5.4 Puff Dumping (IRON and GREASD)
Mass transfer from puff to grid can happen in two ways: slowly, termed “leaking”, or suddenly,
termed “dumping.” As described earlier, all mass is transferred to the vertical grid structure in
a density‐weighted fashion according to each puff’s fractional layer coverage. The process of
leaking ensures that puff mass is transferred to the grid continuously, rather than in discrete
lumps of pollutants with very different concentrations than those in the grid. Sudden dumping
can cause unphysical numerical shocks in the grid and can lead to unrealistic gridded
concentration patterns that appear as “bulls‐eyes”. The idea behind puff leakage is to account
for turbulent shearing of mass from the main plume and its subsequent dispersion to the grid
scale. This rate of transfer should be directly proportional to the puff size relative to the grid
scale.
Puff leakage is controlled by comparing the horizontal area of a puff to a specified leakage
parameter, defined as a fraction of horizontal grid cell area. When a puff is first emitted there
is no leakage. As the puff grows in volume the concentrations within the reactors are reduced
accordingly by dilution. When the puff area exceeds the leakage onset parameter, a fraction of
the mass in each puff reactor is transferred to the grid. This fraction is determined by the
relative exceedance of the leakage parameter; initial leakage is slow as the exceedance is
relatively small, but leakage rates grow as the puff continues to grow beyond the leakage
parameter.
The reduced mass from leakage is compensated by a reduced effective volume, so that
concentrations are not artificially diluted by leakage (an essential chemical imperative). Thus,
two distinct volumes are tracked: the actual volume (defined by the puff spread ) and the
effective volume. While these are identical before leakage, they deviate after leakage is
initiated, and thereafter the relative deformation of the actual puff volume (via diffusion,
shearing, etc.) is used to scale the deformation of effective puff volume.
Eventually the horizontal span of the puff will exceed the grid cell area, and the remaining mass
is then dumped all at once to the grid. However, because of the combination of photochemical
processing and leakage, by the time a puff dumps the potential for producing numerical shocks
is much reduced. Furthermore, if the puff exceeds a user‐defined maximum age, puff mass is
transferred to the grid at the rate of 10% per timestep.
6.5.5 PiG Rendering (IRON and GREASD)
While the mass confined to the puffs at any given time has not yet affected the grid
concentrations, it will eventually, so it can be somewhat misleading to sequester this mass from
visualizations of a model simulation. The puff mass can be optionally incorporated into the
model average output files for visualization purposes (referred to as “PiG rendering”).
Rendering employs a “virtual dump” of the puff masses into the average output concentration
array each time step. As described for chemistry, virtual puff mass is added as an increment
over the entire grid column according to fractional layer‐density weighting over puff depth,
thus diluting it’s concentrations relative to that within the puff. The actual puff mass remains
within each puff over the course of its lifetime, and the actual grid mass in unaffected until
puffs are killed and their masses truly dumped into the grid. This visualization is available for
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

141

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
6. Plume‐in‐Grid Submodel

either 2‐D surface or 3‐D average output files, and can produce some rather startling effects in
output displays, including very narrow virtual plumes, or streaks, representing mass moving
through the grid in sub‐grid puffs, but not subject to grid‐scale eddy diffusion.
6.5.6 High Resolution Puff Sampling (IRON and GREASD)
PiG optionally employs surface‐layer puff sampling on a user‐defined grid of arbitrary horizontal
resolution, similarly to the way nested grids are defined. Sampling grids are entirely passive,
and intended to provide a display of the plume concentrations at scales much smaller than
typically used for the finest computational grids (i.e., <1 km), primarily around and downwind
of a source complex. Sampled PiG concentrations are time‐averaged like the output
concentrations provided on the computational grids, and are written to files with similar format
so that they may be readily viewed and manipulated with CAMx post‐processing software.
Additional information on configuring and using PiG sampling grids is provided in Sections 2 and
4.
Given that the puffs constantly evolve via diffusive growth and reshaping due to deforming
shears, the sampling procedure includes trigonometric calculations to define which sampling
points are influenced by each puff. This influence is determined according to the puffs’ two‐
dimensional horizontal Gaussian shape shown in Figure 6‐1. To include a sufficiently large
percentage of mass across each puff for sampling, limits of ±3x/y in both horizontal dimensions
are used to define the puffs’ total elliptical area coverage. Puffs are only sampled if they extend
vertically within 10 m of the ground.
Sampling grids are defined in the CAMx control file (see Section 2), and array dimensions must
be set sufficiently large in the CAMx Fortran parameters file in ./Includes/camx.prm (see
Section 2). An example of the type of plume detail that can be visualized using a sampling grid
is provided in Figure 6‐3. In this case, a very fine 200 m sampling grid is set within a 4‐km
computational grid.

6.6 Deposition
The CAMx PiG treats the removal of gas and PM species from each puff via deposition
processes. Both dry and wet deposition calculations presented unique implementation issues
for puffs. The most difficult issue for both forms of deposition was how to manage deposition
exchange between puffs and the ground in the case of negative puff concentration increments.
6.6.1 Dry Deposition
Dry deposition needs to consider the following: (1) the point at which puffs begin to deposit to
the surface; (2) how to handle deposition through potentially deep puffs that may straddle
several layers of varying stability since the puffs do not themselves resolve these stratifications
or vertical concentration distributions; (3) managing deposition fluxes of negative
concentration increments. Our solution to issue (1) was to ignore dry deposition within puffs
until they diffusively grow to the ground, although in reality deposition occurs on roughness
elements that extend some distance above the ground (trees, buildings, etc.). We
implemented a criterion that the bottom of the puff must extend to or below the midpoint of
the surface layer, or below 10 m (whichever is larger), in order for dry deposition to be active.
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS
www.camx.com
142

March 2016

CAMx User’s Guide Version 6.3
6. Plume‐in‐Grid Submodel

Figure 6‐3. Example of a single point source PiG plume as depicted by a sampling grid with
200 m resolution (shown by the extent of the plot; 40 km by 32 km total extent). This
sampling grid was set within a CAMx computational grid with 4‐km resolution. The source
location is arbitrary and is emitting an inert tracer.

Issue (2) can be handled in a variety of ways and complexity. The current implementation
institutes a simpler solution and we will consider more complicated improvements for future
developments if evidence suggests that they would be necessary. PiG utilizes pre‐computed
species‐dependent deposition velocities derived for the grids. Each puff in a particular grid cell
is provided the host cell’s deposition velocities for each species, and these are used to
determine the flux of mass through the fraction of puff depth occupying the model’s surface
layer.
Issue (3) is unique to the incremental chemistry concept introduced with IRON PiG. The flux of
material depositing to the ground is given by F  c  vd , where by the normal definition a
positive deposition velocity vd leads to a positive deposition flux to the ground. If the puff
increment c is negative, then a negative flux is calculated (flux from ground to puff). This is
appropriate if we consider the following argument. Dry deposition applied to a grid cell
removes some pollutant mass from the entire volume. If there is a puff existing in that cell with
a negative concentration increment, then the amount of mass removed from the cell was over
estimated if we consider the puff’s contribution to total cell mass. The negative deposition flux
calculated for this puff leads to the addition of mass to the puff increment. Adding mass to a
negative increment reduces the magnitude of the increment, as expected for a deposition
process. This mass is taken from the grid cell’s accumulated deposited mass to maintain
accurate mass accounting within the model.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

143

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
6. Plume‐in‐Grid Submodel

6.6.2 Wet Deposition
Wet deposition needs to consider the following: (1) how to handle scavenging of pollutants
through potentially deep puffs that may straddle several layers of varying cloud and rain water
contents but that do not themselves resolve vertical concentration distributions; (2) managing
deposition fluxes of negative concentration increments in combination with the potential for
mass to move in and out of rainwater as it falls (e.g., for slightly soluble gasses); (3) accounting
for the initial pollutant concentrations in rainwater as they enter the top of each puff.
It was important to maintain consistency between the treatment of wet deposition and the
approach for puff chemistry. The chemistry relies on the assumption of vertically well‐mixed
puff reactors that can span multiple layers, and this is why layer‐density weighted average
ambient conditions are passed to the chemistry routines. To maintain this assumption for wet
deposition, a single scavenging rate is applied through the entire puff depth as effectively a
single layer of pollutant. This was found to be the simplest implementation approach. This
single scavenging rate is calculated according to layer‐density weighted average ambient cloud
and rainwater contents.
Wet scavenging is performed throughout the entire depth of the puff to determine the amount
of flux in or out of rainwater. Total concentrations (puff + background) are used to determine
species‐dependent scavenging rates using the identical algorithm as for grid removal. The rates
are used to derive removal fractions, and these fractions are then applied directly to the puff
incremental mass for each species. Removal fractions are considered positive for the standard
case of mass moving from puff to rain.
We further assume that the top boundary condition for rainwater entering the top of each puff
is zero. This means that the removal fraction is always positive (from puff to rain) in the single‐
layer puff. In contrast, for gridded concentrations the layer‐by‐layer buildup of slightly soluble
species can lead to a reversal of fluxes (from rain to grid) if super saturation is diagnosed in a
particular layer.
Note that negative puff mass increments in combination with a positive removal fraction lead
to a reversal of the flux direction (rain to puff), but that is not allowed and in such cases wet
scavenging is set to zero. We account for impacts on the mass budget appropriately by adding
to the wet deposition mass array according to the net fluxes into rainwater.

6.7 PiG Configuration
Selecting the individual elevated point sources to receive the PiG treatment is accomplished by
setting their stack diameters negative within the header (time‐invariant point list) section of the
CAMx input point source file. CAMx will run correctly with these negative diameters even if the
PiG algorithm is not invoked. CAMx preprocessors exist to ease the procedure of ranking
elevated point sources by emission rate and flagging the sources that the user wishes to treat.
Invoking the CAMx PiG sub‐model is controlled by keywords in the CAMx control file
(CAMx.in), as described in Section 2. The choices are:
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

144

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
6. Plume‐in‐Grid Submodel

PiG_Submodel
PiG_Submodel
PiG_Submodel

= 'NONE',
= 'GREASD',
= 'IRON',

Note that the single choice between GREASD and IRON applies to all flagged point sources. It is
not possible to make a single CAMx run with IRON PiG applied to a certain group of sources,
and GREASD PiG applied to another group. Also note that GREASD must be selected to run PiG
with OSAT, PM and PSAT; IRON must be selected to run PiG with the RTRAC Probing Tool.
Several additional parameters are used to configure the PiG. It is important to note that all PiG
configuration parameters exist in the CAMx Fortran parameters file
(./Includes/camx.prm), as described in Section 2. PiG parameters are grouped together
and briefly described at the end of that file. By configuring the PiG submodel in the code, the
default PiG configuration (as recommended by the model developers) is preset within the
model distribution and alleviates the need for users to select settings on their own.
The default values are shown below:
parameter
parameter
parameter
parameter
parameter
parameter
parameter
parameter
parameter
parameter

(
(
(
(
(
(
(
(
(
(

MXPIG
MXRECTR
FLEAK
LEAKON
LVISPIG
OVERLAP
DXYMAX
AGEMAX
PIGMXDT
SHRFLG

=
=
=
=
=
=
=
=
=
=

50000 )
1 )
.25 )
.FALSE. )
.FALSE. )
.FALSE. )
-10000. )
18.*3600. )
300. )
1 )

Users should exercise thoughtful consideration when altering these default values. A
description of each of the remaining parameters is provided below, along with guidance in
setting values.
6.7.1 Guidance on the Use of CAMx PiG
6.7.1.1 PiG Keyword
The PiG keyword controls whether the PiG option is to be invoked in a CAMx simulation, and
whether the emissions are treated with the GREASD or IRON options. This keyword can be
switched from NONE to GREASD or IRON on a model restart to invoke the PiG treatment at any
point during a multi‐day simulation. To allow for this, it is not mandatory to provide CAMx with
a pre‐existing PiG output file upon a model restart – CAMx will not stop if this file is missing. It
is recommended that this file be provided on all subsequent restarts since the PiG output file is
needed to reinitialize the PiG module, otherwise all mass contained in puffs at the end of the
previous run will be lost. If the PiG keyword is switched to NONE on a model restart, CAMx will
continue the simulation without PiG, but all mass contained in puffs at the end of the previous
run will be lost.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

145

www.camx.com

March 2016



CAMx User’s Guide Version 6.3
6. Plume‐in‐Grid Submodel

Guidance:
- Invoke GREASD or IRON PiG at any point during a multi‐day simulation, or none at all.
Once PiG is started, provide CAMx with the PiG output file from the previous run for
all subsequent model restarts so that no point source mass is lost.
- GREASD PiG should be invoked for large NOx point sources only, since it does not
provide any organic chemistry. GREASD PiG supports PM chemistry (CF but not CMU).
It can be run in conjunction with the OSAT/PSAT. It does not support DDM, PA, or
RTRAC.
- IRON PiG can be invoked for any point source to treat gas‐phase chemical evolution
using any of the CAMx photochemical mechanisms. IRON PiG does not treat
particulate chemistry. It can be run in conjunction with the RTRAC Probing Tool. It
does not support OSAT/PSAT, DDM, or PA.

Both GREASD and IRON options use the LSODE chemistry solver exclusively, so users will notice
an impact on run time, particularly if many (thousands) puffs are to be tracked, and IRON PiG is
invoked (2 solutions of full photochemistry for each puff), and IRON puffs are configured with
many puff reactor cells (full photochemistry solutions each). Since GREASD chemistry is simpler
and the lifetime of GREASD puffs are much shorter than their IRON counterparts, GREASD PiG
will run faster than IRON PiG for the same number of flagged sources. PiG chemistry is
internally parallelized using OMP to maximize PiG speed performance.
6.7.1.2 Number of PiG Puffs
MXPIG sets the maximum number of PiG puffs to be expected during a simulation. It is used to
statically allocate memory arrays for the PiG sub‐model. A value of 10,000 is usually sufficient
for most applications in which PiG is used; set this parameter to 1 if PiG is not used to conserve
memory. If this parameter is exceeded during a simulation, the model will halt. If this happens,
simply increase MXPIG, recompile the model executable, and restart the simulation.



Guidance: Use the default value for most simulations, or set to 1 if PiG is not to be used.
If the model stops because MXPIG is exceeded, increase its value, recompile, and restart
the model.

6.7.1.3 Number of PiG Reactors (IRON only)
MXRECTR sets the number of puff reactors; when greater than 1, each puff is separated into
that number of reactor cells and primary emissions are apportioned among them using a
Gaussian distribution. Since chemistry is performed for each individual reactor cell (both
background and puff+background), this parameter can affect the speed of chemical
computations in the PiG. We have not seen a significant sensitivity to values greater than 1, but
testing for each application is warranted.



Guidance: Use the default of 1 for initial simulations, but test the sensitivity to this
parameter for each unique application.



Reactors greater than 1 are not allowed for GREASD PiG.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

146

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
6. Plume‐in‐Grid Submodel

6.7.1.4 Leakage Parameters
FLEAK, LEAKON, and DXYMAX together control puff leakage to the computational grid and
ultimately puff termination. When LEAKON is true, FLEAK and DXYMAX control when puffs
begin to leak portions of reactor mass to the grid along their trajectory. When LEAKON is false,
no leaking is performed and puffs maintain all of their mass until they reach sizes for
termination, at which point all mass is directly introduced to the grid at that point. DXYMAX
sets the maximum dimension that puff size will be compared to for leaking and termination:
when it is zero, puff size will be compared to grid area only; when it is positive, puff size will be
compared to the value of DXYMAX regardless of grid resolution; when it is negative, puff size
will be compared to DXYMAX or grid resolution, whichever is smaller. FLEAK is the relative
fraction of horizontal puff area to cell area (or DXYMAX) above which leaking will begin and
continue until sufficient mass is shed and the puff is terminated. In the example above, puffs
will begin to leak mass when they reach 25% of the host grid cell’s area.



Guidance: If LEAKON is set to true, maintain FLEAK at the default value of 0.25. Then test
model sensitivity to different values of FLEAK and/or DXYMAX.



Guidance: We suggest leaving DXYMAX = ‐10000, meaning puffs will be terminated when
they reach the grid scale or 10 km, whichever is smaller. Puffs exceeding this size are
usually well‐aged and go beyond reasonable assumptions of puff coherence (also see
AGEMAX parameter below).



Leaking is not allowed when PiG is run with PM.

6.7.1.5 Overlap Flag (IRON only)
OVERLAP controls whether puff overlap is to be treated in the background chemistry step. As
stated earlier, puffs only overlap if they meet the size criteria for leaking; all puffs smaller than
this size do not overlap any other puffs in the same grid cell.



Guidance: We recommend that the OVERLAP flag remain set to the default value of
“false”.



Overlap is not allowed for GREASD PiG.

6.7.1.6 Virtual Puff Rendering
LVISPIG is a flag that turns on puff “rendering” to the computational grid average
concentrations. When it is false, the chemical effects of puff mass are not seen on the output
average files until they either begin to leak mass to the grid and/or they are terminated and
their mass is entirely introduced to the computation grid. However, when the flag is true, all
puff mass that resides in each grid column is summed, apportioned vertically to each grid cell
according to puff vertical extent (via density and layer‐depth weighting), converted to
concentrations, and added to the average gridded concentrations for output. This process is
referred to as rendering since the effects of all puff mass can be readily visualized in the CAMx
output.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

147

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
6. Plume‐in‐Grid Submodel



Guidance: This option has no impact on the actual CAMx chemical solution. However,
output average concentration files will be affected by puff rendering, and therefore could
impact graphics of CAMx results and model performance measures.



Virtual dumps are not allowed when PiG is run with PM.

6.7.1.7 Maximum Puff Age
AGEMAX is the age limit for all PiG puffs (IRON and GREASD). When puffs reach this age limit,
they are terminated and all of their mass is transferred to the host grid. The assumption of a
stream of coherent puffs becomes less valid with time as horizontal and vertical wind shears
increase plume spread. At some point the plume mass is better resolved on the grid than
within puffs. The maximum puff age provides a safety check to ensure that puffs do not persist
for unrealistic times in stable environments. The maximum puff age should be set long enough
to allow puffs to persist overnight, but a lifetime of longer than a day is probably not realistic.



Guidance: limit puff age to 12‐24 hours – we find that 18 hours works best since it will
allow puffs emitted in the late afternoon to last through the night and into the following
morning. Twelve hours is seen to be too short in this regard; puffs usually do not reach 24
hours of age before being terminated by grid constraints.

6.7.1.8 Maximum Puff Release Interval
PIGMXDT sets the maximum frequency of release and by default is set to 300 seconds (5
minutes). This value should be adequate for most applications. However, if the user wishes to
improve plume resolution by increasing the number of puffs, the frequency of release can be
increased by reducing the value of PIGMXDT. This value supersedes the automated puff release
rate that is determined by wind speed and grid size.



Guidance: Maintain the default value of 300 s and allow PiG to use the automated PiG
release frequency. Set to a lower value if better plume resolution is desired; note that
more puffs will be released and this could slow the model markedly.

6.7.1.9 Effects of Wind Shear on Puff Growth Rates
SHRFLG sets the approach by which to apply the effects of explicitly resolved (grid scale) wind
shear on puff growth rates. There are three options available to the user:
0 = shear is never applied;
1 = shear is applied only within the boundary layer;
2 = shear is always applied.



Guidance: The application of wind shear can lead to large growth rates, especially above
the boundary layer where stability squelches turbulent growth, and this may over‐dilute
puff concentrations, lead to early transfer of puff mass to the grid, and have markedly
reduced impacts downwind. Shear has less relative impact on growth rates in neutral/
unstable conditions because turbulent growth on its own leads to rapid plume dilution.
For these reasons, the default is to ignore the effects of shear when puffs are above the
boundary layer.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

148

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
7. Source Apportionment

7. SOURCE APPORTIONMENT
Photochemical grid models are often used to develop emission reduction strategies to attain air
quality objectives. Traditional methods involve running numerous iterative reduction or “zero‐
out” simulations (i.e., “brute force” methods) to identify the contributions from specific
pollutants, source categories and source regions. This process quickly becomes impractical, but
the lack of such information might lead to implicating sources that contribute little to high
pollutant levels or, conversely, not properly identifying sources that do contribute.
CAMx includes a source apportionment (SA) or attribution capability that estimates the
contributions from multiple source areas, categories, and pollutant types to the spatial and
temporal distribution of ozone and PM in a single model run. The main challenges in
implementing a methodology to track the relationships between separate groups of precursor
emissions and subsequent non‐linear formation of target pollutants include:



Accounting not only for the presence of precursors from a given source region at a given
receptor location, but also accurately estimating their cumulative contribution to target
pollutants while they were en‐route to the receptor;



Ensuring compatibility with the underlying air quality model formulation so that derived
source‐receptor relationships are consistent with model results for total concentrations;



Providing sufficient spatial and temporal resolution while managing, within practical
constraints, the computer resources required to run the source apportionment tool.

SA uses sets of tracer species to track the fate of precursor emissions and the ozone and PM
compounds formed from these emissions. The tracers operate as “spectators” to the normal
CAMx calculations so that the underlying relationships between total emissions and
concentrations are not perturbed. SA tracers are not “passive”: rather they track the effects of
chemical reaction, transport, diffusion, emissions and deposition within a CAMx simulation and
are thus referred to as “reaction tracers.” A source can be defined in terms of geographical
area (or region) and/or emission category (or group). Figure 7‐1 provides an example of the
way a CAMx domain can be sub‐divided into multiple source areas – 40 in this example. Also,
the emission inventory could be sub‐divided into several source categories; for example, three
emission categories (mobile, industrial, biogenic) over 40 source regions would produce 120
separate sets of tracers. All sources of precursors, ozone, PM must be accounted for, so CAMx
intial and boundary conditions are also tracked as separate source groups. The methodology is
designed such that all ozone, PM and precursor concentrations are attributed among the
selected source regions/groups at all times and throughout all grids. The methodology also
estimates the fractions of ozone formed en‐route under VOC‐ or NOX‐limited conditions,
indicating whether ozone at a particular time and locations will respond to reductions in VOC or
NOX precursor emissions.
An important feature of the reaction tracer approach is that the normal CAMx calculations are
not perturbed; thus, SA estimates the same total ozone, PM and precursor concentrations as
CAMx. Further, since the same inputs are used for meteorology, emissions etc., and the same
numerical methods are employed throughout the model, the source‐receptor relationships
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

149

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
7. Source Apportionment

45

40

18

31

37

14

17

34

22

24
10
3

21
26

8

40
11

9

23

12

4

35

32

35

15

36

20

27

16
29

28

25

2
30
19

1

38

7

13

33

30
6
39

-95

-90

-85

-80

-75

-70

Figure 7‐1. Example of the sub‐division of a CAMx domain into separate areas for geographic
source apportionment.

developed by SA inherently have a high degree of consistency with those generated by CAMx.
The biggest limitation of this (or any other) source apportionment approach relates to non‐
linear chemical interactions between emissions from different sources, which by extension
means that any perturbation to the emission inventory changes source‐receptor relationships
and attribution in a non‐linear way. Thus, for pollutants like ozone and some PM, SA results
only apply to a particular emissions scenario, and cannot be used to extrapolate effects
resulting from emission changes among the tracked source regions/groups.

7.1 Ozone Source Apportionment
Yarwood et al (1996a,b) developed an ozone source attribution approach that has become
known as the “Ozone Source Apportionment Technology” (OSAT). This method was originally
implemented in the Urban Airshed Model (UAM) and was built into the first version of CAMx.
The second version (OSAT2 – although this term was not widely used) was released with CAMx
v4.20 in 2005 along with the addition of Particulate Source Apportionment Technology (PSAT).
The OSAT2 update accounted for simultaneous production and destruction of ozone by
photochemistry and tended to allocate less ozone to long‐range transport (because of
destruction during transport) and more to local production. The third version (OSAT3) was
released with CAMx v6.30 in 2016, and includes an improved approach to handle NOx recycling
(Yarwood and Koo, 2015). The OSAT3 update tends to allocate more ozone to long‐range
transport (due to contributions from NOx during downwind transport) and less to local
production.
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

150

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
7. Source Apportionment

7.1.1 OSAT Formulation
The original OSAT uses four tracers per source region/group to account for contributions to
ozone formation. Ozone formation involves both NOx and VOC, and OSAT uses two tracer
families (Ni and Vi) to apportion NOx and VOC by source region/group i. The ozone formation
process is controlled by the relative availability of NOx and VOC, and so ozone formation is
described either as NOx‐limited or VOC‐limited, respectively. The ratio of the production rates
of hydrogen peroxide (H2O2) and nitric acid (HNO3) is the indicator used to classify ozone
formation as being instantaneously limited by NOx or VOC. Ozone formation is classified as
being NOx‐limited when P(H2O2)/P(HNO3) < 0.35 (Sillman, 1995). When ozone production at a
given location and time is NOx‐limited, it makes sense to attribute that production to source
region/groups based on their contributions to the local NOx, and similarly to attribute
production based on VOC contributions when ozone formation is VOC‐limited. Consequently,
separate ozone tracer families (O3Ni and O3Vi) are used to track ozone formed under NOx and
VOC‐limited conditions.
The OSAT tracers by source region/group i are:
Ni
Vi
O3Ni
O3Vi

Nitric oxide (NO) and nitrogen dioxide (NO2) emissions
VOC emissions
Ozone formed under NOx‐limited conditions from Ni
Ozone formed under VOC‐limited conditions from Vi

The original OSAT tracer scheme is illustrated in Figure 7‐2. Net ozone change due to chemistry
(ΔO3) is tracked by the tracer families O3N and O3V. Ozone destruction (ΔO3 < 0) reduces all
O3N and O3V proportionately. Ozone production (ΔO3 > 0) is classified either as NOx‐limited or
VOC‐limited using the indicator H2O2/HNO3 and assigned either to O3N or O3V, respectively,
in proportion to the precursor tracers present, respectively N or V. The precursor tracers N and
V are removed by chemical decay.

ΔH2O2/ΔHNO3

O3N

O3V
V

ΔO3

N

Figure 7‐2. The original OSAT scheme for ozone apportionment. Information flows along
arrows. Changes in core model species are shown in blue, OSAT tracers are in black, the
diamond represents the OSAT algorithm that determines ozone tracer changes.
H2O2/HNO3 is the indicator ratio used to determine NOx‐ or VOC‐limited ozone production.
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

151

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
7. Source Apportionment

7.1.2 OSAT2 Formulation
The original OSAT algorithm allocated the net ozone change (ΔO3) to tracers O3N and/or O3V.
However, ozone production and destruction reactions operate simultaneously and so the net
ozone change is the balance of production and destruction. For example, VOC oxidation can
cause photochemical ozone production at the same time that O3 + VOC reactions directly
consume ozone, and these processes may lead to a net ozone increase or decrease depending
mainly upon availability of NOx and sunlight.
OSAT2 accounts for the following ozone destruction mechanisms:
1)
2)
3)
4)

O3 + VOC reactions since these remove ozone;
O(3P) + VOC reactions since these effectively remove ozone;
O(1D) + H2O reaction since this effectively removes ozone;
HOx + O3 reactions that do not re‐form ozone.

Ozone destruction is calculated as the smaller (i.e., more negative) of the sum of these four
mechanisms or ΔO3. Ozone production is then calculated as the difference between ΔO3 and
the ozone destruction. The O3V and O3N tracers are adjusted first for ozone destruction
(applied to all tracers) and second for ozone production (applied using the OSAT rules).
The amount of ozone destruction is calculated from the time‐integrated rates of the four
chemical processes listed above. It is easy to account for processes 1‐3 since the ozone
destroyed is simply the time‐integral of the reactions involved. Process 4 is less easy to
quantify because ozone can be re‐formed. For example:
O3 + OH → HO2
HO2 + NO → OH + NO2
NO2 + hν → NO + O
O + O2 → O3
However, process 4 is an important ozone destruction mechanism in low NOx (e.g., rural)
environments. Therefore, accounting for process 4 is important to understanding long‐range
ozone transport. The main reaction pathways between ozone and HOx (OH and HO2) are
shown in Figure 7‐3.
The ozone destruction rate due to O3 + HOx reactions is computed from:

O3 Destruction





Rate HO2 term 

Rate O3  HOx   
 Rate HO  NO   Rate HO term  
2
2



The OSAT2 tracers are the same as the original OSAT. The OSAT2 scheme for ozone
apportionment is illustrated in Figure 7‐4. Ozone production and destruction are treated
separately and can occur simultaneously. Ozone destruction (−ΔO3) reduces all O3N and O3V
proportionately. Ozone production (+ΔO3) is classified either as NOx‐limited or VOC‐limited
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

152

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
7. Source Apportionment

OH
HO2

HOx
Radical
Pool

O3
OH

HO2

hV
NO
NO2

HO2,
RO2

Peroxides

Figure 7‐3. Daytime reactions of ozone with HOx (OH and HO2) showing potential for
reformation of ozone or ozone destruction via peroxide formation.

−ΔO3
ΔH2O2/ΔHNO3

O3N

O3V
V

+ΔO3

N

Figure 7‐4. The OSAT2 scheme for ozone apportionment. Information flows along arrows.
Changes in core model species are shown in blue, OSAT tracers are in black, the diamond
represents the OSAT algorithm that determines ozone tracer producton. H2O2/HNO3 is the
indicator ratio used to determine NOx‐ or VOC‐limited ozone production.

using the indicator H2O2/HNO3 and assigned either to O3N or O3V, respectively, in
proportion to the precursor tracers present, respectively N or V. The precursor tracers V and N
are removed by chemical decay.
7.1.3 OSAT3 Formulation
OSAT3 improves the accuracy of the OSAT methods by keeping track of the source(s) of ozone
removed by reaction with NO to form NO2 and subsequently returned as ozone when NO2 is
destroyed by photolysis. Accomplishing this objective requires maintaining source attribution
of odd‐oxygen through the chemical reactions that link ozone, NO and NO2. This is illustrated in
the following chemical reactions where ozone is written as OOO, NO2 is written as ONO, and
the source attributed odd‐oxygen is shown in red:
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS
www.camx.com
153

March 2016

CAMx User’s Guide Version 6.3
7. Source Apportionment

NO + OOO → ONO
ONO + hv → NO + O
O + OO → OOO
Source attribution of the odd‐oxygen content of NO2 is performed by tracer families OON and
OOV that are introduced in OSAT3. Two tracer families are needed in order to keep track of the
source profile of ozone consumed, which was represented by O3V and O3N.
Source attribution of the nitrogen in NO and NO2 must also be performed in order to apply the
apportionment algorithms that track ozone production using O3N and O3V. Accordingly,
OSAT3 simultaneously attributes both the N and odd‐oxygen in NO2 to sources, and the source
signatures of these two apportionments will almost always differ. This is illustrated below,
where chemical source attribution is shown in blue for nitrogen and red for odd‐oxygen:
NO + OOO → ONO
ONO + hν → NO + O
O + OO → OOO
The chemical conversion pathways between oxidized nitrogen species (NOy) in CB6 are
summarized in Figure 7‐5. Arrows show the direction of conversion, which is bi‐directional in
some cases. Other chemical mechanisms have similar NOy conversion pathways to CB6. Also
shown in Figure 7‐5 are the OSAT3 tracer families. Color coding shows the correspondence
between OSAT3 tracer families and the NOy species that they represent (note that the purpose
for color coding in Figure 7‐5 is different from colors used in the chemical reactions above).

CB6

N2O5

OSAT3

NO3

NO

HONO

NIT

NO2

PNA

RGN

HNO3

HN3

PAN

TPN
NTR

RNO3

Figure 7‐5. Correspondence between NOy species in CB6 and tracer families in OSAT3 with
conversions between species/tracers shown by arrows.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

154

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
7. Source Apportionment

Tracking source attribution of nitrogen through all forms of NOy enables OSAT3 to account for
NOx recycling when NOx is converted to another form of NOy (e.g., PAN or HNO3) and later
converted back to NOx. OSAT3 uses the following 10 tracers by source region/group i:
Vi
NITi
RGNi
TPNi
NTRi
HN3i
O3Ni
O3Vi
OONi
OOVi

VOC emissions
Nitric oxide (NO) and nitrous acid (HONO) emissions
Nitrogen dioxide (NO2), nitrate radical (NO3) and dinitrogen pentoxide (N2O5)
Peroxyl acetyl nitrate (PAN), analogues of PAN and peroxy nitric acid (PNA)
Organic nitrates (RNO3)
Gaseous nitric acid (HNO3)
Ozone formed under NOx‐limited conditions from Ni
Ozone formed under VOC‐limited conditions from Vi
Odd‐oxygen in NO2 formed from O3Ni
Odd‐oxygen in NO2 formed from O3Vi

The OSAT3 scheme for ozone apportionment is illustrated in Figure 7‐6. The VOC precursor
tracer family V is unchanged in OSAT3 and removed by chemical decay, while the tracer N is
replaced with NIT. The fate of NOx emissions is tracked by the nitrogen tracer families NIT,
RGN, TPN, NTR and HN3. Ozone production and destruction are treated separately and can
occur simultaneously (as in OSAT2). Ozone production (+ΔO3) is classified either as NOx‐limited
or VOC‐limited using the indicator H2O2/HNO3 and assigned either to O3N or O3V,
respectively, in proportion to the precursor tracers present, respectively NIT or V. Ozone
destruction (−ΔO3) reduces all O3N and O3V proportionately. When ozone destruction results
from reaction with NO to form NO2, the amounts of O3N and O3V removed are transferred to
the respective odd‐oxygen tracers OON and OOV. When NO2 is removed by photolysis to form
ozone, the amounts of OON and OOV removed are transferred to the respective tracers O3N
and O3V.
7.1.4 Alternative Ozone Apportionment Using APCA
An alternative ozone apportionment technique called Anthropogenic Precursor Culpability
Assessment (APCA) differs from OSAT in recognizing that certain emission categories are not
controllable (e.g., biogenic emissions) and that apportioning ozone production to these
categories does not provide information that is relevant to development of control strategies.
To address this, in situations where OSAT would attribute ozone production to non‐controllable
emissions, APCA re‐allocates that ozone production to the controllable precursors that
participated in ozone formation with the non‐controllable precursor. For example, when ozone
formation is due to biogenic VOC and anthropogenic NOx under VOC‐limited conditions (a
situation where OSAT would attribute ozone production to biogenic VOC), APCA attributes
ozone production to the anthropogenic NOx present. Using APCA instead of OSAT results in
more ozone formation attributed to anthropogenic NOx sources and less ozone formation
attributed to biogenic VOC sources.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

155

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
7. Source Apportionment

−ΔO3
ΔH2O2/ΔHNO3

O3N

O3V
V

+ΔO3

NIT

OOV − OON − RGN
HN3

TPN
NTR

Figure 7‐6. The OSAT3 scheme for ozone apportionment. Information flows along arrows.
Changes in core model species are shown in blue, OSAT tracers are in black, the diamond
represent the OSAT algorithms that determine ozone tracer production. H2O2/HNO3 is the
indicator ratio used to determine NOx‐ or VOC‐limited ozone production. RGN apportions
the nitrogen in NO2 whereas OON and OOV apportion the odd‐oxygen in NO2.

The only difference between APCA and OSAT is the algorithm used to allocate ozone production
under VOC or NOx‐limited conditions. The OSAT3 update does not revise the allocation of
ozone production under VOC or NOx‐limited conditions and therefore the APCA algorithm
works with the OSAT3 update.

7.2 Particulate Source Apportionment
Particulate Source Apportionment (PSAT) uses multiple tracer families to track the fate of
primary and secondary PM (Yarwood et al., 2004). PSAT is designed to apportion the following
classes of CAMx PM species (CF mode only):



Sulfate (PSO4)



Particulate nitrate (PNO3)



Ammonium (PNH4)



Secondary organic aerosol (SOA)



Particulate mercury (HgP)



Six categories of primary PM:
- Elemental carbon (PEC)
- Primary organic aerosol (POA)
- Crustal fine (FCRS)

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

156

www.camx.com

March 2016

-

CAMx User’s Guide Version 6.3
7. Source Apportionment

Other fine (FPRM)
Crustal coarse (CCRS)
Other coarse (CPRM)

A single tracer family can apportion primary PM species whereas secondary PM species require
several tracer families to track the relationship between gaseous precursors and the resulting
PM. PNO3 and SOA are the most complex PM categories to apportion because the emitted
precursor gases (NO, VOC) are several steps removed from the resulting PM species (PNO3,
SOA).
The PSAT “reactive tracers” for each type of PM by source region/group i are described below.
PSAT tracer names for particulate species begin with the letter “P.”
Sulfur
SO2i Primary SO2 emissions
PS4i Particulate sulfate from primary emissions plus secondarily formed sulfate
Nitrogen
NITi Nitric oxide (NO) and nitrous acid (HONO)
RGNi Nitrogen dioxide (NO2), nitrate radical (NO3), and dinitrogen pentoxide (N2O5)
TPNi Peroxyl acetyl nitrate (PAN), analogues of PAN and peroxy nitric acid (PNA)
NTRi Organic nitrates (RNO3)
HN3i Nitric acid (HNO3)
PN3i Particulate nitrate from primary emissions plus secondarily formed nitrate
NH3i Ammonia (NH3)
PN4i Particulate ammonium (NH4)
Secondary Organics
AROi Aromatic (benzene, toluene and xylene) secondary organic aerosol precursors
Isoprene secondary organic aerosol precursors
ISPi
TRPi Terpene secondary organic aerosol precursors
SQT Sesquiterpene secondary organic aerosol precursors
CG1i Condensable gases from aromatics (low volatility products)
CG2i Condensable gases from aromatics (high volatility products)
CG3i Condensable gases from isoprene (low volatility products)
CG4i Condensable gases from isoprene (high volatility products)
CG5i Condensable gases from terpenes (low volatility products)
CG6i Condensable gases from terpenes (high volatility products)
CG7i Condensable gases from sesqiterpenes
PO1i Particulate organic aerosol associated with CG1
PO2i Particulate organic aerosol associated with CG2
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

157

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
7. Source Apportionment

PO3i Particulate organic aerosol associated with CG3
PO4i Particulate organic aerosol associated with CG4
PO5i Particulate organic aerosol associated with CG5
PO6i Particulate organic aerosol associated with CG6
PO7i Particulate organic aerosol associated with CG7
POHi Particulate non‐volatile organic aerosol from aromatic precursors
PPAi Anthropogenic organic aerosol polymers (SOPA)
PPBi Biogenic organic aerosol polymers (SOPB)
Mercury
HG0i Elemental Mercury vapor
HG2i Reactive gaseous Mercury vapor
PHGi Particulate Mercury
Primary Particulates
PECi Primary Elemental Carbon
POAi Primary Organic Aerosol
PFCi Fine Crustal PM
PFNi Other Fine Particulate
PCCi Coarse Crustal PM
PCSi Other Coarse Particulate
Both ozone and PNO3 are associated with NOx emissions. The oxidized nitrogen tracer families
for OSAT3 and PSAT are equivalent with the only difference being the additional tracer for
particulate species in PSAT. Therefore, PSAT uses the same OSAT3 tracer family for oxidized
nitrogen.
PSAT includes a total of 40 tracers for each source region/group if applied to all PM types.
Since source apportionment may not always be needed for all species, the PSAT
implementation is flexible and allows source apportionment for any or all of the chemical
classes in each CAMx simulation (i.e. the PSO4, PNO3, PNH4, SOA, HgP and primary PM classes
listed above). For example, source apportionment for sulfate, nitrate and ammonium requires
just 10 tracers per source region/group.
A fundamental assumption in PSAT is that PM should be apportioned to the primary precursor
for each type of PM. For example, PSO4 is apportioned to SOx emissions, PNO3 is apportioned
to NOx emissions, PNH4 is apportioned to NH3 emissions, etc. As a source apportionment
method, PSAT must account for all modeled sources of a PM species. Consider two model
species A and B that are apportioned by reactive tracers ai and bi, respectively. Reactive tracers
must be included for all sources of A and B including emissions, initial and boundary conditions
so that complete source apportionment is obtained, i.e., A = ai and B = bi.
In PSAT, the general approach to modeling change over a model time step t is illustrated for a
chemical reaction AB. The general equation for species destruction is:
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS
www.camx.com
158

March 2016

CAMx User’s Guide Version 6.3
7. Source Apportionment

ai t  t   ai t   A

ai
 ai

Here, the relative apportionment of A is preserved as the total amount changes. This equation
applies to chemical removal of A and also to physical removal of A by processes such as
deposition or transport out of a specific grid cell.
The general equation for species production (e.g., chemical production by the chemical reaction
AB) is:
a
bi t  t   bi t   B i
 ai
Here, production of B inherits the apportionment of the precursor A. The same equation
applies for “production” of B in a specific grid cell due to emissions or transport. For the case
where B increases due to emissions, ai is the apportionment of the emissions inventory. For
the case where B increases due to transport, ai is the apportionment of the upwind grid cell.
In some cases, source category specific weighting factors (wi) must be added to the equation
for species destruction:
wa
a i t  t   a i t   A i i
 wi a i
An example is chemical decay of the aromatic VOC tracers (ARO), which must be weighted by
the average OH rate constant of each AROi. ARO tracers for different source groups have
different average VOC reactivities because the relative amounts of benzene, toluenes and
xylenes differ between source categories.
In some cases, source category specific weighting factors (wi) must be added to the equation
for species production:

bi t  t   bi t   B

wi a i
 wi a i

An example is chemical production of condensable gases (CG1 or CG2) from aromatic VOC
tracers, which must be weighted by aerosol yield weighting factors. The aerosol yield weighting
factors depend upon the relative amounts of benzene, toluenes and xylenes in each source
group.
Several aerosol reactions are treated as equilibria, AB. If A and B reach equilibrium at each
time step, it follows that their source apportionments also reach equilibrium:

ai t  t  

ai t  bi t 

bi t  t  

ai t  bi t 

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

A 

 A B
B 

 A B

159

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
7. Source Apportionment

Examples are the equilibrium between gas phase nitric acid and aerosol nitrate, gas phase
ammonium and aerosol ammonium, and condensable organic gases (CG) and secondary
organic aerosols (SOA).

7.3 Running CAMx With SA
7.3.1 CAMx Control File
Source apportionment is invoked similarly to the other Probing Tools within the CAMx control
file. In the &CAMx_Control namelist module, the variable Probing_Tool must be set to
“SA” if OSAT, APCA or PSAT are to be run. An additional namelist module called
&SA_Control must then be provided in the control file to configure the SA portion of the
model. The additional namelist module is described below. The order of the variables follows
the template available with the source code. Several examples of the SA portion of the CAMx
run control file are shown in Figures 7‐7a‐c.
Description of SA Control in the CAMx Run Control File
&SA_Control

Label for the Probing Tool namelist module that configures the
SA option; it must begin in column 2

&

Flag ending a namelist module; it must be in column 2

SA_Summary_Output

Logical flag used to limit the species written to the tracer
concentration file to a subset of the SA tracers. If set to true,
the output will be restricted to O3N and O3V for OSAT/APCA,
and the following species for PSAT: PS4, PN3, PN4, PO1, PO2,
PO3, PO4, PO5, PO6, PO7, POH, PPA, PPB, PEC, POA, PFC, PFN,
PCC, PCS, HG0, HG2, PHG

SA_Treat_SULFATE_Class

Logical flag to turn on the sulfate class of tracer species

SA_Treat_NITRATE_Class

Logical flag to turn on the nitrate class of tracer species

SA_Treat_SOA_Class

Logical flag to turn on the SOA class of tracer species

SA_Treat_PRIMARY_Class

Logical flag to turn on the primary PM class of tracer species

SA_Treat_MERCURY_Class

Logical flag to turn on the mercury class of tracer species

SA_Treat_OZONE_Class

Logical flag to turn on the ozone class of tracer species (uses
OSAT attribution by default)

SA_Use_APCA

Logical flag to use APCA attribution rather than OSAT
(SA_Treat_OZONE_Class must be set to TRUE)

SA_File_Root

Character root output path/filename

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

160

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
7. Source Apportionment

SA_Master_Sfc_Output

Logical flag for master grid surface output (TRUE=SA
concentration file will be output, FALSE=SA concentration file
will not be output)

SA_Nested_Sfc_Output

Logical flag for nested grid surface output (TRUE=SA
concentration file will be output, FALSE=SA concentration file
will not be output)

SA_Stratify_Boundary

Logical flag to stratify boundary types (TRUE=separate tracer
types will be used for the North, South, East, West and Top
boundaries, FALSE=a single tracer type will be used for all 5
boundaries)

SA_Deposition_Output

Logical flag to output deposited tracer mass to a file
(TRUE=output deposited tracer mass, FALSE=do not generate a
tracer deposition output file)

SA_Number_of_Source_Regions Integer number of source regions for this run. This must be
the same as the number of source areas defined in the
SA_Source_Area_Map file
SA_Number_of_Source_Groups Integer number of emission groups (categories) for this run.
Together with the Use_Leftover_Group flag, this
determines the number of paired gridded and point emission
files that must be supplied (additional details below)
Use_Leftover_Group

Logical flag to define a “leftover” emissions group (TRUE=
calculate a “leftover” emissions group from the difference
between the sum of the emission group files and the regular
CAMx emission files, FALSE=do not calculate a “leftover”
emissions group)

SA_Receptor_Definitions

Character input SA receptor definition path/filename. (This is an
optional file)

SA_Source_Area_Map

Character array (by CAMx grid) input SA original source area
definition path/filename uniquely assigning each grid cell to a
single source region (required for master grid, optional for
nested grids)

SA_Use_Partial_SourceMap

Logical flag for fractional (or partial) source region (or area)
maps (TRUE= use fractional maps, FALSE = use original source
area definition only)

Partial_Source_Area_Map

Character array (by SA emissions group and CAMx grid) input SA
fractional source area definition path/filename assigning each
grid cell to multiple source regions by emission group (optional)

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

161

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
7. Source Apportionment

SA_PT_Override

Logical flag to allow point source override (TRUE = look for and
use the point source override flags in sector‐specific point
source files, FALSE = ignore point source override flags)

SA_Master_Restart

Character input master grid SA restart path/filename (ignored if
Restart=FALSE)

SA_Nested_Restart

Character input nested grid SA restart path/filename (ignored if
Restart=FALSE or Number_of_Grids=1)

SA_Points_Group

Character array (by source group) input SA elevated point
source emissions path/filename (optional, ignored if
Point_Emissions=FALSE)

SA_Emiss_Group_Grid

Character array (by source group, by CAMx grid) input SA
gridded emissions path/filename (optional, ignored if
Gridded_Emissions=FALSE)

Each partial source area map to be used in the run must be listed by source group and grid: e.g.,
Partial_Source_Area_Map(3,2) refers to SA emissions group 3 and grid 2. These map files must
be listed in the same order as the group emission input files (i.e., the map assigned to category
1 must be consistent with the emissions assigned to category 1).
7.3.2 Specifying Emission Groups
SA can apportion ozone, PM and precursor concentrations among several emission categories
(or “groups”). To achieve this, the emissions for each group must be supplied in separate
emission files, both for low level (gridded) emissions for the master and each nested grid, and
for elevated point sources. The additional emission files must be in the CAMx gridded and
point emission file formats, as described in Section 3. If a category does not include point
sources (e.g. biogenics), the point source file name for the group can be left blank. If a category
has no gridded emissions, the gridded file name for the group can be left blank for all grids.
APCA requires at least two emission groups, and the first group must be biogenic emissions.
For example, in the case where emissions are tracked by three groups, three sets of emission
files should be supplied that when summed equal the total emissions in the regular CAMx
emission files supplied to the core model. CAMx also allows for an alternative option: two sets
of files could be supplied and the third group can be calculated from the “leftover” emissions
(i.e., the difference between the regular CAMx emissions and the two specified emission
groups). The leftover option is set according to the input flag “Use_Leftover_Group”. If
the leftover option is selected, the model verifies that the leftover group is not too small to
calculate within the numerical precision of the computer (this also traps cases where the flag
was set in error). If the leftover option is not selected, the model verifies that the total
emissions for the groups supplied are equal to the regular model emissions, i.e., that a leftover
group is not needed. In both cases, if appropriate conditions are not met, the model stops with
a descriptive error message.
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

162

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
7. Source Apportionment

&SA_Control
SA_File_Root

= './OSAT_output/CAMx.OSAT.020603',

SA_Master_Sfc_Output
SA_Nested_Sfc_Output
SA_Summary_Output
SA_Stratify_Boundary
SA_Deposition_Output
SA_Number_of_Source_Regions
SA_Number_of_Source_Groups
Use_Leftover_Group
SA_Treat_SULFATE_Class
SA_Treat_NITRATE_Class
SA_Treat_SOA_Class
SA_Treat_PRIMARY_Class
SA_Treat_MERCURY_Class
SA_Treat_OZONE_Class
SA_Use_APCA

=
=
=
=
=
=
=
=
=
=
=
=
=
=
=

.true.,
.true.,
.true.,
.false.,
.false.,
19,
1,
.false.,
.false.,
.false.,
.false.,
.false.,
.false.,
.true.,
.false.,

SA_Receptor_Definitions = './OSAT_input/receptor.cities ',
SA_Source_Area_Map(1)
= './OSAT_input/OSAT.source.area.map',
SA_Source_Area_Map(2)
= ' ',
SA_Use_Partial_SourceMap = .false.,
Partial_Source_Area_Map(1,1) = ' ',
! Map for SA group 1, grid 1
Partial_Source_Area_Map(1,2) = ' ',
! Map for SA group 1, grid 2
SA_PT_Override
= .false.,
SA_Master_Restart
SA_Nested_Restart

= ' ',
= ' ',

SA_Points_Group(1)

= ' ',

SA_Emiss_Group_Grid(1,1) = ' ',
SA_Emiss_Group_Grid(1,2) = ' ',
&

Figure 7‐7a. An example of SA input records in the CAMx run control file. The options for this
OSAT run are as follows: this is a two‐grid run, master and nested grid surface concentrations
are written to file, a single tracer type is to be used for all boundaries, 19 source regions, and
one emission group (i.e., zero additional emission files and no leftover group). This is the first
day of the simulation (i.e., restart is false), so no OSAT restart files are supplied.
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

163

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
7. Source Apportionment

&SA_Control
SA_File_Root

= './OSAT_output/CAMx.OSAT.020604',

SA_Master_Sfc_Output
SA_Nested_Sfc_Output
SA_Summary_Output
SA_Stratify_Boundary
SA_Deposition_Output
SA_Number_of_Source_Regions
SA_Number_of_Source_Groups
Use_Leftover_Group
SA_Treat_SULFATE_Class
SA_Treat_NITRATE_Class
SA_Treat_SOA_Class
SA_Treat_PRIMARY_Class
SA_Treat_MERCURY_Class
SA_Treat_OZONE_Class
SA_Use_APCA

=
=
=
=
=
=
=
=
=
=
=
=
=
=
=

.true.,
.true.,
.true.,
.false.,
.false.,
19,
3,
.true.,
.false.,
.false.,
.false.,
.false.,
.false.,
.true.,
.false.,

SA_Receptor_Definitions = './OSAT_input/receptor.cities ',
SA_Source_Area_Map(1)
= './OSAT_input/OSAT.source.area.map',
SA_Source_Area_Map(2)
= ' ',
SA_Use_Partial_SourceMap = .false.,
Partial_Source_Area_Map(1,1) = ' ',
! Map for SA group 1, grid 1
Partial_Source_Area_Map(1,2) = ' ',
! Map for SA group 1, grid 2
SA_PT_Override
= .false.,
SA_Master_Restart
SA_Nested_Restart

= './OSAT_output/CAMx.OSAT.020603.sa.inst',
= './OSAT_output/CAMx.OSAT.020603.sa.finst',

SA_Points_Group(1)
SA_Points_Group(2)

= ' ',
= './OSAT_input/utils.020604',

SA_Emiss_Group_Grid(1,1)
SA_Emiss_Group_Grid(1,2)
SA_Emiss_Group_Grid(2,1)
SA_Emiss_Group_Grid(2,2)

=
=
=
=

'./OSAT_input/bio.grd1.020604',
'./OSAT_input/bio.grd2.020604',
'./OSAT_input/util.grd1.020604',
'./OSAT_input/util.grd2.020604',

&

Figure 7‐7b. As in Figure 7‐7a, but in this case the run is a continuation day of a run with
three emission groups. The three emission groups are defined by supplying extra emission
files for point and area sources for each grid (emission groups 1 and 2), and setting the
“Use_Leftover_Group” flag to TRUE so that the model calculates the third group internally.
The point source group 1 filename is blank because group 1 is a category with no point source
emissions (e.g., biogenics).
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

164

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
7. Source Apportionment

&SA_Control
SA_File_Root

= './OSAT_output/CAMx.APCA.020604',

SA_Master_Sfc_Output
SA_Nested_Sfc_Output
SA_Summary_Output
SA_Stratify_Boundary
SA_Deposition_Output
SA_Number_of_Source_Regions
SA_Number_of_Source_Groups
Use_Leftover_Group
SA_Treat_SULFATE_Class
SA_Treat_NITRATE_Class
SA_Treat_SOA_Class
SA_Treat_PRIMARY_Class
SA_Treat_MERCURY_Class
SA_Treat_OZONE_Class
SA_Use_APCA

=
=
=
=
=
=
=
=
=
=
=
=
=
=
=

.true.,
.true.,
.true.,
.false.,
.false.,
19,
3,
.false.,
.true.,
.true.,
.false.,
.false.,
.false.,
.true.,
.true.,

SA_Receptor_Definitions = './OSAT_input/receptor.cities ',
SA_Source_Area_Map(1)
= './OSAT_input/OSAT.source.area.map',
SA_Source_Area_Map(2)
= ' ',
SA_Use_Partial_SourceMap = .false.,
Partial_Source_Area_Map(1,1) = ' ',
! Map for SA group 1, grid 1
Partial_Source_Area_Map(1,2) = ' ',
! Map for SA group 1, grid 2
SA_PT_Override
= .false.,
SA_Master_Restart
SA_Nested_Restart

= './OSAT_output/CAMx.APCA.020603.sa.inst',
= './OSAT_output/CAMx.APCA.020603.sa.finst',

SA_Points_Group(1)
SA_Points_Group(2)
SA_Points_Group(2)

= ' ',
= './OSAT_input/utils.020604',
= './OSAT_input/other.020604',

SA_Emiss_Group_Grid(1,1)
SA_Emiss_Group_Grid(1,2)
SA_Emiss_Group_Grid(2,1)
SA_Emiss_Group_Grid(2,2)
SA_Emiss_Group_Grid(3,1)
SA_Emiss_Group_Grid(3,2)

=
=
=
=
=
=

'./OSAT_input/bio.grd1.020604',
'./OSAT_input/bio.grd2.020604',
'./OSAT_input/util.grd1.020604',
'./OSAT_input/util.grd2.020604',
'./OSAT_input/othr.grd1.020604',
'./OSAT_input/othr.grd2.020604',

&

Figure 7‐7c. This figure follows from Figure 7‐7b: it is a continuation day of a 2‐grid run with
three emission groups, and all three emission groups are defined explicitly by supplying extra
emission files; therefore, the “Use_Leftover_Group” flag is set to FALSE. The point source
group 1 filename is blank because group 1 is a category with no point source emissions (e.g.,
biogenics). APCA is used to attribute ozone sources, so biogenic emissions MUST be present
as group 1. PSAT will trace PM sulfate and nitrate species.
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

165

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
7. Source Apportionment

The number of emission files that need to be supplied for different model configurations is
summarized in Table 7‐1; the table also shows how the emissions groups are numbered, which
is reflected in the tracer species names (defined below).

Table 7‐1. Numbers of emission file sets (i.e., gridded files and point source file) needed for
different model configurations. APCA requires at least two emission groups, and the first
group must be biogenic emissions.
Number of
Emission Groups
n=1
n>1
n>1

Use
Leftover Group
Not Applicable
False
True

Number of Emission
File Sets Needed
0
n
n‐1

Numbering of Emission Groups
and Tracer Species
0
1,2,3,...n
1,2,3,...n

When specifying point source files to resolve source categories, the list of point sources on each
file must be identical (i.e., same number of sources, same order) to the regular model point
source file. This formal restriction is necessary to ensure that point sources are correctly cross‐
referenced within CAMx. Thus, a point source file for a specific source group may need to
contain records for sources that are not in the group: these records should have zero
emissions.
7.3.3 Source Area Mapping
SA can apportion ozone, PM and precursor concentrations among several geographic regions
within the modeling domain, as shown in Figure 7‐1. SA requires a digital map of the modeling
grid that defines how tracers are allocated spatially – this “source area map” file assigns each
grid cell to one or more geographic source regions. A source area map must be defined for the
master grid and optionally any nested grids. The source area map formats are identical among
all grids, but maps for nested grids must include the boundary (“buffer”) rows and columns.
The source regions defined on each nest take precedence over those defined for the master
grid. If a source area map is not provided for a specific nest then the source region definition
will be defined by the source area map for the parent grid.
There are two ways to define source area maps. The first (original) approach is to uniquely
assign the entirety of each grid cell to a single geographic region with which to apportion all
source categories present in that grid cell. The second option allows for the fractional
allocation of each grid cell to multiple regions, for example, in cases where several geopolitical
boundaries intersect within a single cell. Furthermore, separate fractional area maps may be
developed that uniquely define source region distributions for each emission category to be
tracked by SA. The original source area maps are required to run SA, but can be superceded by
the optional fractional source area maps. The original maps provide the default SA region
definition in case a fractional region map file is not provided for one or more source categories.
If no fractional area maps are provided to CAMx, then the entire SA treatment defaults back to
the original area map definition. CAMx includes reports in the output diagnostic file to allow
the user to review the SA region configuration.
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

166

www.camx.com

March 2016

31
31
31
31
31
31
31
31
31
31
31
31
21
21
21
21
21
21
21
21
21
11
11
11
11
11
11
11
11
11
27
27
27
27
27
27
27
27
27
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33

CAMx User’s Guide Version 6.3
7. Ozone Source Apportionment Technology

31
31
31
31
31
31
31
31
31
31
31
31
21
21
21
21
21
21
21
21
21
11
11
11
11
11
11
11
11
11
27
27
27
27
27
27
27
27
27
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33

31
31
31
31
31
31
31
31
31
31
31
31
21
21
21
21
21
21
21
21
21
11
11
11
11
11
11
11
11
11
27
27
27
27
27
27
27
27
27
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33

31
31
31
31
31
31
31
31
31
31
31
31
31
21
21
21
21
21
21
21
21
11
11
11
11
11
11
11
11
11
27
27
27
27
27
27
27
27
27
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33

18
31
31
31
31
31
31
31
31
31
31
31
31
21
21
21
21
21
21
21
21
11
11
11
11
11
11
11
11
11
27
27
27
27
27
27
27
27
27
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
39
39
39
39
39

18
18
18
18
18
18
18
18
18
18
10
10
10
10
10
21
21
21
21
21
21
11
11
11
11
11
11
11
11
11
27
27
27
27
27
27
27
27
27
27
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
39
39
39
39
39
39

18
18
18
18
18
18
18
18
18
18
10
10
10
10
10
10
10
10
10
21
21
11
11
11
11
11
11
11
11
11
27
27
27
27
27
27
27
27
27
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
39
39
39
39
39
39
39

18
18
18
18
18
18
18
18
18
18
10
10
10
10
10
10
10
10
10
20
20
11
11
11
11
11
11
11
11
11
27
27
27
27
27
27
27
27
27
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
33
39
39
39
39
39
39
39
39

18
18
18
18
18
18
18
18
18
18
10
10
10
10
10
10
10
10
10
20
20
20
20
11
11
11
11
11
11
11
27
27
27
27
27
27
27
27
27
27
33
33
33
33
33
33
33
33
33
33
33
33
33
33
39
39
39
39
39
39
39
39
39

18
18
18
18
18
18
18
18
18
18
10
10
10
10
10
10
10
10
10
20
20
20
20
20
20
20
20
20
20
20
20
20
2
2
2
2
2
2
2
2
33
33
33
33
33
33
33
33
33
33
33
33
33
39
39
39
39
39
39
39
39
39
39

18
18
18
18
18
18
18
18
18
18
10
10
10
10
10
10
10
10
10
20
20
20
20
20
20
20
20
20
20
20
20
20
2
2
2
2
2
2
2
2
2
2
13
13
13
13
13
33
33
33
33
13
13
39
39
39
39
39
39
39
39
39
39

18
18
18
18
18
18
18
18
18
18
10
10
10
10
10
10
10
10
10
20
20
20
20
20
20
20
20
20
20
20
20
20
2
2
2
2
2
2
2
2
2
2
13
13
13
13
13
13
13
13
13
13
39
39
39
39
39
39
39
39
39
39
39

18
18
18
18
18
37
37
18
18
18
10
10
10
10
10
10
10
10
10
20
20
20
20
20
20
20
20
20
20
20
20
20
2
2
2
2
2
2
2
2
2
2
13
13
13
13
13
13
13
13
13
13
13
39
39
39
39
39
39
39
39
39
39

18
37
37
37
37
37
37
37
18
18
10
10
10
10
10
10
10
10
10
20
20
20
20
20
20
20
20
20
20
20
20
20
2
2
2
2
2
2
2
2
2
2
13
13
13
13
13
13
13
13
13
13
13
39
39
39
39
39
39
39
39
39
39

37
37
37
37
37
37
37
37
37
18
10
10
10
10
10
10
10
10
10
20
20
20
20
20
20
20
20
20
20
20
20
20
2
2
2
2
2
2
2
2
2
2
13
13
13
13
13
13
13
13
13
13
13
39
39
39
39
39
39
39
39
39
39

37
37
37
37
37
37
37
37
37
37
37
10
10
10
10
10
10
10
10
8
8
8
20
20
20
20
20
20
20
20
20
20
2
2
2
2
2
2
2
2
2
2
13
13
13
19
19
19
13
13
13
13
13
13
39
39
39
39
39
39
39
39
39

37
37
37
37
37
37
37
37
37
37
37
37
37
10
10
10
10
8
8
8
8
8
8
20
20
20
20
20
20
20
20
20
2
2
2
2
2
2
19
19
19
19
19
19
19
19
19
19
13
13
13
13
13
13
39
39
39
39
39
39
39
39
39

37
37
37
37
37
37
37
37
37
37
37
37
37
8
8
10
8
8
8
8
8
8
8
8
20
20
20
20
20
20
20
20
2
2
2
32
19
19
19
19
19
19
19
19
19
19
19
19
13
13
13
13
13
13
39
39
39
39
39
39
39
39
39

17
17
37
37
37
37
37
37
37
37
37
37
37
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
20
20
20
20
20
32
32
32
19
19
19
19
19
19
19
19
19
19
19
19
19
13
13
13
13
13
39
39
39
39
39
39
39
39
39

17
17
37
37
37
37
37
37
37
37
37
37
37
37
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
20
32
32
32
32
32
19
19
19
19
19
19
19
19
19
19
19
19
19
19
19
13
13
13
13
39
39
39
39
39
39
39
39

17
17
17
37
37
37
37
37
37
37
37
37
37
37
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
12
32
32
32
32
32
19
19
19
19
19
19
19
19
19
19
19
19
19
19
19
39
39
13
39
39
39
39
39
39
39
39
39

17
17
17
37
37
37
37
37
37
37
37
37
37
37
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
12
12
32
32
32
32
32
19
19
19
19
1
1
1
1
1
1
1
1
1
1
1
39
39
39
39
39
39
39
39
39
39
39
39

17
17
17
17
17
37
37
37
37
37
37
37
37
37
8
8
8
8
8
8
8
8
8
8
8
8
9
12
12
12
12
32
32
32
32
32
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
39
39
39
39
39
39
39
39
39
39
39
39

17
17
17
17
17
37
37
37
37
37
37
37
37
37
8
9
9
9
9
9
9
9
9
9
9
9
9
12
12
12
12
32
32
32
32
32
1
1
1
1
1
1
1
1
1
1
1
1
6
6
6
39
39
39
39
39
39
39
39
39
39
39
39

17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
9
9
9
9
9
9
9
9
9
9
9
9
12
12
12
12
32
32
32
32
32
1
1
1
1
1
1
1
1
1
1
1
1
6
6
39
39
39
39
39
39
39
39
39
39
39
39
39

17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
9
9
9
9
9
9
9
9
9
9
9
12
12
12
12
32
32
32
32
32
1
1
1
1
1
1
1
1
1
1
1
1
6
6
6
39
39
39
39
39
39
39
39
39
39
39
39

17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
9
9
9
9
9
9
9
9
9
9
12
12
12
12
12
32
32
32
32
32
1
1
1
1
1
1
1
1
1
1
1
1
6
6
6
6
39
39
39
39
39
39
39
39
39
39
39

17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
9
9
9
9
9
9
9
9
9
12
12
12
12
12
12
32
32
32
32
32
7
7
7
7
7
1
1
1
1
1
1
1
6
6
6
6
6
39
39
39
39
39
39
39
39
39
39

17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
26
26
26
26
26
26
26
26
12
12
12
12
12
12
12
32
32
32
32
32
7
7
7
7
7
7
7
7
7
7
7
7
7
6
6
6
6
39
39
39
39
39
39
39
39
39
39

40
17
17
17
17
17
17
17
17
17
17
17
17
17
17
17
26
26
26
26
26
26
26
26
12
12
12
12
12
12
12
32
32
32
32
32
7
7
7
7
7
7
7
7
7
7
7
7
7
6
6
6
39
39
39
39
39
39
39
39
39
39
39

40
40
17
17
17
17
17
17
17
17
17
17
17
17
17
17
26
26
26
26
26
26
26
26
26
12
12
12
12
12
12
32
32
32
25
25
7
7
7
7
7
7
7
7
7
7
7
7
7
6
6
6
39
39
39
39
39
39
39
39
39
39
39

40
40
40
40
17
17
17
17
17
17
17
17
17
17
17
17
26
26
26
26
26
26
26
26
26
12
12
12
12
12
12
32
32
32
25
25
30
7
7
7
7
7
7
7
7
7
7
7
7
6
6
6
6
6
39
39
39
39
39
39
39
39
39

40
40
40
40
40
17
17
17
17
17
17
17
17
17
17
26
26
26
26
26
26
26
26
26
26
12
12
12
12
12
35
32
32
25
25
25
30
30
7
7
7
7
7
7
7
7
7
7
7
6
6
6
6
6
6
6
6
6
6
6
39
39
39

40
40
40
40
40
40
40
40
40
40
40
40
40
40
40
26
26
26
26
26
26
26
26
26
26
36
36
36
12
35
35
32
32
25
25
30
30
30
30
30
7
7
7
7
7
7
7
7
7
6
6
6
6
6
6
6
6
6
6
6
6
6
39

40
40
40
40
40
40
40
40
40
40
40
40
40
40
26
26
26
26
26
26
26
26
26
26
36
36
36
36
36
35
35
32
25
25
25
30
30
30
30
30
30
7
7
7
7
7
7
7
7
6
6
6
6
6
6
6
6
6
6
6
6
6
6

40
40
40
40
40
40
40
40
40
40
40
40
40
40
26
26
26
26
26
26
26
26
26
36
36
36
36
36
36
35
35
25
25
25
25
25
30
30
30
30
30
30
30
30
7
7
7
7
7
6
6
6
6
6
6
6
6
6
6
6
6
6
6

40
40
40
40
40
40
40
40
40
40
40
40
40
40
26
26
26
26
26
26
26
36
36
36
36
36
36
36
36
35
35
25
25
25
25
25
30
30
30
30
30
30
30
30
30
7
38
38
38
38
38
38
38
6
6
6
6
6
6
6
6
6
6

40
40
40
40
40
40
40
40
40
40
40
40
40
40
28
28
28
28
28
28
28
28
36
36
36
36
36
36
35
35
35
25
25
25
25
25
25
30
30
30
30
30
30
30
30
38
38
38
38
38
38
38
38
38
38
38
38
6
6
6
6
6
6

40
40
40
40
40
40
40
40
40
40
40
40
40
24
24
28
28
28
28
28
28
28
36
36
36
36
35
35
35
35
35
25
25
25
25
25
25
30
30
30
30
30
30
30
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38

40
40
40
40
40
40
40
40
40
40
40
40
40
24
24
28
28
28
28
28
28
28
15
36
36
35
35
35
35
35
35
25
25
25
25
25
25
25
30
30
30
30
30
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38

40
40
40
40
40
40
40
40
40
40
24
24
24
24
24
28
28
28
28
28
28
28
36
36
35
35
35
35
35
35
35
25
25
25
25
25
25
25
25
30
30
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38

40
40
40
40
40
40
40
40
40
40
24
24
24
24
24
28
28
28
28
28
28
28
36
35
35
35
35
35
35
35
35
25
25
25
25
25
25
25
25
25
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38

40
40
40
40
40
40
40
40
40
40
24
24
24
24
24
28
28
28
28
28
28
28
15
35
35
35
35
35
35
35
35
25
25
25
25
25
25
25
25
25
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38

40
40
40
40
40
40
40
40
40
40
24
24
24
24
24
28
28
28
28
28
28
28
15
15
15
15
35
35
35
35
35
25
25
25
25
25
25
25
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38

40
40
40
40
40
40
40
40
40
40
24
24
24
24
24
28
28
28
28
28
28
28
15
15
15
15
15
35
35
35
35
25
25
25
25
25
25
25
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38

40
40
40
40
40
40
40
40
24
24
24
24
24
24
24
28
28
28
28
28
28
28
15
15
15
15
15
35
35
35
35
25
25
25
25
25
25
25
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38

40
40
40
40
40
40
40
24
24
24
24
24
24
24
24
28
28
28
28
28
28
4
4
4
15
15
15
35
35
35
35
25
25
25
25
25
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38

40
40
40
40
40
40
24
24
24
24
24
24
24
24
24
28
28
28
28
28
28
28
23
23
4
4
15
35
38
38
38
38
38
25
25
25
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38

40
40
40
40
40
40
24
24
24
24
24
24
24
24
24
24
24
23
23
23
23
23
23
23
23
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38

40
40
40
40
40
40
24
24
24
24
24
24
24
24
24
24
24
24
23
23
23
23
23
23
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38

40
40
40
40
40
40
24
24
24
24
24
24
24
24
24
24
24
24
24
24
23
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38

40
40
40
40
40
40
34
34
34
34
34
34
24
16
16
3
3
3
24
24
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38

40
40
40
40
40
40
34
34
34
34
34
34
34
16
16
3
3
3
24
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38

40
40
40
40
40
40
34
34
34
34
22
22
22
16
16
3
3
24
24
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38

40
40
40
40
40
40
34
34
22
22
22
22
22
16
16
29
29
24
24
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38

40
40
40
40
40
22
22
22
22
22
22
22
22
16
16
29
29
29
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38

40
40
40
40
14
14
14
14
14
14
14
14
16
16
16
16
16
16
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38

40
40
14
14
14
14
14
14
14
14
14
14
16
16
16
16
16
16
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38

14
14
14
14
14
14
14
14
14
14
38
38
38
38
38
16
16
16
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38

14
14
14
14
14
14
14
14
14
14
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38

14
14
14
14
14
14
14
14
14
14
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38

14
14
14
14
14
14
14
14
14
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38

14
14
14
14
14
14
14
14
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38

40
40
40
40
40
14
14
14
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38
38

Figure 7‐8. Example of the original source area map file for the domain and source areas shown in Figure 7‐1.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

167

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
7. Ozone Source Apportionment Technology

The original SA map format is simple: an array of 3‐digit integers (i3) spanning the entirety of a
particular CAMx grid. Figure 7‐8 shows the source area mapping file for the single grid
corresponding to Figure 7‐1. Since the CAMx domain in Figure 7‐1 has 63 rows and 64 columns
of cells, the file shown in Figure 7‐8 has 63 lines with 64 numbers on each line. The first
number in the top left corner always corresponds to the northwest corner of the domain. This
file is typically generated using GIS software by overlaying the modeling grid onto geopolitical
maps and using the dominant coverage in each grid cell as its source region assignment.
To facilitate multiple intersecting regions within each grid cell, a fractional area map for a single
grid may include multiple “panels”, where the total number of panels is determined by the
maximum number of region overlaps found among all grid cells. For example, if a particular
grid cell contains a grid‐wide maximum of four overlapping regions, then the fractional map
contains four panels, each listing one of the four regions and its fractional coverage in that cell.
The fractional SA map file has the following format:
Loop over number of panels
/SRCMAPnn-mm/

Header keyword, where nn is source
category/group ID, mm is panel ID

Loop from ny grid rows to 1
(regn(i,j),frc(i,j),i=1,nx)
End loop over rows
End loop over panels
/END/

Loop over nx grid columns,
500(i3,1x,f5.1)

End of file keyword

The integer variable array regn is the region index that exists in cell (i,j) and the real variable
array frc is the fraction (percent) of cell (i,j) covered by that region. For non‐zero cell
fractions, both regn and frc must be listed, otherwise regn is shown as 0 and frc is blank
to maximize visual clarity of the file. The total coverage among all regions in each grid cell
equals 100.0% when summed over all panels. An example is shown in Figure 7‐9 for a small
grid of 10x10 cells.
The original source area map, and possibly even the fractional map, may not adequately resolve
the region to which certain point sources should be assigned. To provide finer control of point
source assignments to geographic areas, the region index can be specified for any point source
using the kcell variable in the point source file (see file description in Section 3). This feature
is referred to as “point source override.”
7.3.3.1 Generating Fractional Area Maps From SMOKE Reports
A Fortran tool called REGNMAP has been developed to support the development of fractional
area maps using information derived from the Sparse Matrix Operator Kernel Emissions
(SMOKE) processing system. SMOKE can be configured to output information to “report files”
that list the spatial allocation of county‐level emissions to a particular modeling grid by criteria
pollutant (NOx, VOC, SOx, and PM). A separate fractional area map can be developed for each
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

168

www.camx.com

March 2016

/SRCMAP03-01/
5 100.0 5 100.0
5 100.0 5 100.0
5 100.0 5 100.0
5 100.0 5 100.0
5 100.0 5 100.0
5 100.0 5 100.0
5 100.0 5 100.0
5 100.0 5 100.0
5 100.0 5 100.0
5 100.0 5 100.0
/SRCMAP03-02/
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
/END/

CAMx User’s Guide Version 6.3
7. Ozone Source Apportionment Technology

5
5
5
5
5
5
5
5
5
5
0
0
0
0
0
0
0
0
0
0

100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0

5
5
5
5
5
5
5
5
5
5
0
0
0
0
0
0
0
0
0
0

100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0

5 25.0
5 25.0
5 30.0
5 35.0
5 50.0
5 60.0
5 80.0
5 100.0
5 100.0
5 100.0

6
6
6
6
6
6
6
5
5
5

6
6
6
6
6
6
6
0
0
0

0
0
0
0
0
0
0
6
6
6

75.0
75.0
60.0
65.0
50.0
40.0
20.0

100.0
100.0
100.0
100.0
100.0
100.0
100.0
20.0
50.0
60.0

6
6
6
6
6
6
6
6
6
6

80.0
50.0
40.0

0
0
0
0
0
0
0
0
0
0

100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0

6
6
6
6
6
6
6
6
6
6
0
0
0
0
0
0
0
0
0
0

100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0

6
6
6
6
6
6
6
6
6
6
0
0
0
0
0
0
0
0
0
0

100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0

6
6
6
6
6
6
6
6
6
6

100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0

0
0
0
0
0
0
0
0
0
0

Figure 7‐9. Example fractional area map file for a small (10x10) grid. This file is for source
category/group #3 and includes 2 map panels. The grid covers source region #5 and #6 and
these regions overlap in the middle of the domain. Panel 2 shows just the remaining overlap
information for region #6.

source sector or group of sectors depending on how the user runs SMOKE and configures the
list of sectors to be tracked for SA (e.g., consider spatial differences between urban area
sources and agricultural non‐road sources). SMOKE spatial allocation reports must be invoked
in order to utilize the REGNMAP program to develop fractional area maps for CAMx.
Mobile emissions are not spatially allocated in the same way as non‐road and stationary
sources, so SMOKE reports are not available for the on‐road sector if SMOKE‐MOVES is used.
Therefore, the on‐road sector must continue to be tracked in SA using the original source area
map. Additionally, SMOKE does not allocate elevated point sources to the modeling grid like
surface county‐level sources, and so SMOKE spatial allocation reports are not available for point
sources. All category‐specific point source files to be tracked by SA are assigned to the original
region map definition by default, except for those individual point sources flagged for source
region override (see Section 3).
REGNMAP reads SMOKE spatial allocation reports for a specific modeling grid and source
category (or group of categories), extracts emissions data by grid cell and state/county Federal
Information Processing Standards (FIPS) code, and generates a fractional area map file for that
grid and source category/group. The list of SA regions to process are externally defined as a
county or group of counties, a state or group of states, or all other undefined areas. REGNMAP
provides an option to select among the criteria pollutants NOx, VOC, SOx, or PM2.5 as the basis
to define the fractional grid cell areas in case the specific source category/group to be
processed is uniquely characterized by one of these species (e.g., NOx for mobile sources, SOx
for power plants). Alternatively, the user may select “All” criteria pollutants, in which case the
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS
www.camx.com
169

March 2016

CAMx User’s Guide Version 6.3
7. Ozone Source Apportionment Technology

fractional allocation is based on the sum of all emissions reported per grid cell for that source
category/group. Additional information on how to run REGNMAP is provided with the
program.
7.3.4 Receptor Definition
Tracer concentrations can be optionally output to a text file for selected receptor locations at
the model’s output frequency (usually 1 hour). The receptors for each model run are defined in
the “receptor definition” input file. Three types of receptors are supported:
POINT

a point specified in the CAMx projection coordinate system.
Concentrations at the point are determined by bi‐linear
interpolation of the surrounding four surface grid cells.

SINGLE CELL

a single surface grid cell identified by grid cell index.

CELL AVERAGE

a group of surface grid cells identified by a range of grid indices
that are averaged together to provide multi‐cell average tracer
concentrations.

WALL OF CELLS

a group of grid cells identified by a range of grid and layer indices
that define a wall (i.e., a flux plane).

For the receptor types that are defined by grid cell it is necessary to specify the grid containing
the receptor on the receptor definition record. Grid numbers are defined using the internal
CAMx grid ordering. The grid numbering as defined by CAMx is shown in a table in the .diag
file. Each receptor can be identified by a 10 character name. The formats for specifying each
receptor type are given in Table 7‐2. An example receptor file is shown below:

POINT
SINGLE CELL
CELL AVERAGE
31
32
33
34
31
32
33
34
WALL OF CELLS

City 1
Cell 1
Region 10

1024.0
1
2

-272.0
45
8

18

10
18

20
18

1

5

19
19
19
19
18
18
18
18
Boundary1

2

7.3.5 Output File Formats
SA writes several output files that are in the CAMx Fortran binary format, as described in
Section 3. These include the master and nested grid tracer instantaneous concentration files
(.sa.inst and .sa.finst), the grid‐specific surface tracer average concentration file
(.sa.grdnn), and the grid‐specific surface deposited mass file (.sa.depn.grdnn). In
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

170

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
7. Ozone Source Apportionment Technology

Table 7‐2. Format for the receptor definition file.
Receptor Type
POINT

SINGLE CELL

CELL AVERAGE

WALL OF CELLS

Line
1
1
1
1
1
1
1
1
1
1
1
1
1
2‐M
2‐M
1
1
1
1
1
2
2
3
3

Columns
1‐15
21‐30
31‐40
41‐50
1‐15
21‐30
31‐40
41‐50
51‐60
1‐15
21‐30
31‐40
41‐50
1‐10
11‐20
1‐15
21‐30
31‐40
41‐50
51‐60
41‐50
51‐60
41‐50
51‐60

Data
The word “POINT”
Receptor name
X co‐ordinate
Y co‐ordinate
The word “SINGLE CELL”
Receptor name
Grid Number
X cell number
Y cell number
The words “CELL AVERAGE”
Receptor name
Grid number
The number of cells to average (M)
X cell number
Y cell number
The words “WALL OF CELLS”
Receptor name
Grid number
X‐cell begin
X‐cell end
Y‐cell begin
Y‐cell end
Z‐cell begin
Z‐cell end

addition, SA writes out tracer concentrations for selected receptor locations to an text file
(.sa.receptor). The naming conventions for tracer species and the format of the receptor
concentration file are discussed below.
7.3.5.1 Tracer Species Names
The names of tracer species uniquely identify the information carried by each species and
together identify the SA configuration. Species names have less than ten characters, consistent
with the CAMx convention. The naming conventions are as follows:
Emission Sources
SSSeeerrr
where:
SSS Species type, e.g., NOX, VOC, O3V, O3N, PSO4, etc.
eee Emissions group:
Single group, always 000
Multiple groups, 001, 002, etc.
rrr Region tracer released from, 001, 002, 003, etc.
Initial/Boundary
SSSeeerrr
where:
SSS Species type, e.g., NOX, VOC, O3V, O3N, PSO4, etc.
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

171

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
7. Ozone Source Apportionment Technology

eee

rrr
Examples:

Initial Concentrations: always 000
Boundary Concentrations not stratified by boundary: always 000
Boundary Concentrations stratified by boundary: WST, EST, STH, NTH,
TOP indicating boundary of origin
IC for Initial Concentrations, BC for Boundary Concentrations
NOX000015, VOC002015, O3V000IC, O3NTOPBC

7.3.5.2 Receptor Concentration File
Tracer concentrations at user‐specified receptor locations are output to the “receptor
concentration” file. The file is in comma delimited text format suitable for importing into a
spreadsheet. An example output file is shown in Figure 7‐10. Two header lines at the top of
the file identify the model version and the date the run was performed. Next, two lines identify
the time period covered by the file and the averaging interval (generally one hour, determined
by the CAMx simulation control file). Next, three lines define the SA configuration, followed by
the numbers of tracer species that result from this configuration. The names of each tracer
species are listed by tracer type: the order in which species are listed here is the same as the
order in which tracer concentrations are given later in the file.
The tracer species names are followed by the number of receptors and receptor names as
specified in the “receptor definition” file. The tracer concentrations are reported in blocks with
a date and time stamp at the head of each block. Within each block, receptors are reported in
numerical order. For each receptor, there are data for the tracer species identified at the
heading “Tracer Names”. All values are in CAMx units of ppm for gases and g/m3 for PM.
7.3.6 Postprocessing
The tracer concentrations in the gridded surface concentration files can be displayed using any
post‐processing software normally used for displaying CAMx average file output formats.
The receptor concentration file contains information for all receptors and all hours within the
model run that created the file. It is left to the user to develop post‐processing tools to analyze
the information contained in this file.

7.4 Steps In Developing Inputs And Running SA
Below is a simple methodological list of steps to follow in setting up and running SA. The
process is similar among the OSAT/APCA, PSAT, and DDM Probing Tools.
1) Define the source groups and regions that you wish to track. Keep in mind that memory
resources increase dramatically as the number of tracers grows. Probing Tool
applications with large numbers of tracers, tracer classes, nested grids or grid cells may
exceed available memory.
2) Build a source region map (Figure 7‐8) that defines the spatial allocation of tracer
emissions. For small domains or small number of regions, this can be done by hand. We
suggest using GIS software to develop complex source region maps on large grids.
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

172

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
7. Ozone Source Apportionment Technology

CAMx,CAMx 6.30 Test Problem -- Mech6 CF CB05 SA.OMP,Source Apportionment, SA 160408,
Thu Mar 31 13:18:53 2016
File Duration ,
Average Interval ,
Number
Number
Number
Number
Number
Number
Number
Number
Number
Number
Number
Number
Number
Number
Number
Number

of
of
of
of
of
of
of
of
of
of
of
of
of
of
of
of

02154,
1.0000

timing periods
source areas
emission groupings
tracer species
VOC species
O3N species
O3V species
OON species
OOV species
NIT species
RGN species
TPN species
NTR species
HN3 species
INERT TIME species
DECAY TIME species

0.00,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,

02154,

24.00,

0
4
4
180
18
18
18
18
18
18
18
18
18
18
0
0

Tracer Names,
VOC000IC ,VOC000BC,VOC001001,VOC001002,VOC001003,VOC001004,VOC002001,…
O3N000IC ,O3N000BC,O3N001001,O3N001002,O3N001003,O3N001004,O3N002001,…
O3V000IC ,O3V000BC,O3V001001,O3V001002,O3V001003,O3V001004,O3V002001,…
(List continues for remaining tracer species names)
Number
No,
1,
2,
3,

4,

of receptors ,
4
Name,
Type,
Grid#,
Xloc,
Yloc,
City 1
,
0, ,
1024.0,
-272.0,
Cell 1
,
1,
1,
45,
18,
Region 10,
8,
2,
31,
19,
32,
19,
33,
19,
34,
19,
31,
18,
32,
18,
33,
18,
34,
18,
Boundary1,
3,
2,
10,
20,
18,
18,
1,
5,

Time Varying Tracer Data,
Data for Period,
02154,
0.00,
02154,
1.00,
Receptor,
1,
1.3265E-02, 1.3544E-09, 1.0000E-16, 1.0974E-15, 1.0000E-16, 1.0000E-16,…
1.2237E-01, 3.3869E-08, 1.0000E-16, 1.6165E-14, 1.0000E-16, 1.0000E-16,…
8.7304E-02, 1.1926E-08, 1.0000E-16, 1.0000E-16, 1.0000E-16, 1.0000E-16,…
9.0300E-02, 1.5269E-08, 1.0188E-16, 2.6997E-15, 1.0213E-16, 1.0162E-16,…
1.0036E-16, 4.0640E-15, 1.0036E-16, 1.0036E-16, 1.0036E-16, 1.0029E-16,…
1.0000E-16, 3.7563E-15, 1.0000E-16, 1.0000E-16, 1.0000E-16, 1.0000E-16,…
Receptor,
2,
(File continues with data for remaining receptors and hours)

Figure 7‐10. Example receptor concentration file. Lines ending with “…” are truncated to fit
the page, and the file would continue with data for additional receptors and hours in the
same format.
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

173

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
7. Ozone Source Apportionment Technology

3) Process the emissions inventory into the separate source group files that you want to
track (e.g., mobile, area, point, biogenic, etc.).
a) Consideration of potential source apportionment or sensitivity applications prior to
any emissions processing can be very beneficial so that files by group are available
for later use.
b) Elevated point sources will automatically be assigned to the source region in which
they reside. However, you may override the region to which each individual point
source is assigned (see the definition of kcell in Section 3, Elevated Point Source
File). A point source region does not need to be defined in the source region map,
e.g., you could have a map with two regions that split the domain in half, with a
third region assigned arbitrarily to represent elevated point sources only.
4) Edit the CAMx control namelist file (Section 2).
a) Set the Probing_Tool variable to “SA”; this will activate the &SA_Control
namelist module.
b) Edit or add the &SA_Control namelist module (described earlier). Provide the
required information, including:
 output paths
 whether to stratify boundary conditions
 flags to turn on specific ozone or PM classes
 number of source regions
 number of source groups
 whether to use the leftover group option
 receptor definitions
 list of input emission files by group.
c) Note that APCA requires that the biogenic emission files for each grid are listed first.
Several examples are shown in Figure 7‐7.
5) Configure the CAMx source code to define the number of tracers, and build an
executable. This will ensure that you have sufficient memory for the Probing Tool
application.
a) Edit the file Includes/camx.prm
b) Change the parameter MXTRSP, following the instructions provided in the file.
CAMx is distributed with MXTRSP=1 to minimize memory requirements for
standard applications of the model. If you run SA with an insufficient value, the
model will stop and tell you the required value of MXTRSP for your application.
c) Execute the CAMx Makefile to build an executable program (Section 2).
6) Run CAMx and review the diagnostic output files to ensure that the model is correctly
interpreting and running the Probing Tool configuration that you have specified. Ensure
that CAMx is generating the proper output files that you are expecting. Review the
table of emissions by source group and region.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

174

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
7. Ozone Source Apportionment Technology

7) Review gridded tracer fields using commonly available plotting programs. Utilities such
as PAVE or Verdi will read Probing Tool files directly. Use of any other software may
require specialized re‐formatting procedures.
8) You may post‐process and analyze SA receptor files using your own spreadsheet or
database software.
9) Probing Tool gridded tracer output files are written in the same Fortran binary format as
the regular CAMx concentration output files. You can post‐process gridded output fields
using any software that reads CAMx files, or you can adapt those programs or build your
own software to generate specialized analysis and graphical products.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

175

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
8. Decoupled Direct Method

8. DECOUPLED DIRECT METHOD FOR SENSITIVITY ANALYSIS
Photochemical modelers have traditionally used sensitivity analysis both for model
performance evaluation and emission control strategy design. The simplest approach to
sensitivity analysis, often referred to as the “brute‐force” approach, involves changing a model
input parameter, rerunning the model, and then evaluating the change in model output for
each parameter to be investigated. For example, a model performance evaluation may use
sensitivity simulations to evaluate the impact of changing initial or boundary conditions (ICs and
BCs), biogenic emissions, anthropogenic emissions, etc. Control strategy evaluation may
reduce VOC and NOx emissions to determine whether VOC and/or NOx reduction strategies are
the most effective path to reduce ozone.
The advantages of the “brute force” method for sensitivity analysis are:



Applicable to any model input parameter;



Results are conceptually easy to explain and interpret.

The limitations of the “brute force” method are:



Computationally inefficient;



Sensitivity depends upon the magnitude of the perturbation if the model response is non‐
linear;



Sensitivity derived from small perturbations may contain significant levels of uncertainty
(numerical noise).

The last two points bear further explanation. If the model response to an input parameter
depends upon non‐linear components within the model (e.g., chemistry), then the relative
magnitude or even sign of the output response may change for perturbations of different sizes.
An example is the ozone response to NOx reductions in a VOC‐limited environment: smaller
reductions in NOx emissions increase ozone levels whereas larger NOx reductions decrease
ozone.
This situation can be illustrated mathematically. We define a “sensitivity coefficient” (s) which
represents the change in concentration (c) with respect to some input parameter (), evaluated
relative to the base state (=0),
s 

c


o

In general,  can be a vector (denoted as ), which contains multiple parameters related to
processes in the model (e.g., rate constants) or inputs to the model (e.g., emissions). The
concentration response to a change in  can be represented by a Taylor series of sensitivity
coefficients:

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

176

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
8. Decoupled Direct Method
n

c  x, t ;    c  x, t ;  0   
i 1



c
i

i  i 0 
o

 2c
1

2 i 1 j 1 i  j
n

i  i 0  j   j 0 

n

o

 third and higher order terms...
where n is the number of  vector elements, x is the spatial dimension vector, and t is time. In
the ozone sensitivity example above, the non‐linear ozone response to large NOx emission
reductions indicates that high‐order sensitivity coefficients (curvatures and inflections) are
significant relative to the first order sensitivity (linear response). As the magnitude of the input
perturbation tends to zero, the output response will become dominated by the first‐order
sensitivity. Therefore, very small changes in the input parameter may be required to use the
“brute force” method to estimate the first‐order (local) sensitivity. The practical limitation to
this approach is that since the change in output must be determined from the difference
between two simulations, small levels of numerical uncertainty (noise) in two very similar
outputs will contaminate the sensitivity calculation.
An alternative methodology for evaluating model sensitivity was developed by Dunker (1980,
1981) called the decoupled direct method (DDM). The DDM can be used to calculate the same
type of sensitivity coefficient as the “brute force” method. The difference is that with DDM,
sensitivity coefficients are calculated explicitly by specialized algorithms implemented in the
host model. Thus, the DDM offers several advantages over the brute force method:



Improved computational efficiency, especially as multiple sensitivities can be calculated
simultaneously;



Improved accuracy since sensitivities are not contaminated by numerical noise.

8.1 Implementation
The original CAMx implementation of the DDM considered only first‐order sensitivity for gas‐
phase species. Dunker et al. (2002) performed a rigorous analysis of DDM and demonstrated
excellent agreement against brute force tests. High‐order DDM (HDDM; Hakami et al., 2003;
Cohan et al., 2005) has since been implemented in CAMx. HDDM enables CAMx to calculate
second‐order sensitivities along with first‐order values for gas‐phase species (Koo et al.,
2007a,2008). The first‐order DDM sensitivity has been extended to PM species (Koo et al.,
2007b,2009). In the following discussion we use the term DDM generically to mean first and/or
higher order sensitivity.
The CAMx DDM calculates concentration sensitivity to several sources (i.e., emissions, ICs and
BCs) and to chemical rate constants. The sensitivity to be evaluated may bear a simple
relationship to a model input parameter, such as scaling ozone BCs by a factor (BCnew =  × BC0),
or additively increasing the ozone BC’s by a constant amount everywhere (BCnew =  + BC0). To
allow complete flexibility, the sensitivity perturbations are specified by providing additional IC,
BC, and/or emission input files with the same format as the regular model input files.
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

177

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
8. Decoupled Direct Method

As a result, the user can perform many different sensitivity calculations in a single run according
to the content of the DDM input files. For example, if the same BC file is specified for both
CAMx and DDM, the output sensitivity fields represent the sensitivity of the predicted
concentrations to those particular BCs. Simply scaling the output sensitivity coefficients fields
provides the incremental concentrations resulting from scaling the BCs. If a DDM BC file
contains constant concentrations everywhere, the sensitivity will correspond to a uniform
absolute increase in the BCs rather than a percentage increase as described above. Another
possibility includes a DDM BC file with a different spatial pattern than the CAMx input file. The
sensitivity coefficient would then correspond to changing both the geographic distribution and
magnitude of BCs. In short, the DDM input files can be arbitrary – different from the CAMx
input file in the overall magnitude of concentrations or emissions, different in the geographic
and temporal distribution, and different in the relative proportions of the chemical species.
However, the user must understand what perturbations are being considered in order to
properly interpret the resulting output sensitivity coefficient fields.
In mathematical terms, a regular model input file, for example the BC input file, represents
some set of functions of space and time fi(x,t), where each chemical species i can be defined by
a unique function. An additional input file provided to the DDM represents another set of
functions of space, time, and chemical species gi(x,t) that can be different from the regular
input file. The scalar parameter i is then defined by

Fi  x, t  

f i  x, t    i  g i  x, t  .

Here, i×gi(x,t) is the perturbation, and the user desires information on how the model would
respond if the input fi(x,t) is replaced by the input Fi(x,t). In the case of sensitivity to rate
constants, no user‐defined input file is provided and the perturbation is always defined as i×k
where k is a vector of selected rate constants. The DDM calculates the first‐order sensitivity
si(1)(x,t) and second‐order sensitivity si(2)(x,t) with respect to the scalar parameter i. The Taylor
series to second order then gives the estimate:

c l  x, t ;  i   c l  x, t ;  i  0    i  s i(1)  x, t  

1 2 ( 2)
 i  s i  x, t 
2

where cl(x;t;i) is the estimated model result for species l when Fi(x,t) is used as input, and
cl(x,t;i=0) is the base case model result when fi(x,t) is used as input.
For example, to calculate the sensitivity of the predicted ozone concentration to scaling
boundary ozone by a factor, CAMx would be provided with a DDM BC file that has the same
ozone values as the regular model BC file. The sensitivity coefficient fields output by CAMx
could then be used to estimate the resulting ozone concentration if the ozone BCs were
increased by 20%, as follows (for simplicity hereafter, the dependence on space, time, and
chemical species will be omitted):

c 0.2

 c 0  0.2  s (1) 

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

178

1
 0.2 2  s ( 2 )
2
www.camx.com

March 2016

CAMx User’s Guide Version 6.3
8. Decoupled Direct Method

To calculate the sensitivity of the predicted ozone concentration to increasing boundary ozone
by 10 ppb, CAMx would be provided with a DDM BC file that has a constant ozone value of 10
ppb. The sensitivity coefficient fields could be used to estimate the ozone concentration if the
ozone BCs were increased by 10 ppb as follows:

c 1.0

 c  0  1 s (1) 

1 2 ( 2)
1  s
2

where each sensitivity is calculated according to the 10 ppb addition carried in the DDM BC file
(thus sensitivities are scaled by unity). An alternative approach would be to provide CAMx with
a DDM BC file that has a constant ozone value of 1 ppb, and to estimate ozone response if the
ozone BCs were increased by 10 ppb would require that each sensitivity above be scaled by 10.
These are examples of relatively simple sensitivities. A more complex example would be to
calculate ozone sensitivity to scaling morning (6‐9 AM) NOx emissions in a specific group of grid
cells. In this case you would provide CAMx with a DDM emissions file where all values are zero
except for the NOx emissions in the selected grid cells between 6 AM and 9 AM, which would
have the same value as the regular emissions file. The sensitivity coefficient could be used to
predict the concentration after a scalar change () in the morning NOx emissions using the
same general equation as given above:

c

 c 0    s (1) 

1 2 ( 2)
 s
2

Any type of sensitivity perturbation can be described via an input file. However, the CAMx user
interface also provides easy ways to define some sensitivities that are likely to be used
frequently. In the first example above, the DDM BC file was described as having the “same
ozone values as the regular model BC file.” To avoid the effort of preparing an input file that is
trivially different from the regular model file, the user interface allows you to select specific
species from an input file to track ‐ in this case ozone. It is possible to separately track the
sensitivity to more than one species from the same file (e.g. ozone and NO). It is also possible
to track the combined sensitivity to a group of species, such as NOx, VOC, HRVOC, or ALL. The
user interface also provides a simple way to track sensitivities to emissions from specific grid
cells or groups of cells (sub‐regions).
8.1.1 Tracking Sensitivity Coefficients Within CAMx
DDM sensitivity coefficients are calculated in parallel to the core CAMx processes (emissions,
advection, diffusion, chemistry, deposition, etc.) that step the three‐dimensional concentration
fields forward in time. For some processes (e.g., chemistry and horizontal advection), the
sensitivity routines make use of information saved from the corresponding core model routines
in cases where the results depend non‐linearly upon species concentrations. In other cases, the
sensitivity algorithm is identical to the CAMx algorithm (e.g., horizontal diffusion) and both
concentrations and sensitivity coefficients can be processed by the single roiutine. Finally,
there are cases where a specialized module has been written for the sensitivity coefficients to
improve the computational efficiency (e.g., vertical advection).
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS
www.camx.com
179

March 2016

CAMx User’s Guide Version 6.3
8. Decoupled Direct Method

Priorities in the DDM coding implementation were:



Ensuring accuracy by using consistent numerical methods for the concentrations and
sensitivities;



Ensuring accuracy by calculating the concentrations and sensitivities within the chemistry
solvers over the same chemistry sub‐steps (for original DDM);



Optimizing the efficiency of the sensitivity coefficient calculations without compromising
accuracy;



Providing a flexible User Interface that allows calculation of sensitivities to all sources and
precursors;



Ensuring that the DDM algorithms have minimal impact on computer resource
requirements (memory and CPU time) when the DDM is not being used.

DDM and HDDM can be used with either of the two horizontal advection solvers available in
CAMx. The original first‐order DDM implementation was developed only for the EBI chemistry
solver; it cannot be used with LSODE. However, HDDM can be used with EBI or LSODE.
DDM and HDDM algorithms are not currently implemented for the CAMx Plume‐in‐Grid (PiG)
submodel or the ACM2 vertical diffusion schemes.
8.1.2 Flexi‐DDM
Although DDM is computationally much more efficient than the Brute‐Force method, it does
require much more additional CPU time and memory space over and above a standard CAMx
run, which can be significant especially when many first and second‐order sensitivities are
requested for a nested grid run with multiple source categories and multiple source regions.
The increased computational cost may not always be worthwhile if only part of the modeling
domain is of interest. One way to enhance computational efficiency in such cases is to use one‐
way nesting, where BCs for a nested grid are extracted from the parent grid, and so subsequent
runs with sensitivities are performed without outer grids. However, differences between the
two nesting schemes (i.e. 1‐way vs. 2‐way) sometimes cause discrepancies in the model results.
An alternative approach is to run the full 2‐way nested model while “turning off” sensitivity
calculations outside nested grids of interest.
CAMx provides a feature called “Flexi‐DDM”, which allows the user to turn off sensitivity
calculations for selected grids (normally grids far outside the area of interest) to improve
computational efficiency of DDM runs (at the expense of accuracy). This reduces CPU times but
will not reduce memory requirements. Also, note that turning off sensitivity calculations for
outer grids is only appropriate for certain types of sensitivity calculations: e.g., sensitivity to
master grid BCs cannot be calculated with Flexi‐DDM.

8.2 Running CAMx With DDM and HDDM
The DDM user interface was designed along similar lines to the Source Apportionment (SA) user
interface. This makes it easier to learn how to use both options and promotes consistency in
analyses performed using SA and DDM. DDM is invoked similarly to the other Probing Tools
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

180

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
8. Decoupled Direct Method

within the CAMx control file. In the &CAMx_Control namelist module, the variable
Probing_Tool must be set to either “DDM” to utilize the original first‐order
implementation, or “HDDM” to utilize the high‐order implementation. An additional namelist
module called &DDM_Control must then be provided in the control file to configure the DDM
portion of the model. The additional namelist module is described below. The order of the
variables follow the template available with the source code. An example of the DDM portion
of the CAMx run control file is shown in Figure 8‐1.
Description of DDM Control in the CAMx Run Control File
&DDM_Control

Label for the Probing Tool namelist module that configures the
DDM option; it must begin in column 2

&

Flag ending a namelist; it must be in column 2

DDM_File_Root

Character root output path/filename

DDM_Master_Sfc_Output

Logical flag for master grid surface output (TRUE=DDM file will
be output for all sensitivities, FALSE=DDM file will not be
output)

DDM_Nested_Sfc_Output

Logical flag for nested grid surface output (TRUE=DDM file will
be output for all sensitivities, FALSE=DDM file will not be
output)

DDM_Stratify_Boundary

Logical flag to stratify boundary types (TRUE=separate
sensitivity types will be used for the N, S, E, W, and Top
boundaries, FALSE=a single sensitivity type will be used for all 5
boundaries)

DDM_Number_of_Source_Regions
Integer number of source regions to be tracked. This
must be the same as the number of source areas defined in the
DDM_Source_Area_Map file. This value must be greater
than zero when sensitivity to emissions is requested.
DDM_Number_of_Source_Groups

Integer number of emission groups to be tracked. This
determines the number of emission files that must be supplied
(additional details below). This value must be greater than zero
when sensitivity to emissions is requested.

Number_of_IC_Species_Groups

Integer number of species or species groups in the
initial conditions to be tracked. This number may be between
zero and the number of species being simulated plus four
(allowing for the four species groups VOC, HRVOC, NOX, ALL).

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

181

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
8. Decoupled Direct Method

IC_Species_Groups

Character array (by IC group) names of the species or species
groups in the initial conditions to be tracked. Allowed names
are any species being simulated by the mechanism in use (e.g.,
O3, PAR, NO, etc.) plus the species groups NOX, VOC, HRVOC,
and ALL. It is permissible to specify both a species and a group
containing that species, e.g., both NO and NOX. Each name
may have up to 10 characters. Note that if you select a species
that is not present on the IC file provided, the initial sensitivities
for that species will be set to zero. This variable may be left
blank if the number of initial condition species groups is zero.

Number_of_BC_Species_Groups

Integer number of species or species groups in the
boundary conditions to be tracked. This number may be
between zero and the number of species being simulated plus
four (allowing for the four species groups VOC, HRVOC, NOX,
ALL).

BC_species_Groups

Character array (by BC group) names of the species or species
groups in the boundary conditions to be tracked. See
description for IC_Species_Group above.

Number_of_EM_Species_Groups

Integer number of species or species groups in the
emissions to be tracked. This number may be between zero
and the number of species being simulated plus four (allowing
for the four species groups VOC, HRVOC, NOX, ALL).

Emis_Species_Groups

Character array (by emissions group) names of the species or
species groups in the emissions to be tracked. See description
for IC_Species_Group above.

Number_of_Rate_Const_Groups

Integer number of reaction rate sensitivity groups to be
tracked. This number may be zero.

Rate_Const_Groups

Character string containing each reaction rate sensitivity group
name and reaction numbers that belong to the group. Group
name and reaction numbers are separated by colon (:) and each
reaction number is separated by comma (,).

Number_of_HDDM_Sens_Groups Integer number of second‐order sensitivity groups to be tracked
(additional details below). This number may be zero.
HDDM_parameters

Character array names of the first‐order sensitivity parameters
to which second‐order sensitivity is computed. The naming of
the first‐order parameters is the same as the long name of
sensitivities with the first 4 characters omitted (see DDM
sensitivity naming conventions/formats below). For each
HDDM sensitivity group, two first‐order parameters are
required (the same can be used twice). All the first‐order
parameters must be included in the modeling.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

182

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
8. Decoupled Direct Method

DDM_Receptor_Definitions

Character input DDM receptor definition path/filename. (This is
an optional file).

DDM_Source_Area_Map

Character array (by CAMx grid) input DDM source area
definition path/filename (required for master grid, optional for
nested grids). Source regions are defined using a map in the
same format as the integer SA source area map (Section 7);
fractional source area maps are not supported by DDM. Unlike
SA, DDM does not require that all parts of the modeling domain
be tracked, therefore it is permissible to define an area
numbered zero in the source area map (emissions from those
areas will not be tracked). The non‐zero source region numbers
must be between 1 and the number of regions.

DDM_PT_Override

Logical flag to allow point source override (TRUE = look for and
use the point source override flags in sector‐specific point
source files, FALSE = ignore point source override flags)

DDM_Calc_Grid

Logical array containing Flexi‐DDM flag for each grid (.TRUE. =
calculate sensitivities in the grid; .FALSE. = do not calculate
sensitivities in the grid).

DDM_Initial_Conditions

The name of the sensitivity initial condition file. Leave the file
name blank for restart days or if sensitivity to initial conditions
is not being calculated.

DDM_Boundary_Conditions

The name of the sensitivity lateral boundary condition file.
Leave the file name blank if sensitivity to boundary conditions is
not being calculated.

DDM_Master_Restart

Character input master grid DDM restart path/filename
(ignored if Restart=FALSE)

DDM_Nested_Restart

Character input nested grid DDM restart path/filename
(ignored if Restart=FALSE or Number_of_Grids=1)

DDM_Points_Group

Character array (by source group) input DDM elevated point
source emissions path/filename (optional, ignored if
Point_Emissions=FALSE)

DDM_Emiss_Group_Grid

Character array (by source group, by CAMx grid) input DDM
gridded emissions path/filename (optional, ignored if
Gridded_Emissions=FALSE)

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

183

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
8. Decoupled Direct Method

&DDM_Control
DDM_File_Root
DDM_Master_Sfc_Output
DDM_Nested_Sfc_Output
DDM_Stratify_Boundary
DDM_Number_of_Source_Regions
DDM_Number_of_Source_Groups

=
=
=
=
=
=

'./DDM_output/CAMx.020604’,
.true.,
.true.,
.false.,
4,
2,

Number_of_IC_Species_Groups
IC_Species_Groups(1)
Number_of_BC_Species_Groups
BC_species_Groups(1)
Number_of_EM_Species_Groups
Emis_Species_Groups(1)
Emis_Species_Groups(2)
Number_of_Rate_Const_Groups
Rate_Const_Groups(1)
Number_of_HDDM_Sens_Groups
HDDM_parameters(1,1)
HDDM_parameters(1,2)
HDDM_parameters(2,1)
HDDM_parameters(2,2)
HDDM_parameters(3,1)
HDDM_parameters(3,2)

=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=

1,
'O3',
1,
'O3',
2,
'NOX',
'VOC',
1,
'RXN1: 120,121,122',
3,
'EM0201NOX_',
'EM0201NOX_',
'EM0201VOC_',
'EM0201VOC_',
'EM0201NOX_',
'EM0201VOC_',

DDM_Receptor_Definitions
DDM_Source_Area_Map(1)
DDM_Source_Area_Map(2)
DDM_PT_Override
DDM_Calc_Grid(1)
DDM_Calc_Grid(2)

=
=
=
=
=
=

'./DDM_input/receptor.cities',
'./DDM_input/source_map.DDM.4areas',
' ',
.false.,
.true.,
.true.,

DDM_Initial_Conditions
DDM_Boundary_Conditions
DDM_Master_Restart
DDM_Nested_Restart

=
=
=
=

'./DDM_input/IC.020603',
'./DDM_input/BC.020604',
'./DDM_output/CAMx.020603.ddm.inst',
'./DDM_output/CAMx.020603.ddm.finst',

DDM_Points_Group(1)
DDM_Points_Group(2)

= ' ',
= './DDM_input/utils.020604',

DDM_Emiss_Group_Grid(1,1)
DDM_Emiss_Group_Grid(1,2)
DDM_Emiss_Group_Grid(2,1)
DDM_Emiss_Group_Grid(2,2)

=
=
=
=

'./OSAT_input/bio.grd1.020604',
'./OSAT_input/bio.grd2.020604',
'./OSAT_input/util.grd1.020604',
'./OSAT_input/util.grd2.020604',

&

Figure 8‐1. Example of DDM inputs in the CAMx control file. CAMx is run with two grids, and DDM is
configured to track emissions from four source regions and two source groups. Sensitivity to ozone
initial and boundary conditions are tracked, while sensitivities to NOx and VOC emissions are tracked.
Sensitivity for a single rate constant group will be calculated involving mechanism reaction numbers
120, 121, and 122. Three groups of second‐order sensitivities to anthropogenic NOx and VOC
emissions (from emissions group 2, source region 1) will be computed (d2/dNOx2, d2/dVOC2 and
d2/dNOxdVOC). No source region map is provided for the nested grid (the region assignments on the
nest are defined by the master grid). Only the group 2 point sources are tracked (no biogenic point
sources are available).
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

184

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
8. Decoupled Direct Method

8.3 DDM Output Files
The output file types for a DDM simulation are described in Table 8‐1. These files have the
same format as corresponding concentration output files, described in Section 3.

Table 8‐1. DDM output file suffix names.
File Name Suffix
.ddm.inst
.ddm.finst
.ddm.grdnn
.ddm.receptor

DDM File Type
Binary master grid instantaneous sensitivity file at end of simulation (used for restart), 3‐
D, all sensitivities, in µmol m‐3 for gases and µg m‐3 for PM.
Binary nested grid instantaneous sensitivity file at end of simulation (used for restart), 3‐
D, all sensitivities, in µmol m‐3 for gases and µg m‐3 for PM.
Binary average sensitivity file for grid nn, 2‐D, surface layer sensitivities only for affected
species requested in the CAMx average file, in ppm for gases and µg m‐3 for PM.
Text hourly average sensitivities at user specific receptor locations. This file is in comma
delimited text format suitable for importing into a spreadsheet.

8.4 DDM Sensitivity Coefficient Names
Each DDM sensitivity coefficient tracks the influence of a species from a specific source (the
influencing species) on a predicted concentration (the affected species). The sensitivity
coefficient names are constructed to show this relationship, as follows:
{Affected Species}{Pollutant Source}{Influencing Species}
This is a lot of information to encode in a name that must conform to the ten character limit
imposed by the binary I/O file formats. Because of this, two naming systems are used in CAMx:



Long Names ‐ these names are easy to read, but since they are more than ten characters
in length they cannot be used in sensitivity coefficient binary output files. If an alternate
I/O format is implemented in the future it may be possible to use the long names on
sensitivity output files.



Short Names ‐ these convey the same information as the long names but require more
practice to learn. They are used in the sensitivity coefficient binary output files.

At the start of each CAMx run a concordance of Long and Short sensitivity coefficient names is
written to the diagnostic output file (.diag file). An example concordance is shown in Figure 8‐
2, and a detailed explanation of the naming convention follows.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

185

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
8. Decoupled Direct Method

Affected
Influencing
Source
Long
Short
Species
Species
Type
Group Region
Name
Name
------------------------------------------------------------------------------NO
ALL
EM
1
2
NO__EM0102ALL_ 0160102ALL
NO2
ALL
EM
1
2
NO2_EM0102ALL_ 0170102ALL
O3
ALL
EM
1
2
O3__EM0102ALL_ 0180102ALL
PAN
ALL
EM
1
2
PAN_EM0102ALL_ 0190102ALL
PANX
ALL
EM
1
2
PANXEM0102ALL_ 0200102ALL
PNA
ALL
EM
1
2
PNA_EM0102ALL_ 0210102ALL
FACD
ALL
EM
1
2
FACDEM0102ALL_ 0220102ALL
FORM
ALL
EM
1
2
FORMEM0102ALL_ 0230102ALL
H2O2
ALL
EM
1
2
H2O2EM0102ALL_ 0240102ALL
HNO3
ALL
EM
1
2
HNO3EM0102ALL_ 0250102ALL
HONO
ALL
EM
1
2
HONOEM0102ALL_ 0260102ALL
IOLE
ALL
EM
1
2
IOLEEM0102ALL_ 0270102ALL
ISOP
ALL
EM
1
2
ISOPEM0102ALL_ 0280102ALL
ISPD
ALL
EM
1
2
ISPDEM0102ALL_ 0290102ALL
MEOH
ALL
EM
1
2
MEOHEM0102ALL_ 0300102ALL
MEPX
ALL
EM
1
2
MEPXEM0102ALL_ 0310102ALL
MGLY
ALL
EM
1
2
MGLYEM0102ALL_ 0320102ALL
AACD
ALL
EM
1
2
AACDEM0102ALL_ 0330102ALL
ALDX
ALL
EM
1
2
ALDXEM0102ALL_ 0340102ALL
CO
ALL
EM
1
2
CO__EM0102ALL_ 0350102ALL
ALD2
ALL
EM
1
2
ALD2EM0102ALL_ 0360102ALL
NTR
ALL
EM
1
2
NTR_EM0102ALL_ 0370102ALL

Figure 8‐2. Example concordance of long and short sensitivity coefficient names from the
CAMx diagnostic output file.
8.4.1 Initial Condition Sensitivity Names
Long Name

NNNNIC____MMMM

where:
NNNN
IC
____
MMMM

Affected species name with trailing underscore to pad blanks
Indicates the sensitivity coefficient is for initial conditions
Four underscores to pad the name to 14 characters
Influencing species name with trailing underscore to pad blanks

Examples:

O3__IC____O3__
HNO3IC____NOX_
ETH_IC____HRVO

Short Name

nnnI___mmm

where:
nnn
I
___
mmm

Affected species number
Indicates the sensitivity coefficient is for initial conditions
Three underscores to pad the name to 10 characters
Influencing species number or name of a species group (NOX,
VOC, HRVOC or ALL).

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

186

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
8. Decoupled Direct Method

Examples:

018I___018 (where O3 is species number 18)
025I___NOX (where HNO3 is species number 25)
042I___HRV (where ETH is species number 42)

8.4.2 Boundary Condition Sensitivity Names
Long Name

NNNNBCRRR_MMMM

where:
NNNN
BC
RRR
_
MMMM

Affected species name with trailing underscore to pad blanks
Indicates the sensitivity coefficient is for boundary conditions
NTH, STH, EST, WST or TOP if stratified by boundary; ALL if not
stratified by boundary
Underscore to pad the name to 14 characters
Influencing species name with trailing underscore to pad blanks

Examples:

O3__BCTOP_O3__
HNO3BCEST_NOX_
ETH_BCALL_HRVO

Short Name

nnnBRRRmmm

where:
nnn
B
RRR
mmm

Examples:

Affected species number
Indicates the sensitivity coefficient is for initial conditions
NTH, STH, EST, WST or TOP if stratified by boundary; ALL if not
stratified by boundary
Influencing species number or name of a species group (NOX,
VOC, HRVOC or ALL)
018BTOP018 (where O3 is species number 18)
025BESTNOX (where HNO3 is species number 25)
042BALLHRV (where ETH is species number 42)

8.4.3 Emissions Sensitivity Names
Long Name

NNNNEMGGRRMMMM

where:
NNNN
EM
GG
RR
MMMM

Affected species name with trailing underscore to pad blanks
Indicates the sensitivity coefficient is for emissions
Emissions group number
Emissions region number
Influencing species name with trailing underscore to pad blanks

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

187

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
8. Decoupled Direct Method

Examples:

O3__EM0101O3__
HNO3EM0201NOX_
ETH_EM0103HRVO

Short Name

nnnGGRRmmm

where:
nnn
GG
RR
mmm

Examples:

Affected species number
Emissions group number
Emissions region number
Influencing species number or name of a species group (NOX,
VOC, HRVOC or ALL)
0180101018 (where O3 is species number 18)
0250201NOX (where HNO3 is species number 25)
0420103HRV (where ETH is species number 42)

8.4.4 Reaction Rate Sensitivity Names
Long Name

NNNNRATE__MMMM

where:
NNNN
RATE
__
MMMM

Affected species name with trailing underscore to pad blanks
Indicates the sensitivity coefficient is for rate constants
Two underscores to pad the name to 14 characters
Reaction rate sensitivity group name with trailing underscore to
pad blanks

Examples:

NO__RATE__RXN1
O3__RATE__R28_

Short Name

nnnRATEmmm

where:
nnn
RATE
mmm
Examples:

Affected species number
Indicates the sensitivity coefficient is for rate constants
Reaction rate sensitivity group number
016RATE001 (where NO is species number 16)
018RATE002 (where O3 is species number 18)

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

188

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
8. Decoupled Direct Method

8.4.5 HDDM Sensitivity Names
Long Name

NNNNHDDMLLLMMM

where:
NNNN
HDDM
LLL
MMM

Affected species name with trailing underscore to pad blanks
Indicates the sensitivity coefficient is second‐order
The index of the first 1st‐order sensitivity parameter in the internal
list of the 1st‐order parameters
The index of the second 1st‐order sensitivity parameter in the
internal list of the 1st‐order parameters

Examples:

NO__HDDM001001
O3__HDDM001002

Short Name

nnnHlllmmm

where:
nnn
H
lll
mmm

Examples:

Affected species number
Indicates the sensitivity coefficient is second‐order
The index of the first 1st‐order sensitivity parameter in the internal
list of the 1st‐order parameters
The index of the second 1st‐order sensitivity parameter in the
internal list of the 1st‐order parameters
016H001001 (where NO is species number 16)
018H001002 (where O3 is species number 18)

8.5 Steps In Developing Inputs And Running DDM
Below is a simple methodological list of steps to follow in setting up and running DDM. The
process is similar among the SA and DDM Probing Tools.
1) Define the source groups and regions that you wish to track. Keep in mind that memory
resources increase dramatically as the number of sensitivities grows. Probing Tool
applications with large numbers of sensitivities, nested grids or grid cells may exceed
available memory.
2) Build an integer source region map (see Section 7) that defines the spatial allocation of
emission sensitivities. For small domains or small number of regions, this can be done by
hand. We suggest using GIS software to develop complex source region maps on large
grids.
3) Process the emissions inventory into the separate source group files that you want to
track (e.g., mobile, area, point, biogenic, etc.).

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

189

www.camx.com

March 2016

4)

5)

6)

7)

8)

CAMx User’s Guide Version 6.3
8. Decoupled Direct Method

a) Consideration of potential source apportionment or sensitivity applications prior to
any emissions processing can be very beneficial so that files by group are available
for later use.
b) Elevated point sources will automatically be assigned to the source region in which
they reside. However, you may override the region to which each individual point
source is assigned (see the definition of kcell in Section 3, Elevated Point Source
File). A point source region does not need to be defined in the source region map,
e.g., you could have a map with two regions that split the domain in half, with a
third region assigned arbitrarily to represent elevated point sources only.
Edit the CAMx control namelist file (Section 2).
a) Set the Probing_Tool variable to the technology you wish to use (DDM, HDDM).
This will activate the &DDM_Control namelist module.
b) Edit or add the &DDM_Control namelist module (described earlier). Provide the
required information, including:
 output paths
 whether to stratify boundary conditions
 number of source regions
 number of source groups
 numbers and names of IC, BC, emissions, rate constant, and HDDM groups
 receptor definitions
 IC/BC input files
 list of input emission files by group.
Configure the CAMx source code to define the number of tracers, and build an
executable. This will ensure that you have sufficient memory for the Probing Tool
application.
a) Edit the file Includes/camx.prm
b) Change the parameters MXTRSP and MXFDDM, following the instructions provided
in the file. CAMx is distributed with MXTRSP = 1 and MXFDDM = 1 to minimize
memory requirements for standard applications of the model. If you run DDM with
an insufficient value, the model will stop and tell you the required value of MXTRSP
and MXFDDM for your application.
c) Execute the CAMx Makefile to build an executable program (Section 2).
Run CAMx and review the diagnostic output files to ensure that the model is correctly
interpreting and running the Probing Tool configuration that you have specified. Ensure
that CAMx is generating the proper output files that you are expecting. Review the table
of concordance of long and short sensitivity coefficient names.
Review gridded tracer fields using commonly available plotting programs. Utilities such as
PAVE or Verdi will read Probing Tool files directly. Use of any other software may require
specialized re‐formatting procedures.
Probing Tool gridded tracer output files are written in the same Fortran binary format as
the regular CAMx concentration output files. You can post‐process gridded output fields
using any software that reads CAMx files, or you can adapt those programs or build your
own software to generate specialized analysis and graphical products.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

190

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
9. Process Analysis

9. PROCESS ANALYSIS
Process Analysis (PA) allows for in‐depth analysis of photochemical model performance by
revealing the contributions from individual physical and chemical processes operating within
the model (Jeffries and Tonnesen, 1994). Using PA, one can more fully understand the complex
interactions between the different processes, explain simulation results within the context of
the model formulation, and improve the design of control strategies.
A conventional model performance evaluation employs statistical and graphical methods to
analyze predicted concentrations against observed concentrations. This answers the basic
question: “How well is the model replicating measurements?” While such comparisons are
necessary to summarize performance, they are far from sufficient to determine whether the
model is adequately representing the real situation. This is because compensating errors
among various model processes can result in predictions that serendipitously agree with limited
observations but for the wrong reasons. In contrast PA provides information on how the
specific model predictions were obtained, which can be interpreted to improve model
performance and/or inform control strategy decisions.

9.1 Process Analysis In CAMx
Three components of PA are implemented in CAMx:
1) Integrated Processes Rate (IPR) analysis. The IPR method provides detailed process rate
information for each physical process in CAMx (i.e., advection, diffusion, deposition,
emissions, and chemistry) for selected grid cells and selected species (Wang, Langstaff,
and Jeffries, 1995). The IPR outputs can be analyzed to determine what processes
governed the model‐predicted concentrations at any time and place. IPR information has
often been plotted as a time series of process contributions for specific cells or groups of
cells. IPR outputs have also been used to check the mass balance in the host model, i.e.,
to determine whether model concentrations are fully explained by the diagnosed process
information or whether unexpected artifacts are occurring. The IPR data are relatively
easy to interpret and can be analyzed using simple tools such as spreadsheets. IPR works
for all gas and PM mechanisms and with PiG. IPR does not work with the ACM2 vertical
diffusion option.
2) Integrated Reaction Rate (IRR) analysis. The IRR method provides detailed reaction rate
information for all reactions in the chemical mechanism for selected grid cells (Jeffries and
Tonnesen, 1994). The IRR data can be analyzed to determine how the chemical changes
occurring in the model are related to the chemical mechanism. For example, by analyzing
rate information over groups of reactions it is possible to quantify chemically meaningful
attributes such as radical initiation rates, radical propagation efficiencies, chain lengths,
etc. Since these analyses tend to be complex, IRR data generally require post‐processing
to be useful. IRR is implemented for the CB05 chemical mechanism, and partially for
CB6r2; it is not implemented for other CB6 variants or for SAPRC07 (see Chapter 5, Table
5‐1).

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

191

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
9. Process Analysis

3) Chemical Process Analysis (CPA). CPA is related to the IRR method but is designed to be
more user friendly and accessible. With CPA, a selection of useful parameters is
calculated from IRR data within CAMx and then output to gridded files for entire grids
(Tonnesen and Dennis, 2000). The gridded CPA files use the same format as the gridded
concentration outputs and can be visualized and processed using standard post‐
processing tools. CPA is implemented for the CB05 chemical mechanism, and partially for
CB6r2; it is not implemented for other CB6 variants or for SAPRC07 (see Chapter 5, Table
5‐1).
9.1.1 Integrated Process Rate Analysis
The specific processes that are reported by IPR are listed in Table 9‐1. This information is
output for each chemical species selected for inclusion in the average concentration output file,
and for each grid cell selected for analysis. The process rates are integrated across each model
output time interval (normally hourly). Taken together, this information provides a complete
description of how the species concentration changed across the output time interval and the
magnitude of all of the processes that caused this change. Information is output in the
concentration units used internally within CAMx (µmole/m3 for gases, µg/m3 for PM species). A
gas conversion factor (ppm/µmole/m3) specific to the grid cell/time period is also output to
allow conversion to mixing ratio (ppm) for comparison of gas species with CAMx average
concentration outputs. For PM species, the conversion factor is always 1. Grid cell volume is
also output to allow aggregation across grid cells.
For most of the process rates listed in Table 9‐1 the interpretation is straightforward, the rate is
simply the concentration change caused by the named process across the output time interval.
The sign convention is such that a positive flux always tends to increase the cell concentration.
Further explanation is provided for several processes below:
Plume‐in‐Grid change: The grid cell concentration change caused by Plume‐in‐Grid puffs
that transferred mass to the grid cell during the output time interval.
Point source emissions: Does not include point sources selected for PiG treatment as
these are reported in Plume‐in‐Grid Change.
Dilution in the vertical: CAMx allows for layer interface heights to change over time
which can lead to a “dilution” term for affected grid cells.
Boundary diffusion: In some cases this term will be zero by definition, namely: the
bottom boundary of surface layer grid cells; the top boundary of top layer grid cells; any
lateral boundary that coincides with a nest boundary.
Dry deposition: This term is zero by definition for all grid cells above the surface layer.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

192

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
9. Process Analysis

Table 9‐1. Process information reported by the IPR option.
IPR Parameter
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26

Process Information
Initial concentration
Gas phase chemistry
Gridded emissions
Point source emissions
Plume‐in‐Grid change
West boundary advection
East boundary advection
South boundary advection
North boundary advection
Bottom boundary advection
Top boundary advection
Dilution in the vertical
West boundary diffusion
East boundary diffusion
South boundary diffusion
North boundary diffusion
Bottom boundary diffusion
Top boundary diffusion
Dry deposition
Wet deposition
Inorganic aerosol chemistry
Organic aerosol chemistry
Aqueous aerosol chemistry
Final concentration
Units conversion
Average cell volume

Unitsa
µmole/m3 (µg/m3)
µmole/m3 (µg/m3)
µmole/m3 (µg/m3)
µmole/m3 (µg/m3)
µmole/m3 (µg/m3)
µmole/m3 (µg/m3)
µmole/m3 (µg/m3)
µmole/m3 (µg/m3)
µmole/m3 (µg/m3)
µmole/m3 (µg/m3)
µmole/m3 (µg/m3)
µmole/m3 (µg/m3)
µmole/m3 (µg/m3)
µmole/m3 (µg/m3)
µmole/m3 (µg/m3)
µmole/m3 (µg/m3)
µmole/m3 (µg/m3)
µmole/m3 (µg/m3)
µmole/m3 (µg/m3)
µmole/m3 (µg/m3)
µmole/m3 (µg/m3)
µmole/m3 (µg/m3)
µmole/m3 (µg/m3)
µmole/m3 (µg/m3)
ppm/(µmole/m3) (N/A)b
m3

a Units in the parentheses are for PM species.
b Unit conversion factor for PM species is always 1.

9.1.2 Integrated Reaction Rate Analysis
IRR provides the integrated rate of each gas‐phase chemical reaction in units of ppm hr‐1 for
each grid cell selected for process analysis. Reaction rates are accumulated (integrated) within
the chemistry solver at the time steps being used to solve the chemical equations, and output
at the CAMx output time interval (usually 1 hour).
9.1.3 Chemical Process Analysis
The CPA method calculates a pre‐determined set of parameters as listed in Table 9‐2. The CPA
parameters are calculated for all grid cells in either the surface layer or all layers. The selection
between surface layer or all layer CPA outputs is determined by the “3‐D average file” flag
specified in the CAMx Control File (see Section 2). This is based on the premise that 3‐D CPA
information will be interpreted in conjunction with 3‐D concentration fields.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

193

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
9. Process Analysis

Table 9‐2. Chemical Process Analysis (CPA) variables calculated in CAMx for the CB05 and
CB6r2 mechanisms. Concentrations are ppb; production and destruction are ppb/hr;
photolysis rates are hr‐1, ratios are unitless.
CB05
CB6r2
Description
Ozone and Oxidant Production and Loss
OxProd
OxProd
Production of Ox = Ozone + NOy‐NO
OxLoss
OxLoss
Destruction of Ox
PO3_net
PO3_net
Net ozone produced
PO3_VOCsns
PO3_VOCsns Net ozone produced under VOC sensitive conditions
PO3_NOxsns
PO3_NOxsns Net ozone produced under NOx sensitive conditions
Ratio of H2O2 produced / HNO3 produced. Greater than 0.35 means NOx
PH2O2_PHN3
PH2O2_PNH3
sensitive ozone production
O3_dest
O3_dest
Ozone destruction by chemical reactions
Radical Initiation
OH_new
New OH produced (initiated)
HO2_new
New HO2 produced
HOx_new
New HOx (HOx = OH+HO2) produced
newOH_O1D
Production of OH from ozone photolysis
newOH_HONO
Production of OH from HONO photolysis
nOH_O3_OLE
Production of OH from ozone‐alkene reactions
nwHO2_HCHO
Production of HO2 from formaldehyde photolysis
RO2_new
New RO2 produced
Radical Propagation
OHw_CO
OHw_CO
OH reacted with carbon monoxide
OHw_CH4
OHw_CH4
OH reacted with methane
OHw_ECH4 OH reacted with locally emitted methane
OHw_ETHA
OHw_ETHA
OH reacted with alkanes
OHw_PAR
OHw_PAR
OHw_PRPA OH reacted with propane
OHw_BENZ OH reacted with benzene
OHw_TOL
OHw_TOL
OH reacted with toluene and mono‐substituted aromatics
OHw_XYL
OHw_XYL
OH reacted with xylenes and poly‐substituted aromatics
OHw_ETH
OHw_ETH
OH reacted with ethene
OHw_ETHY OH reacted with ethylene
OHw_OLE
OHw_OLE
OH reacted with terminal alkenes (R–HC=CH2, e.g. propene)
OHw_IOLE
OHw_IOLE OH reacted with internal alkenes (R–HC=CH–R, e.g. 2‐butene)
OHw_ISOP
OHw_ISOP OH reacted with isoprene
OHw_TERP
OHw_TERP OH reacted with terpenes
OHw_all_HC
OHw_all_HC OH reacted with all organic compounds (including CO)
ISOPwOx
Isoprene reacted with O3, NO3 and O(3P)
TERPwOx
Terpenes reacted with O3, NO3 and O(3P)
OH_rctd
OH_rctd
Total OH reacted
HO2_rctd
Total HO2 reacted
HOx_rctd
Total HOx reacted
RO2_rctd
Total RO2 reacted
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

194

www.camx.com

March 2016

CB05
OHfromHO2
Y_OHperHO2
OH_term
HO2_term
HOx_term
RO2_term
HOx_CL
HCHOp_eth
HCHOp_ole
HCHOp_iole
HCHOp_terp
HCHOp_isop
HCHOp_ispd
HCHOp_Tot
HNO3_OHNO2
HNO3_NO3HC
HNO3_N2O5
PANprodNet
PANlossNet

RNO3_prod
NOxrecycl
NOw_HO2
NOw_RO2s
NOw_RCO3s
J_NO2
J_O3O1D
J_CLDADJ
OH
HO2

CAMx User’s Guide Version 6.3
9. Process Analysis
CB6r2

Description
OH produced from reactions of HO2
Yield of OH per HO2 reacted (= OHfromHO2 / HO2_rctd)
Radical Termination and HOx Chain Length
OH terminated
HO2 terminated
HOx terminated
RO2 terminated
HOx chain length ( = HOx_rctd / {2 x HOx_new} )
Formaldehyde Production
Formaldehyde produced from ethene
Formaldehyde from terminal alkenes (R–HC=CH2, e.g. propene) in the first
generation of products
Formaldehyde from internal alkenes (R–HC=CH–R, e.g. 2‐butene) in the first
generation of products
Formaldehyde from terpenes in the first generation of products
Formaldehyde from isoprene in the first generation of products
Formaldehyde from isoprene daughter products (isoprod, methacrolein and
methylvinylketone)
Total formaldehyde produced
NOy Reactions
HNO3_OHNO2 Nitric acid produced from OH reacting with NO2
HNO3_NO3HC Nitric acid produced from NO3 reacting with organics
HNO3_N2O5 Nitric acid produced from N2O5 reacting with water
Net PAN produced
Net PAN destroyed
NTR1_prod Organic nitrate production
NTR2_prod Organic nitrate production
INTR_prod Organic nitrate production
NTR1toNTR2 Organic nitrate conversion
INTRtoNTR2 Organic nitrate conversion
Organic nitrates (RNO3) produced
NOxrecycl
Nitrates (HNO3 and RNO3) recycled to NOx
NO reacted with HO2 (forming NO2)
NO reacted with RO2 (forming NO2)
NO reacted with RCO3 (forming NO2)
Photolysis
J_NO2
NO2 photolysis rate
J_O3O1D
O3 photolysis rate to O(1D) atoms
Cloud/haze adjustment factor
Radical Concentrations
OH radical concentration
HO2 radical concentration

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

195

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
9. Process Analysis

9.2 Running Process Analysis
PA can be used with most of the physical options available for the “core” CAMx model, e.g., the
various advection and chemistry mechanisms/solvers. However, PA cannot be used at the
same time as the other CAMx “Probing Tool” options (e.g., SA, DDM, or RTRAC) because the
Probing Tools share internal data structures to minimize the total memory resources required
by CAMx. IPR cannot be used with the ACM2 diffusion option.
PA is invoked similarly to the other Probing Tools within the CAMx control file. In the
&CAMx_Control namelist module, the variable Probing_Tool must be set to either “PA”
(generates all PA output), “IPR”, or “IRR”. Table 9‐3 summarizes the types of process analysis
performed for each keyword and the output files that are produced.

Table 9‐3. Process analysis keywords and associated CAMx output files.
Process Analysis Key Word
IPR

IRR

Output
PA

Filename

Yes

No

Yes

*.ipr

No

Yes

Yes

*.irr

No

Yes

Yes

*.cpa.grdnn

File
Contains

Integrated process rate (IPR)
information for all selected cells
Integrated reaction rate (IRR)
information for all selected cells
Chemical process analysis
(CPA) parameters for grid nn

An additional namelist module called &PA_Control must then be provided in the control file
to configure the PA portion of the model. The additional namelist module is described below.
The order of the variables follow the template available with the source code. An example of
the PA portion of the CAMx run control file is shown in Figure 9‐1.
The rules for defining PA sub‐domains are as follows:
1)
2)
3)
4)

They must be contained within a single CAMx grid;
They may not include cells that contain a nested grid;
They may contain as few as 1 grid cells;
They may contain up to all of the grid cells in a CAMx grid provided that this does not
violate the second rule;
5) They may intersect or overlap – the same grid cell may be in several process analysis
domains.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

196

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
9. Process Analysis

Description of PA Control in the CAMx Run Control File
&PA_Control

Label for the Probing Tool namelist module that configures the
PA option; it must begin in column 2

&

Flag ending a namelist; it must be in column 2

PA_File_Root

Character root output path/filename

Number_of_PA_Domains

Integer number of PA analysis domains to be evaluated during
the simulation.

Within_CAMx_Grid

Integer array (by PA domain) pointer into the CAMx grid within
which the PA domain exists (1=master grid, etc.). Use the CAMx
internal grid number reported in the *.diag file. Note that this
may differ from the nest order provided by the user in the
CAMx control file.

PA_Beg_I_Index

Integer array (by PA domain) grid column containing western
edge of PA domain.

PA_End_I_Index

Integer array (by PA domain) grid column containing eastern
edge of PA domain.

PA_Beg_J_Index

Integer array (by PA domain) grid row containing southern edge
of PA domain.

PA_End_J_Index

Integer array (by PA domain) grid row containing northern edge
of PA domain.

PA_Beg_K_Index

Integer array (by PA domain) grid layer containing bottom of PA
domain.

PA_End_K_Index

Integer array (by PA domain) grid layer containing top of PA
domain.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

197

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
9. Process Analysis

&PA_Control
PA_File_Root

= 'CAMx.OTAG.950707.PA',

Number_of_PA_Domains
Within_CAMx_Grid(1)
PA_Beg_I_Index(1)
PA_End_I_Index(1)
PA_Beg_J_Index(1)
PA_End_J_Index(1)
PA_Beg_K_Index(1)
PA_End_K_Index(1)

=
=
=
=
=
=
=
=

2,
1,
8,
12,
9,
13,
1,
5,

Within_CAMx_Grid(2)
PA_Beg_I_Index(2)
PA_End_I_Index(2)
PA_Beg_J_Index(2)
PA_End_J_Index(2)
PA_Beg_K_Index(2)
PA_End_K_Index(2)

=
=
=
=
=
=
=

2,
107,
110,
78,
82,
1,
7,

&

Figure 9‐1. Example section of a CAMx control file specifying options for Process Analysis.
9.2.1 Setting CAMx Parameters
PA stores information in data structures that are dimensioned using Fortran parameter
statements. These parameters must be large enough to accommodate the PA configuration
specified in the CAMx control file. If one of these parameters is exceeded CAMx will stop with
an error message stating that a parameter must be changed and the model recompiled. It is
always a good idea to do a complete rebuild (use the Unix command “make clean”) when a
parameter is changed. The parameters that may need to be changed are in two include files,
“procan.inc” and “camx.prm”.
procan.inc
MXPADOM – The maximum number of Process Analysis domains.
MXPACEL – The maximum number of Process Analysis cells over all domains
camx.prm
MXTRSP – This parameter defines gridded data structures that are used by several
probing tools. For PA the data structures store chemical process analysis (CPA)
variables, so MXTRSP must be set to at least the value of MXCPA (set in procan.inc)
which is 99.
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

198

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
9. Process Analysis

9.2.2 Output File Formats
CAMx may output up to four files containing PA information according to the PA option
selected (see Table 9‐3). All of these files are in Fortran binary format to conserve disk space.
Two files (the *.ipr and *.irr files) contain information for just the grid cells selected for
PA. The formats for these files are not described here since two Fortran post‐processor
programs are provided to extract data from these files in a comma‐delimited text format. The
files containing CPA information (the *.cpa.grdnn) are gridded files covering the same area
as the regular model average files (*.avrg.grdnn). The gridded CPA files have the same
format as a regular model average file as described in Section 3.

9.3 Postprocessing
Two post‐processors are provided to read the binary *.ipr and *.irr output files and
extract PA data for further analysis. The CPA output files can be visualized directly.
9.3.1 IPR Output Files
The Fortran program “ext_ipr” extracts IPR data from one or more CAMx *.ipr binary files
and reformats the data to comma delimited text format (.csv) suitable for subsequent
analysis (e.g., using spreadsheets). The “ext_ipr” program performs the following tasks:



Reads and outputs the descriptive header of the *.ipr file;



Optionally combines data from several consecutive *.ipr files to provide multi‐day
output;



Selects data for an individual cell within a PA sub‐domain or aggregates data over multiple
cells within a PA sub‐domain;



Outputs the selected IPR data in .csv format in either ppb or molar units for gas species;
PM species are in either g/m3 or mass units.

A sample script to run the “ext_ipr” program is provided with its source code, and the script
includes a description of how to use the program.
The “ext_ipr” program can combine IPR information across several cells. This is useful for
analyzing the contributions of model processes to a geographic area that spans multiple cells
and layers (e.g., an urban area). For simplicity, the multi‐cell area must be defined as a
rectangular box. The capability of aggregating IPR information across vertical layers is
particularly important during the day because vertical columns of cells within the mixed layer
become strongly coupled on time scales shorter than one hour. Thus, if the process
contributions for a surface grid cell are analyzed during the day vertical diffusion will often
completely dominate all other processes. In this situation, it is more informative to analyze a
column of cells extending from the surface to the approximate height of the mixed layer. When
the “ext_ipr” program aggregates information across grid cells it accounts for differences in
cell volume. If the output for aggregate cells is requested in ppb units, the output from CAMx
in micromole/volume units is converted to ppb using the volume‐weighted average units
conversion factor for the cells being aggregated.
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

199

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
9. Process Analysis

One useful approach to analyzing IPR data is to plot the contributions of several processes as a
time series. Figure 9‐2 presents an example as an illustration of how PA can be used.

Hourly PSO4 Change from Different Processes in Chicago Area.
Run = postproc_test
Grid cells used from grid number 1: (43, 47) to (52, 56) using layers 1 to 5
Top Boundary
Gas-Phase Chemistry
East Boundary

Deposition
Heterogeneous Chemistry
South Boundary

Emissions
West Boundary
North Boundary

1

Change in PSO4 (ug/m3)

0.5

0

-0.5

-1

-1.5
Jun 13, 2002

Jun 14, 2002

Figure 9‐2. Example IPR time series analysis for PSO4; lateral boundary and chemistry terms
are not aggregated.

9.3.2 IRR Output Files
The Fortran program “ext_irr” extracts IRR data from one or more CAMx *.irr binary files
and reformats the data for subsequent analysis. The “ext_irr” program performs the
following tasks:



Reads and outputs the descriptive header of the *.irr file;



Optionally combines data from several consecutive *.irr files to provide multi‐day
output;



Selects data for an individual cell or multiple cells within a PA sub‐domain;



Optionally, outputs the selected IRR data to a .csv format text file;



Optionally, outputs the selected IRR data to a UAM average format binary file.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

200

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
9. Process Analysis

The text IRR data are suitable for subsequent analysis (e.g., using spreadsheets). The binary IRR
output from “ext_irr” can be displayed using any post‐processing software that can display
CAMx average concentration outputs.
9.3.3 CPA Output Files
CPA results are output from CAMx as gridded files in the same format as the average
concentration files, and therefore can be visualized using any post‐processing software that can
display CAMx concentration outputs. These files may optionally include just the surface layer
or all layers according to how the “3‐D output” flag is set in the CAMx control file. Surface layer
species concentrations reflect the balance between several model processes including surface
emissions/deposition, vertical mixing, and chemical reactions in surface (and possibly aloft) grid
cells. In contrast, the CPA output data are grid cell specific and reflect chemical change in single
grid cells. The fact that vertical mixing tends to average species concentrations over multiple
layers whereas CPA variables are layer specific may complicate and bias the interpretation of
CPA results. A solution is to place both the concentrations and CPA variables on a comparable
basis by averaging them over all layers within the planetary boundary layer.
A post‐processor (VERTAVG) was developed to average CPA variables and concentrations over
multiple layers contained within the depth of the planetary boundary layer (PBL). The PBL
depth varies in space and time according to the strength and vertical extent of turbulent
mixing. Vertical turbulent mixing is specified for CAMx by the input diffusivity (Kv) fields. The
VERTAVG processor reads CAMx Kv, height, temperature, and pressure input files, and then
calculates the PBL depth for each grid column at each hour. VERTAVG also reads a CAMx 3‐D
output file of CPA variables (or species concentrations) and calculates air‐mass weighted PBL
values for each grid column at each hour. The output from VERTAVG is a 2‐D file in average file
format where the single layer represents the PBL average values rather than surface layer
values.
VERTAVG appends two extra variables to the file being processed to aid with interpretation and
to make clear that the data have been vertically averaged. The added variables are:



PBL_Z is the diagnosed height of the PBL for each grid column.



PBL_I is the layer index of the top layer within the diagnosed PBL for each grid column.

CAMx must be run using the option to create 3‐D species concentration (and therefore CPA)
output files in order to use VERTAVG.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

201

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
10. Reactive Tracers

10. REACTIVE TRACERS
The CAMx reactive tracer algorithm provides a flexible approach for simulating the emission,
dispersion, chemistry, and deposition of multiple trace gas and particulate tracers on the
computational grid(s) and within the IRON PiG treatment. Originally developed to model air
toxics, it was extended and generalized to be able to track a variety of user‐defined organic and
inorganic species.
Reactive tracers are carried separately from the core model photochemical/PM chemistry
mechanisms, and thus can be used to address a variety of issues, separately or in combination:



Reactive and inert gaseous and particle air toxic compounds;



Chemical decay of individual VOC compounds into multiple generations of daughter
products;



Source tagging of primary emitted inert and reactive compounds from specific source
types/classes, or from individual stacks, facilities and/or complexes.

The reactive tracer algorithm is implemented as a CAMx “Probing Tool” and thus shares model
data structures with other Probing Tools such as the SA, DDM and PA. This streamlines the
CAMx code, improves efficiency, and maximizes consistency with the core model since it allows
reactive tracer calculations for emissions, transport, and deposition to use the existing CAMx
algorithms. However, this means that reactive tracers cannot be used simultaneously with
other Probing Tools.
The reactive tracer implementation employs two approaches to define tracer chemistry. The
original approach, referred to as RTRAC, allows tracers to decay and form multiple generations
of daughter products through photolysis and user‐specified thermal reactions with ozone and
radicals (OH, NO3) that are extracted from the core model’s gas‐phase chemistry (CB or SAPRC).
A second approach, referred to as the RTRAC Chemical Mechanism Compiler (RTCMC), allows
the user to externally define a full chemistry mechanism with no limits on complexity (within
available computer resources). RTCMC can also access any gas‐phase concentrations from the
core gas‐phase mechanism as well. Neither option allows chemistry for particulate tracers.

10.1 Description of RTRAC
Reactive tracers are defined for each CAMx run by providing an RTRAC chemistry parameters
file similar to that used for the core model. The example in Figure 10‐1 illustrates an example
RTRAC air toxics application (ENVIRON, 2002; Morris et al., 2003). The number and names of
the tracers are arbitrary; i.e. information on the tracer species’ chemical identities, structure,
reaction pathways, and kinetics are kept separate from the core model. Consistent with the
chemistry parameters files used for the core model’s photochemistry, the physical
characteristics for each reactive tracer must be specified for deposition calculations, and their
reaction pathways and rates must be defined.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

202

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
10. Reactive Tracers

CAMx Version
|VERSION6.3
Description
|Example RTRAC Chemsitry for Toxics
No of gas tracers |6
No of aero tracers |8
No photolysis rxns |4
No thermal rxns
|12
Gas Tracers
No. Name
P/S
SNAM lower bnd
H-law
T-fact
1 PACET
PRIM
1.00E-12 6.30e+03
-6492.
2 HCHO
PRIM
1.00E-12 6.30e+03
-6492.
3 BENZ
PRIM
1.00E-12 1.80e-01
0.
4 BUTA
PRIM
1.00E-12 1.00e-02
0.
5 SACET
SEC
ALD2
1.00E-12 6.30e+03
-6492.
6 SFORM
SEC
FORM
1.00E-12 6.30e+03
-6492.
Aero Tracers
No. Name
lower bnd
Density
Low cut Upper cut
7 DSLF
1.00E-09
1.5
0.10
2.50
8 ECF
1.00E-09
1.5
0.10
2.50
9 CRF
1.00E-09
1.5
0.10
2.50
10 CR6F
1.00E-09
1.5
0.10
2.50
11 DSLC
1.00E-09
1.5
2.50
10.00
12 ECC
1.00E-09
1.5
2.50
10.00
13 CRC
1.00E-09
1.5
2.50
10.00
14 CR6C
1.00E-09
1.5
2.50
10.00
Photolysis reactions
Toxic
Rxn #
Factor
PACET
108
1.0
SACET
108
1.0
HCHO
98
1.6
SFORM
98
1.6
Thermal reactions and rates
Toxic
React A(ppm-1min-1)
Ea(K)
B
Tref
PACET
OH
8.2015E+03 -3.1099E+02
0.0
300.0
PACET
NO3
2.0689E+03 1.8599E+03
0.0
300.0
HCHO
OH
1.6699E+03 -6.4815E+02
2.0
300.0
HCHO
NO3
4.1377E+03 2.5161E+03
0.0
300.0
BENZ
OH
3.6944E+03 1.9978E+02
0.0
300.0
BUTA
OH
2.1871E+04 -4.4787E+02
0.0
300.0
BUTA
O3
4.8766E+01 2.5000E+03
0.0
300.0
BUTA
NO3
2.1871E+04 1.4890E+03
0.0
300.0
SACET
OH
8.2015E+03 -3.1099E+02
0.0
300.0
SACET
NO3
2.0689E+03 1.8599E+03
0.0
300.0
SFORM
OH
1.6699E+03 -6.4815E+02
2.0
300.0
SFORM
NO3
4.1377E+03 2.5161E+03
0.0
300.0

Molwt
44.00
30.00
78.00
54.00
44.00
30.00

Reactvty
0.0
0.0
0.0
0.0
0.0
0.0

Rscale
1.0
1.0
1.0
1.0
1.0
1.0

Figure 10‐1. Example RTRAC chemistry input file for modeling specific toxic species.

The structure of the RTRAC tracer definition provides complete flexibility in the selection of the
compounds and tracers to be included in each analysis. The user is able to easily alter or
expand the compounds as needed.
For gas species, the required deposition parameters are the Henry’s Law constant and
molecular weight Mg, the latter of which defines a diffusivity parameter according to
M g / M H O . The deposition calculation for gases that react in plant tissue also needs a
2

reactivity parameter that describes whether a species reacts when dissolved inside leaf tissues
(Wesely, 1989). This parameter is intended for modeling the deposition of reactive species,
such as ozone, and should be set to zero for air toxics. The deposition calculation for gasses
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

203

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
10. Reactive Tracers

also uses an “Rscale” factor to adjust the surface resistance. This is used to set the surface
resistance to zero for strong acids (e.g., HNO3) and should be set to 1.0 for modeling air toxics.
The deposition calculation for particles requires the particle density and size associated with
each species. The particle size is calculated as the geometric mean of the lower and upper cut
points (see Figure 10‐1). If possible, the particle size and density should be based on the
measured size of particles associated with each RTRAC species (e.g., for a compound associated
with soot particles, use a density and size representing the soot).
The emission rates for the RTRAC species are provided by an extra set of emission files (surface
and/or point source). Emissions of gases are in moles per time period (normally moles/hour),
whereas particles are in grams per time period. The emissions file format is the same as for a
regular CAMx emissions file, as described in Section 3.
10.1.1 RTRAC Gas‐Phase Chemistry
The RTRAC chemistry calculations use a special chemistry module. Chemistry may be modeled
for primary and secondary gas species, meaning that tracers can be formed from the decay of
primary tracers or from the decay of host model species (e.g., secondary formaldehyde). The
chemical decay of gaseous tracers can account for thermal reactions with ozone (O3), hydroxyl
radical (OH) and nitrate radical (NO3), as well as photolysis. The algorithms are coded so that all
chemical decay pathways are zero by default and only become non‐zero if decay rates are
explicitly specified in the input file (see Figure 10‐1). The example RTRAC chemistry input file in
Figure 10‐1 shows how thermal reactions are specified by naming the tracer and oxidant, and
providing reaction rate parameters. Note that the RTRAC chemical reaction rates depend on
the rates and parameters provided in the RTRAC input file, and not the rates in the host model
chemical mechanism; however the host model does provide the oxidizing species
concentrations (i.e., O3, OH, and NO3).
10.1.1.1 Thermal Reactions
Thermal reactions with oxidants are modeled as second order reactions:

R  k tracer oxidant 
where R is the decay rate and the rate constant k is defined using the generalized temperature
dependent rate expression:
B

 T 
  Ea 
k  A

 exp
 300 
 T 
The Arrhenius factor (A) must be in units (ppm‐1min‐1), the activation energy (Ea) must be Kelvin
and B is dimensionless. This is the same as expression 3 in Table 3‐3a. Oxidant concentrations
for the decay calculation are obtained from the CAMx photochemical simulation for each grid
cell at each time step. RTRAC can be used with any of the photochemical mechanisms that are
available in the current version of CAMx (see Section 5). Choosing between the core
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

204

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
10. Reactive Tracers

mechanisms will influence the RTRAC chemical decay rates by changing the oxidant
concentrations in the host model.
10.1.1.2 Photolysis
Photolysis reactions are specified by naming the tracer undergoing photolysis and providing a
ratio of the tracer photolysis rate to one of the photolysis reactions in the host photochemical
mechanism. For example, Figure 10‐1 shows that there are both primary and secondary
acetaldehyde reactive tracers (PACET and SACET) and the photolysis rate for both species is set
equal to CB6r2 reaction 108 (photolysis of ALD2), which is based on acetaldehyde. Figure 10‐1
also shows that there are two types of formaldehyde (PFORM and SFORM). Modeling the
photolysis of formaldehyde with RTRAC is complicated by the fact that the CB6r2 mechanism
includes two photolysis reactions for formaldehyde (reactions 97 and 98). The solution shown
in Figure 10‐1 is to model formaldehyde photolysis as 1.6 times the rate of reaction 98. The
CAMx host mechanisms are discussed in Section 5 and are defined by the text chemistry
parameters files (Section 3) and mechanism listings distributed with CAMx and available from
the CAMx web page (www.camx.com).
10.1.1.3 Secondary Species
RTRAC allows for formation of secondary/daughter products related to the chemical decay of
one of the primary tracers. Secondary species can also be subject to chemical decay, just like
primary species, if the user desires. Therefore, the RTRAC chemistry module allows decay
reactions (thermal and photolysis) to be specified for secondary species using the same method
as for primary species. In this manner, concentrations of secondary species are determined by
the balance between chemical production and destruction. RTRAC requires that any secondary
daughter tracers must be specified after their parent tracer in the chemistry parameters input
file.
RTRAC also allows tracers that track the secondary formation of any species that is included in
the host chemical mechanism. For example, in Figure 10‐1 the species SFORM is used to track
secondary formaldehyde, and so SFORM is defined as a secondary species and identified with
the host species FORM. This means that the RTRAC chemistry module will identify the chemical
production of FORM in each grid cell at each time step, and add this chemical production to the
SFORM tracer. Since SFORM is intended to track only secondary formaldehyde, no primary
emissions should be included for SFORM.
10.1.1.4 Chemical Decay Rates for Near‐Source Modeling
The RTRAC algorithm can output hourly chemical decay rates at user‐specified locations to
support external analyses, for example, as input to a Gaussian plume/puff model. The user
provides the locations of each receptor using the CAMx Probing Tools receptor file input
format. Figure 10‐2 displays an example RTRAC receptor input file for the five locations. At
each grid cell, hourly decay rates for each RTRAC compound and every vertical layer are output
and can then be interfaced with a user‐selected plume model.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

205

www.camx.com

March 2016

SINGLE
SINGLE
SINGLE
SINGLE
SINGLE

CAMx User’s Guide Version 6.3
10. Reactive Tracers

CELL
CELL
CELL
CELL
CELL

Test
Test
Test
Test
Test

Cell
Cell
Cell
Cell
Cell

1
1
1
1
1

42
41
39
50
34

44
36
36
43
48

Figure 10‐2. Example RTRAC receptor input file identifying the grid cells with locations where
hourly decay rates will be output for subgrid‐scale point source modeling (see format for SA
receptor file in Table 7‐2).

Separate families of reactive tracer compounds can be simulated by providing separate
emission inputs similarly to SA (see Section 7). Tracking separate families of RTRAC tracers
allows for source apportionment and can be used to avoid double counting when an external
plume model is used to obtain near‐source impacts. For example, separate families of air toxic
tracers can be specified for each point source complex to be modeled by the external plume
model, so that total concentrations could include the local point source impacts (plume model)
plus the regional contributions from all other sources (CAMx RTRAC).

10.2 Description of RTCMC
Like RTRAC, the purpose of RTCMC is to add tracer species to a CAMx “core model” simulation
and have the tracers undergo chemical changes that depend, in part, upon the evolution of
CAMx core model species. The RTCMC approach differs from the original RTRAC approach by
allowing arbitrarily complex chemical reaction schemes, but it is exactly like RTRAC in every
other respect. The current implementation of RTCMC is for gas‐phase reactions, i.e., gas‐phase
tracers reacting with each other and/or gas‐phase host model species. The core model’s
photochemical mechanisms remain intact and separate from the reactive tracer chemistry.
10.2.1 RTCMC Gas‐Phase Chemistry
The RTCMC allows users to input, in a text‐based format, a set of chemical reactions
(mechanism) for certain target species to be treated by the CAMx Reactive Tracer Probing Tool.
RTCMC is an extension of the original RTRAC algorithm that reads (and solves) a completely
independent, user‐defined chemical mechanism for reactive tracers that can utilize
concentrations of any photochemical species from the core model mechanism. Upon startup,
RTCMC compiles information on the chemical mechanism and configures the reactive tracer
chemistry solver. During the model simulation, the RTCMC chemistry solver receives ambient
pollutant information from the core photochemical mechanism and uses this to calculate the
evolution of RTRAC species.
The format of the RTCMC input file is essentially the same as the “IMC” input file format of the
SCICHEM Lagrangian puff model (EPRI, 2000). An example IMC format file is shown in Figure
10‐3. There are four sections in an IMC file that are identified by a keyword at the start of each
section, as follows:
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

206

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
10. Reactive Tracers

#Control
rate_species_units = 'ppm'
rate_time_units = 'min'
solver = 'dlsode'
Jacobian = 'numeric'
#Name, Type, Ambient,
ATol,
Dep, Wet Scav,
MW, Spec Map
O3
A
1.0 1.0E-12
0.0
0.0
0.0
OH
A
1.0 1.0E-12
0.0
0.0
0.0
ATRAC
F
1.0 1.0E-12
0.010
0.0
0.0
BTRAC
F
1.0 1.0E-12
0.001
0.0
0.0
CTRAC
F
1.0 1.0E-12
0.020
0.0
0.0
DTRAC
F
1.0 1.0E-12
0.001
0.0
0.0
ETRAC
F
1.0 1.0E-12
0.030
0.0
0.0
FTRAC
F
1.0 1.0E-12
-0.001
0.0
0.0 NO2
#Table
0
0.
15.
30.
45.
60.
75.
80.
86.
87.
88.
1 4.1590E-04 4.0600E-04 3.7540E-04 3.27E-04 2.6040E-04 9.4990E-05
2.9930E-05 4.8590E-06 8.3030E-08 1.0000E-09
#Equations
1 [ATRAC]
-> (2.0)[BTRAC] ; 0 0.000E-00
2 (1.5)[CTRAC] + [OH] -> (0.5)[DTRAC] ; 1 4.2000E+04
3 [ETRAC] + [O3]
-> [FTRAC]
; 1 1.8000E-02

Figure 10‐3. Example free‐format RTCMC IMC chemistry input file.

#Control
#Species
#Table
#Equations

Configuration information identified by keywords
Names of chemical species and associated data
Photolysis rate data for any photolytic reactions
Chemical reactions and thermal rate constants

The IMC file uses space‐delimited free‐form text format. Leading white space at the start of
any line will be ignored. CAMx reads the IMC file as case insensitive.
The hash symbol (#) before each section keyword marks the start of a section and should be
reserved for this purpose. The four sections should appear in the order shown above. The only
section that may be unnecessary in some cases (i.e., if there are no photolytic reactions) is the
#Table section and guidance on handling this case is provided below.
10.2.1.1 The Control Section
According to the SCICHEM documentation, the #Control section of the IMC file must always
have at least three lines, as follows:
#CONTROL
&CONTROL
&END

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

207

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
10. Reactive Tracers

The first line is the keyword identifying the control section. The second and third lines are
additional keywords denoting the start and end of the control section. Note, however, that
CAMx ignores the &CONTROL and &END lines.
One or more control options may be specified using keywords, inserted one per line, ahead of
the closing &END line, like this:
#CONTROL
&CONTROL
Keyword = ‘option’
Keyword = ‘option’
&END

The keywords used by CAMx and SCICHEM are listed in Table 10‐1 and are case insensitive. Not
all keywords are used by both models, and CAMx will ignore any non‐recognized keyword. A
“=” symbol must separate each keyword and option. The option must be enclosed within
single quotes. In practice, only the first six letters of each keyword and the first three letters of
each option are considered and you may abbreviate accordingly (i.e., keyword = ‘opt’).

Table 10‐1. Keywords, options and default values for the Control section of the IMC file.
Used by
SCICHEM
●
●
●

Used by
CAMx

Rate_species_units

●

●

Rate_time_units

●

●

Keyword
Ambient file
Species_units
Emission_units

●

Solver
Rtol
Atol

●

Jacobian

●
●
●

Options Allowed by CAMx
n/a
n/a
n/a
molecules/cm3 (default)
ppm
seconds (default)
minutes
hours
DLSODE (default)
SLSODE
Rosenbrock
Real number (default = 1.0E‐5)
Real number (default = 1.0E‐18)
Numeric (default)
Algebraic

All CAMx recognized keywords have a default option that will be used if the keyword is omitted,
meaning that the CAMx RTCMC may be run without specifying any keywords provided that that
all other input data (e.g., rate constants) are consistent with the defaults. The allowed keyword
options in Table 10‐1 are discussed below:
Rate_species_units
The concentration units for thermal rate constant expressions.
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

208

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
10. Reactive Tracers

Rate_time_units
The time units for photolysis and thermal rate constant expressions.
Solver
The name of the numerical integrator to be used as the chemistry solver. SLSODE and
DLSODE are, respectively, the single and double precision versions of the Livermore
Solver for Ordinary Differential Equations (Hindmarsh 1983). The Rosenbrock solver is
the double precision RODAS solver (Hairer and Wanner, 1991).
Rtol
The relative error tolerance (convergence criterion) employed for all chemical species by
the chemistry solver.
Atol
The absolute error tolerance (convergence criterion) employed for all chemical species
by the chemistry solver. CAMx does not use species‐specific Atol values that may
appear in the species section, as discussed below.
Jacobian
The chemistry solvers employ a Jacobian matrix of first‐order derivatives of each
chemical species with respect to all species. The Jacobian matrix is constructed
automatically by the RTCMC. This option controls whether the Jacobian is constructed
algebraically or numerically. Both options may be used with the double precision
solvers and numeric may be more efficient. The algebraic option is strongly
recommended for the single precision SLSODE solver (because single precision may be
inadequate for constructing a numeric Jacobian by finite difference).
10.2.1.1.1 Concentration Units
CAMx does not use the Species_units or Emission_units keywords and will ignore
them if they are present. CAMx will output RTCMC species average concentrations in ppm
units. Emissions of RTCMC species must be provided in moles/hour.
10.2.1.1.2 Setting Error Tolerances
All three RTCMC chemistry solvers use the Rtol and Atol parameters specified in the control
section to manage errors in predicted concentrations. CAMx does not use the species‐specific
Atol values that may appear in the species section because it is difficult to select reliable
Atol values for each species. The error (err) in the predicted concentration (con) for species i
should be roughly less than:

err(i) = rtol  con(i) + atol
The combined Rtol and Atol determine accuracy. Setting Atol to zero will result in pure
relative error control. Relative error control has the advantage of being easily understood (the
errors should be smaller than X percent) but suffers the disadvantage of excessive
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

209

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
10. Reactive Tracers

computational resources that may be expended to manage errors in species concentrations
that are essentially zero. Note that RTCMC sets a concentration floor of 1.0E‐16 ppm.
The default settings for Rtol and Atol listed in Table 10‐1 should be generally applicable
because they are conservative and effectively result in pure relative error control. We
recommend against setting Rtol greater than 0.001. Appropriate settings for Atol depend
upon the magnitude of concentration predictions and the need for accurate predictions in high
vs. low concentration areas (e.g., plume centerline vs. out of plume).
Do not request infeasible accuracy from single precision SLSODE by setting Rtol and Atol
smaller than about 1.0E‐7.
10.2.1.2 The Species Section
The species section of the IMC file lists chemical species and associated data. All chemical
species referred to in the equation section must appear in the species section. Extra species
may appear in the species section, but including numerous extra species may cause a run‐time
error by exceeding the memory available for storing species information (if this happens, delete
some of the unused species from the species section).
The first line is the keyword identifying the species section. The following information must be
provided for each listed species:
Name
Species names may be up to 8 characters and must start with a letter. They are case
insensitive. Accurate names are important because other CAMx input data (e.g.,
emissions, boundary conditions) will be matched to RTRAC species by name.
Type
There are four permissible species types identified by first letter: Ambient (A), Fast (F),
Slow (S), and Equilibrium (E). Setting the species type is discussed in more detail below.
Ambient
The ambient value is not used by CAMx, but is used by SCICHEM. Provide a real number
(e.g., 0.0).
Atol
Species specific error tolerances are not used by CAMx, but are used by SCICHEM.
Provide a real number (e.g., 0.0).
Dep
The species deposition velocity in m/s. This deposition velocity will be used for all land
surface types, which is a simplification compared to the CAMx dry deposition scheme.
Provide a real number (e.g., 0.0).
Wet Scav
The wet scavenging coefficient is not used by CAMx. For SCICHEM, this is a washout
ratio. CAMx does not use the washout ratio because this approach is incompatible with
the CAMx wet deposition algorithms. Provide a real number (e.g., 0.0).
MW
The molecular weight is not needed by CAMx. Provide a real number (e.g., 1.0).
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

210

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
10. Reactive Tracers

Spec Map
The core model species name from which to use internally‐calculated dry deposition
velocity for the RTCMC species. This is triggered by a negative “Dep” value in the same
record. Provide a 10‐character string. In the example below, RTCMC tracer O3A will use
deposition velocities calculated by CAMx for the core model species O3:
#Name, Type, Ambient,
ATol,
Dep, Wet Scav,
MW,
O3A
F
1.0 1.000E-08 -3.00E-03
0.0 48.0

Spec Map
O3

10.2.1.2.1 Choosing the Species Type
The Type for each RTCMC species should be set according to: (a) whether the species
concentration should be obtained from the core model or modeled using the RTCMC; and (b)
the most accurate and efficient numerical method for performing chemistry within RTCMC.
All species to be obtained from the core model (e.g., O3, OH, NO, NO2, H2O, M, O2) must be set
to type Ambient. This rule will be enforced by CAMx and, for example, the species O3 must
be set to type A, because it is part of all the core chemical mechanisms.
Species that are solved by the RTCMC may be type F, S or E. The recommended default type
is F (fast) in which case chemistry will be performed using the selected chemistry solver (e.g.,
DLSODE). Species that undergo slow chemical change (lifetime of hours or longer) may be set
to type S (slow) with potential gain in efficiency but some loss in accuracy. Species that
undergo extremely rapid chemical change (lifetime smaller than a second) may be set to type E
(equilibrium) and solved using a steady‐state approximation with some gain in efficiency but
some loss in accuracy. The Rosenbrock solver does not work well with species types S or E.
Equilibrium species may be used effectively with the single precision SLSODE solver to avoid the
need for double precision. You should use types S or E with caution and evaluate both
computational speed and concentration accuracy by comparing against results with using
type F.
10.2.1.3 The Table Section
The table section of the IMC file provides photolysis rates for any photolytic reactions in the
RTCMC mechanism. It must contain at least two lines:
#Table
0 zenith1, zenith2, zenith3, …
The first line is the keyword identifying the table section. The second line must begin with 0
(zero) followed by a list of space‐delimited zenith angles (in degrees) starting with zero degrees
and ascending to the largest angle. If the largest zenith angle specified is less than 90 degrees a
value of 90 degrees is implicitly added to the list. By default, up to 15 zenith angles are allowed
(this may be changed as described under adjustable parameters, below). If the final zenith
angle is not 90 degrees, no more than 14 angles should be listed to allow the 15th angle to be
implicitly set to 90 degrees.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

211

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
10. Reactive Tracers

If the reaction mechanism has no photolytic reactions, include just the first two lines (keyword
followed by zenith angles) in the table section. If the reaction mechanism includes photolytic
reactions, list them one reaction per space‐delimited line after the zenith angles:
reaction_ID, rate1, rate2, rate3, …
The reaction_ID must be the integer identification number of the photolytic reaction
followed by the photolysis rate at each zenith angle, from zero to the largest angle. Photolysis
reactions are first order and have rate units of reciprocal time to be provided in the
Rate_time_units specified in the control section of the IMC file. A photolysis rate of zero
is implicitly assumed at 90 degrees unless you specify otherwise.
A negative “reaction_ID” in the table section causes the photolysis rate to be set to a core
model reaction number, which is set in the position of “rate1”. In the example below,
RTCMC reaction #1 is photolytic and rates will be set according to the CAMx core photolysis
reaction number 9:
#Table
0
-001

0.
9

10.

20.

30.

40.

50.

60.

70.

78.

86.

10.2.1.4 The Equations Section
The equations section of the IMC file lists the chemical reactions and rate constants for the
RTCMC reaction mechanism and must contain at least two lines:
#Equations
reaction_ID [Reactants] > (Stoichiometry) [Products] ; Rate_Constant

The first line is the keyword identifying the equations section and must be followed by at least
one reaction line. Reaction lines list reactions and rate constants and are delimited by white
space and separators. The reaction_ID and the Reactants must be separated by white
space. The Reactants and Products must be separated by a right arrow symbol (the right
arrow may be preceded by characters, e.g., => or −>). The Products and the
Rate_Constant must be separated by a semi‐colon.
The reaction_ID must be an integer value that uniquely identifies each reaction. Reactions
identifiers need not be in order or continuous.
The name of Reactants and Products must be enclosed within square brackets, begin
with a letter, and not exceed 8 characters in length. All species names used in the equations
section must also appear in the species section. Zero to three reactants are allowed. Zero to 20
products are allowed (the maximum is a user adjustable parameter). Reactant and product
names may be preceded by a stoichiometric coefficient enclosed within round brackets. If the
stoichiometric coefficient is omitted it is assumed to be unity.
Rate constants are specified using SCICHEM conventions and must be in the units specified by
the keywords Rate_species_units and Rate_time_units in the control section (the
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

212

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
10. Reactive Tracers

defaults are molecules cm‐3 and seconds, respectively). SCICHEM supports numerous rate
constant expression types and CAMx should interpret all of them correctly, although not all
have been thoroughly tested. Table 10‐2 defines a sub‐set of the SCICHEM rate constant
expression types that are recommended for use with CAMx. The format for specifying rate
expressions is the integer expression type followed by a list of the numerical values required by
that expression type. It is important that rate expressions are defined in units that are
consistent with the reaction order, and Table 10‐3 defines how the reaction order and rate
constant unit dimensions may be determined.

Table 10‐2a. Recommended SCICHEM rate constant expression types for use in CAMx.
Expression Type

Description

Expression

0

Photolysis

k=0

1

Constant

k  k0

2

General temperature
dependence

k  A T  c eB / T 

 G
k 0[M ]
k
F
0

1  k [M ] / k 

3

Troe-type temperature and
pressure dependence

k 0  AT B
k  CT D
F  0.6

 



2 1

G  1  log k 0[ M ] / k  



8

Equilibrium with a previously
defined reaction (kref)

k  k ref A e B / T 

k3[M ]
1 k3[M}/ k2
k0  A e  B / T 
k  C e D / T 

k  k0 
13

Lindemann - Hinshelwood
as used for OH + HNO3

2

k3  E e  F / T 

7

Simple pressure
dependence used for OH +
CO

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

213

k  k0 1  0.6P
www.camx.com

March 2016

CAMx User’s Guide Version 6.3
10. Reactive Tracers

Table 10‐2b. Parameters required by SCICHEM rate constant expression types.
Expression
Type

Parameters
1
0
ko
A
A
kref
A
ko

0
1
2
3
8
13
7

2

3

B
B
A
B

C
C
B
C

4

5

6

E

F

D
D

Table 10‐3. Determining the reaction order and consequent unit dimensions for rate
constants.
Number of Reactants

Concentration Unit
Dimension

Reaction Order

Time Unit Dimension

0

Zero

None

Time‐1

1

First

None

Time‐1

2

Second

Concentration‐1

Time‐1

3

Third

Concentration‐2

Time‐1

The CAMx output “diag” file lists diagnostic information on the mechanism and rate constant
expressions read by from the IMC file. You should review this diagnostic output to ensure that
CAMx correctly read and configured the RTCMC chemistry mechanism.

10.3 Reactive Tracers In IRON PiG
RTRAC/RTCMC calculations for emissions and chemistry have been integrated into the IRON PiG
algorithms. There are two ways in which RTRAC tracers may enter a PiG plume: as primary
emissions from specifically flagged sources within the RTRAC point source file, or by formation
of secondary species from decay of primary plume emissions. There is no entrainment of
tracers from the grid to the plume as this is likely to result in negative tracer concentrations,
especially if the entrained tracer is a secondary product of a host model species (e.g., secondary
formaldehyde). Tracers are assumed to have negligible impact on PiG puff chemistry or oxidant
levels. If the tracer concentration in the plume is high enough to enhance or suppress the
plume oxidant levels, then the photochemical impacts of the tracer can be accounted for by
separately adding the tracer emissions into the host model lumped emissions; e.g., for tracing
high concentrations of propene and butene in a plume, one would track the propene/butene
concentrations using RTRAC tracers but also add CB‐OLE or SAPRC‐OLE1 emissions to the plume
to account for the oxidant impacts. RTRAC checks to ensure that it is reading its own input
point source file. RTRAC and host model point source files must have the same number of
sources in the same order; however, the list of species on each file may be different, and the
sources flagged to receive the PiG treatment may vary. A pre‐processor program was coded to
help prepare consistent RTRAC and host model point source files.
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

214

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
10. Reactive Tracers

Tracers released from PiG sources decay according to the oxidant and photolytic environment
of the plume using user‐supplied chemical rate parameters (as described earlier). Oxidant
concentrations for the decay calculation are obtained from the CAMx PiG incremental
photochemical simulation for each puff at each time step. RTRAC tracers in each puff reactor
are updated based on the total oxidant concentrations for the reactor, i.e., puff increment plus
puff ambient/background. RTRAC enforces a rule that no secondary tracer formation from the
decay of host model species are allowed if IRON PiG is active (e.g., no secondary formaldehyde
tracer formation is allowed with IRON PiG). Secondary tracer production from primary tracer
decay is allowed.
Tracers are transferred from the PiG to the grid using the same approach as for any other host
model species (see Section 6). Tracer concentrations at any point are the superposition of the
grid concentration plus any collocated PiG puffs.
RTRAC optionally employs surface‐layer IRON puff sampling of tracers on a user‐defined
sampling grid (see Section 6). Sampling grids are entirely passive, and intended to provide a
display of the reactive tracer plume concentrations at scales much smaller than typically used
for the finest computational grids (i.e., <1 km).

10.4 Running CAMx With Reactive Tracers
10.4.1 CAMx Control File
RTRAC is invoked similarly to the other Probing Tools within the CAMx control file. In the
&CAMx_Control namelist module, the variable Probing_Tool must be set to “RTRAC” or
“RTCMC”. An additional namelist module called &RT_Control must then be provided in the
control file to configure the RTRAC portion of the model. The additional namelist module is
described below. The order of the variables follows the template available with the source
code. Figure 10‐5 provides an example of the RTRAC control module.

Description of RTRAC Control in the CAMx Run Control File
&RT_Control

Label for the Probing Tool namelist module that configures the
RTRAC option; it must begin in column 2

&

Flag ending a namelist; it must be in column 2

RT_File_Root

Character root output path/filename

RT_Initial_Conditions

Character input master grid RTRAC initial conditions
path/filename (optional, ignored if Restart=TRUE)

RT_Boundary_Conditions

Character input master grid RTRAC boundary conditions
path/filename (optional)

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

215

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
10. Reactive Tracers

RT_Master_Restart

Character input master grid RTRAC restart path/filename
(ignored if Restart=FALSE)

RT_Nested_Restart

Character input nested grid RTRAC restart path/filename
(ignored if Restart=FALSE or Number_of_Grids=1)

RT_Chemistry_Parameters

Character input RTRAC chemistry parameters path/filename, or
RTCMC IMC chemistry definition path/filename

RT_Receptor_Definitions

Character input RTRAC receptor definition path/filename
(optional)

RT_Point_Sources

Character input RTRAC elevated point source emissions
path/filename (optional, ignored if
Point_Emissions=FALSE)

RT_Emiss_Grid

Character array (by CAMx grid) input RTRAC gridded emissions
path/filename (optional, ignored if
Gridded_Emissions=FALSE)

RT_PiG_Sample

Logical sampling grid flag for RTRAC IRON PiG output; sampling
grids are defined in the main &CAMx_Control namelist
(TRUE=sampling grid output will be generated, FALSE=sampling
grid output will not be generated)

&RT_Control
RT_File_Root

= 'CAMx6.test.020614',

RT_Initial_Conditions
RT_Boundary_Conditions
RT_Master_Restart
RT_Nested_Restart

=
=
=
=

' ',
' ',
'CAMx6.test.020613.rt.inst',
'CAMx6.test.020613.rt.finst',

RT_Chemistry_Parameters
RT_Receptor_Definitions
RT_Point_Sources
RT_Emiss_Grid(1)
RT_Emiss_Grid(2)
RT_Emiss_Grid(3)

=
=
=
=
=
=

'CAMx6.chemparam.rtrac_test',
'receptor.rtrac.test',
'pt.rtrac.test',
'emiss.rtrac.36km',
'emiss.rtrac.12km',
'emiss.rtrac.04km',

RT_PiG_Sample

= .true.,

&

Figure 10‐4. Example input of RTRAC options and filenames within the CAMx control file.
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

216

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
10. Reactive Tracers

As with the output for the host model and other Probing Tools, a “root” file name is specified
and suffixes are added depending upon the type of output produced. A separate root name for
RTRAC (and other Probing Tools) allows the user to direct the output to a completely different
path. RTRAC writes several output files that are in Fortran binary format, as described in
Section 3. These include the master and nested grid tracer instantaneous concentration files
(.rt.inst and .rt.finst), the grid‐specific surface tracer average concentration file
(.rt.grdnn). These files are written in the same format as for the regular model species
described in Section 3.
The “RT_Chemistry_Parameter” namelist variable specifies the path/filename of either
the RTRAC chemistry parameters file or the RTCMC IMC chemistry definition file. The choice of
which type of file format is read is set according the main “Probing_Tool” variable (i.e.,
RTRAC or RTCMC).
RTRAC/IRON PiG sampling grids are invoked in the RTRAC namelist by setting a logical flag. If
set to TRUE, the user must provide the number of sampling grids and the grid parameters of
each in the main &CAMx_Control namelist. Sampling grids are set identically to the way
nested grids are specified for the host model, with one exception: there are no vertical levels to
define (sampling grids are currently only 2‐D surface fields). The same rules that apply for the
specification of nested grids holds for the specification of all sampling grids (see Sections 2, 4,
and 6). The “mesh factor” sets the resolution or cell size of the sampling grid relative to the
master grid. The CAMx diagnostic output file provides information on the location and size of
each sampling grid to help ensure proper setup.
10.4.2 User Adjustable Parameters
Once the RTRAC/RTCMC chemistry parameters/definition file is established, the user should be
sure that a sufficient allocation of memory is provided for this Probing Tool. This is done by
examining the main Probing Tool parameter and common block file in
Includes/camx.prm. The parameter MXTRSP should be set to the total number of species
defined in the chemistry parameters file. If sampling grids are to be used, the user should
ensure that sufficient memory is available to define the size of sampling grid arrays. This is also
set in Includes/camx.prm.
User adjustable parameters for RTCMC are set in the CAMx include file
Includes/rtcmcchm.inc. If an error is encountered at model start up because one of
these RTCMC parameter has been exceeded, consult the list of parameters in Table 10‐4 and
then change the parameter appropriately in the rtcmcchm.inc include file. Rebuild the
CAMx executable (we recommend performing a “make clean” before making a new CAMx
executable) after changing any RTCMC parameter.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

217

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
10. Reactive Tracers

Table 10‐4. RTCMC parameters default settings in the Includes/rtcmcchm.inc include
file.
Name
MXRX
MXPHOT
MXZEN
MXRCT
MXPRD
MXEQM
MXSLO

Description
maximum number of RTCMC reactions
maximum number of photolysis reactions
maximum number of photolysis reaction zenith angles
Maximum number of reactants in each reaction
maximum number of products in each reaction
maximum number of equilibrium species
maximum number of slow species

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

218

Default
20
10
15
3
20
5
25

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
11. References

11. REFERENCES
Anthes, R.A. and T.T. Warner. 1978. Development of Hydrodynamic Models Suitable for Air
Pollution and Other Mesometeorological Studies. Mon. Wea. Rev., 106, 1045‐1078.
Arey, J., S.M. Aschmann, E.S.C. Kwok, R. Atkinson. 2001. Alkyl nitrate, hydroxyalkyl nitrate, and
hydroxycarbonyl formation from the NO x‐air photooxidations of C5‐C8 n‐alkanes. J.
Phys. Chem., A 105, 1020‐1027.
Ariya, P.A., A. Khalizov and A. Gidas. 2002. Reactions of gaseous mercury with atomic and
molecular halogens: kinetics, product studies, and atmospheric implications. J. Phys.
Chem., 106, 7310‐7320.
Astitha M., C. Spyrou, G. Kallos, H. Denier Van der Gon, A. Visschedijk and J. Lelieveld. 2009.
Chemical composition change of aerosols along long‐range transport paths. 30th
NATO/SPS ITM on Air Pollution Modelling and its Applications, May 2009, San Francisco,
CA.
Atkinson, R., W.P. Carter, A.M. Winer. 1983. Effects of temperature and pressure on alkyl
nitrate yields in the nitrogen oxide (NOx) photooxidations of n‐pentane and n‐heptane.
J. Phys. Chem., 87(11), 2012‐2018.
Atkinson, R.A., D.L. Baulch, R.A. Cox, J.N. Crowley, R.F. Hampson, R.G. Hynes, M.E. Jenkin, J.A.
Kerr, M.J. Rossi, J. Troe. 2010. “Evaluated kinetic and photochemical data for
atmospheric chemistry ‐ IUPAC subcommittee on gas kinetic data evaluation for
atmospheric chemistry.” January 3, 2010 web version available at http://www.iupac‐
kinetic.ch.cam.ac.uk/index.html.
Barlage, M., F. Chen, M. Tewari, K. Ikeda, D. Gochis, J. Dudhia, R. Rasmussen, B. Livneh, M. Ek,
K. Mitchell. 2010. Noah land surface model modifications to improve snowpack
prediction in the Colorado Rocky Mountains. J. Geophys. Res., 115, D22101,
doi:10.1029/2009JD013470.
Bertram, T.H., Thornton, J.A. 2009. Toward a general parameterization of N2O5 reactivity on
aqueous particles: the competing effects of particle liquid water, nitrate and chloride.
Atmos. Chem. Phys., 9, 8351‐8363.
Bott, A. 1989. A Positive Definite Advection Scheme Obtained by Nonlinear Renormalization of
the Advective Fluxes. Mon. Wea. Rev., 117, 1006‐1015.
Brown, S.S., Ryerson, T.B., Wollny, A.G., Brock, C.A., Peltier, R., Sullivan, A.P., Weber, R.J., Dube,
W.P., Trainer, M., Meagher, J.F., Fehsenfeld, F.C., Ravishankara, A.R. 2006. Variability in
nocturnal nitrogen oxide processing and its role in regional air quality. Science, 311, 67‐
70.
Carlton, A.G., B.J. Turpin, K.E. Altieri, S. Seitzinger, A. Reff, H.‐J. Lim, B. Ervens. 2007.
Atmospheric oxalic acid and SOA production from glyoxal: Results of aqueous
photooxidation experiments. Atmos. Environ., 41, 7588‐7602.
Carpenter, L.J. 2003. Iodine in the marine boundary layer. Chemical Reviews, 103, 4953‐4962.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

219

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
11. References

Carter, W.P.L. 1996. Condensed Atmospheric Photooxidation Mechanisms for Isoprene.
Atmos. Environ., 30, 4275‐4290.
Carter, W.P.L. 2010. Development of the SAPRC‐07 chemical mechanism. Atmos. Environ., 44,
5324‐5335.
Carter, W.P. and G. Heo. 2013. Development of revised SAPRC aromatics mechanisms. Atmos.
Environ., 77, 404‐414.
Chang, J.S., R.A. Brost, I.S.A. Isaksen, S. Madronich, P. Middleton, W.R. Stockwell, and C.J,
Walcek. 1987. A Three‐dimensional Eulerian Acid Deposition Model: Physical Concepts
and Formulation. J. Geophys. Res., 92, 14,681‐14,700.
Clever, H., S.A. Johnson and E.M. Derrick. 1985. The solubility of mercury and some sparingly
soluble mercury salts in water and aqueous solutions. J. Phys. Chem. Ref. Data, 14, 631‐
680.
Cohan, D.S., A. Hakami, Y. Hu, and A.G. Russell. 2005. Nonlinear Response of Ozone to
Emissions: Source Apportionment and Sensitivity Analysis. Environ. Sci. Technol., 39,
6739‐6748.
Colella, P., and P.R. Woodward. 1984. The Piecewise Parabolic Method (PPM) for Gas‐
dynamical Simulations. J. Comp. Phys., 54, 174‐201.
Del Genio, A.D., M.S. Yao, W. Kovari, K.K.W. Lo. 1996. A prognostic cloud water
parameterization for global climate models. J. Climate, 9, 270‐304.
Donahue, N.M., A.L. Robinson, C.O. Stanier, S.N. Pandis. 2006. Coupled partitioning, dilution,
and chemical aging of semivolatile organics. Environ. Sci. Technol., 40, 2635‐2643.
Donahue, N.M., S.A. Epstein, S.N. Pandis, A.L. Robinson. 2011. A two‐dimensional volatility
basis set: 1. organic‐aerosol mixing thermodynamics. Atmos. Chem. Phys., 11, 3303‐
3318.
Donahue, N.M., J.H. Kroll, S.N. Pandis, A.L. Robinson. 2012. A two‐dimensional volatility basis
set – Part 2: Diagnostics of organic‐aerosol evolution. Atmos. Chem. Phys., 12, 615‐634.
Dunker, A.M. 1980. The response of an atmospheric reaction‐transport model to changes in
input functions. Atmos. Environ., 14, 671‐679.
Dunker, A.M. 1981. Efficient calculations of sensitivity coefficients for complex atmospheric
models. Atmos. Environ. 15, 1155‐1161.
Dunker A.M., G. Yarwood, J.P. Ortmann, G.M. Wilson. 2002. The decoupled direct method for
sensitivity analysis in a three‐dimensional air quality model – implementation, accuracy
and efficiency. Environ. Sci. Technol., 36, 2965‐2976.
Edgerton, E.S., B.E. Hartsell and J.J. Jansen. 2001. Atmospheric mercury measurements at a
rural and urban site near Atlanta, GA, USA. 6th International Conference on Mercury as a
Global Pollutant, 15‐19 October 2001, Minamata, Japan.
Ek, M.B., K.E. Mitchell, Y. Lin, E. Rogers, P. Grunmann, V. Koren, G. Gayno, J. D. Tarpley. 2003.
Implementation of Noah land surface model advances in the National Centers for
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

220

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
11. References

Environmental Prediction operational mesoscale Eta model. J. Geophys. Res., 108(D22),
8851, doi:10.1029/2002JD003296.
Elterman, L. 1968. “UV, Visible, and IR Attenuation for Altitudes to 50 km, 1968.” US Air Force
Cambridge Research Laboratory Report, AFCRL 68‐0153.
Emery, C., J. Jung, J. Johnson, G. Yarwood, S. Madronich, G.Grell. 2010. Improving the
Characterization of Clouds and their Impact on Photolysis Rates within the CAMx
Photochemical Grid Model. Prepared for the Texas Commission on Environmental
Quality, Austin, TX. Prepared by ENVIRON International Corporation, Novato, CA and
the National Center for Atmospheric Research, Boulder, CO (August 27, 2010).
Emery, C., E. Tai, G. Yarwood, R. Morris. 2011. Investigation into approaches to reduce
excessive vertical transport over complex terrain in a regional photochemical grid
model. Atmos. Environ., 45, 7341‐7351, doi:10.1016/j.atmosenv.2011.07.052.
Emery, C., J. Jung, B. Koo, G. Yarwood. 2015. Improvements to CAMx Snow Cover Treatments
and Carbon Bond Chemical Mechanism for Winter Ozone. Prepared for the Utah
Department of Environmental Quality, Division of Air Quality, Salt Lake City, UT.
Prepared by Ramboll Environ, Novato, CA (August 2015).
ENVIRON. 2002. Development, Application, and Evaluation of an Advanced Photochemical Air
Toxics Modeling System. Prepared for the Coordinating Research Council, Alpharetta,
GA, and the U.S. Department of Energy, Office of Heavy Vehicle Technologies
(September 27, 2002). Available from www.crcao.com.
EPA. 1990. User’s Guide for the Urban Airshed Model‐Volume I; User’s Manual for UAM(CB‐
IV). U.S. Environmental Protection Agency, Research Triangle Park, NC, EPA‐450/4‐90‐
007a.
EPA. 1998. User’s Guide for the AERMOD Meteorological Preprocessor (Revised Draft).
Prepared by the U.S. Environmental Protection Agency, Research Triangle Park, NC
(November, 1998).
EPRI. 2000. SCICHEM Version 1.2: Technical Documentation. Final Report prepared by
ARAP/Titan Corporation, Princeton, NJ, for EPRI, Palo Alto, CA. December 2000
(1000713).
Fahey, K.M. and S.N. Pandis. 2001. Optimizing model performance: variable size resolution in
cloud chemistry modeling. Atmos. Environ. 35, 4471‐4478.
FLAG. 2000. “Federal Land Managers’ Air Quality Related Values Workgroup (FLAG), Phase I
Report.” Prepared by the US Forest Service, Air Quality Program; National Park Service,
Air Resources Division; and US Fish and Wildlife Service, Air Quality Branch (December
2000).
Gallagher, W.M., K.M. Beswick, J. Duyzer, H. Westrate, T.W. Choularton, J.P. Hummelsho. 1997.
Measurements of aerosol fluxes to Speulder forest using a micrometeorological
technique. Atmos. Environ., 31, 359 – 373.
Ganzeveld, L., D. Helmig, C. W. Fairall, J. Hare, and A. Pozzer. 2009. Atmosphere‐ocean ozone
exchange: A global modeling study of biogeochemical, atmospheric, and waterside
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

221

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
11. References

turbulence dependencies. Global Biogeochem. Cycles, 23, GB4021,
doi:10.1029/2008GB003301.
Gardfeldt, K. and M. Johnson. 2003. Is bimolecular reduction of Hg(II) complexes possible in
aqueous systems of environmental importance? J. Phys. Chem., 107, 4478‐4482.
Gear, C.W. 1971. Numerical Initial Value Problems in Ordinary Differential Equations. Prentice‐
Hall, Englewood Cliffs, NJ.
Gery, M.W., G.Z. Whitten, J.P. Killus, and M.C. Dodge. 1989. A Photochemical Kinetics
Mechanism for Urban and Regional Scale Computer Modeling. J. Geophys. Res., 94,
925‐956.
Goliff, W.S., W.R. Stockwell, C.V. Lawson. 2013. The regional atmospheric chemistry
mechanism, version 2. Atmos. Environ., 68, 174‐185.
Hairer, E., and G. Wanner. 1991. Solving ordinary differential equations II Stiff and differential‐
algebraic problems. Springer‐Verlag, Berlin.
Hakami, A., M.T. Odman, and A.G. Russell. 2003. High‐order, direct sensitivity analysis of
multidimensional air quality models. Environ. Sci. Technol., 37, 2442‐2452.
Hall, B. 1995. The gas‐phase oxidation of elemental mercury by ozone. Water Air Soil Pollut.,
80, 301‐315.
Helmig, D., E.K. Lang, L. Bariteau, P. Boylan, C.W. Fairall, L. Ganzeveld, J.E. Hare, J. Hueber, M.
Pallandt. 2012. Atmospherie‐ocean ozone fluxes during the TexAQS 2006, STRATUS
2006, GOMECC 2007, GasEx 2008 and AMMA 2008 Cruises. J. Geophys. Res., 117,
D04305, doi:10.1029/2011JD015955.
Herman, J.R. and E.A. Celarier. 1997. Earth surface reflectivity climatology at 340‐380 nm from
TOMS data. J. Geophys. Res., 102, No. 23.
Hertel O., R. Berkowics, J. Christensen and O. Hov. 1993. Test of two numerical schemes for
use in atmospheric transport‐chemistry models. Atmos. Env., 27, 2591‐2611.
Hildebrandt Ruiz, L.H, and G. Yarwood. 2013. Interactions between organic aerosol and NOy:
Influence on oxidant production. Prepared for the Texas AQRP (Project 12‐012), by the
University of Texas at Austin, and ENVIRON International Corporation, Novato, CA
(http://aqrp.ceer.utexas.edu/projectinfoFY12_13/12‐012/12‐
012%20Final%20Report.pdf).
Hildebrandt Ruiz, L., Koo, B., Yarwood, G. 2015. Sources of Organic Particulate Matter in
Houston: Evidence from DISCOVER‐AQ data ‐ Modeling and Experiments. Final Report
prepared for the Texas AQRP (Project 14‐024), by the University of Texas at Austin, and
Ramboll Environ, Novato, CA (http://aqrp.ceer.utexas.edu/projectinfoFY14_15/14‐
024/14‐024%20Final%20Report.pdf).
Hindmarsh, A.C. 1983. ODEPACK, a Systematized Collection of ODE Solvers. In Numerical
Methods for Scientific Computation, 55, R.S. Stepleman, Ed., North‐Holland, New York.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

222

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
11. References

Holmes, C.D., D.J. Jacob, and X. Yang. 2006. Global lifetime of elemental mercury against
oxidation by atomic bromine in the free troposphere. Geophys. Res. Lett., 33, L20808,
doi:10.1029/2006GL027176.
Holtslag, A.A.M. and B.A. Boville. 1993. Local versus nonlocal boundary‐layer diffusion in a
global climate model. J. Climate, 6, 1825–1842.
Huebert, B.J., and C.H. Robert. 1985. The Dry Deposition of Nitric Acid to Grass. J. Geophys.
Res., 90(D1), 2085‐2090 (doi:10.1029/JD090iD01p02085).
Hutzell, W.T., Luecken, D.J., Appel, K.W., Carter, W.P.L. 2012. Interpreting predictions from the
SAPRC07 mechanism based on regional and continental simulations. Atmos. Environ.,
46, 417‐429.
IUPAC. 1992. Evaluated Kinetic and Photochemical Data for Atmospheric Chemistry.
Supplement IV. IUPAC Subcommittee on Gas Kinetic Data Evaluation for Atmospheric
Chemistry (R. Atkinson, D.L. Baulch, R. A. Cox, R. F. Hampson, Jr., J. A. Kerr, and J. Troe).
Journal of Physical and Chemical Reference Data, 21, No. 6, 1125‐1568.
IUPAC. 2014a. “Summary of Reactions and Preferred Rate Data, Volume 3 ‐ Inorganic Halogen
Species.” Available at http://iupac.pole‐
ether.fr/datasheets/summary/vol3_summary.xml.
IUPAC. 2014b. “Summary of Reactions and Preferred Rate Data, Volume ‐ Organic Halogen
Species.” Available at http://iupac.pole‐
ether.fr/datasheets/summary/vol4_summary.xml.
IUPAC. 2015. “Task Group on Atmospheric Chemical Kinetic Data Evaluation – Data Sheet
NOx33.” Available at http://iupac.pole‐
ether.fr/htdocs/datasheets/pdf/NOx33_N2O5_H2O.pdf.
Jacob, D.J. 2000. Heterogeneous chemistry and tropospheric ozone. Atmos. Environ., 34,
2131‐2159.
Jaegle, L., D.J. Jacob, W.H. Brune and P.O. Wennberg. 2001. Chemistry of HOx radicals in the
upper troposphere. Atmos. Environ., 35, 469‐489.
Jathar, S.H., Gordon, T.D., Hennigan, C.J., Pye, H.O.T., Pouliot, G., Adams, P.J., Donahue, N.M.,
Robinson, A.L. 2014. Unspeciated organic emissions from combustion sources and their
influence on the secondary organic aerosol budget in the United States. Proc. Natl.
Acad. Sci., 111, 10473‐10478.
Jeffries, H. E., and G.S. Tonnesen. 1994. Comparison of two photochemical reaction
mechanisms using a mass balance and process analysis. Atmos. Environ., 28, 2991‐3003.
Kalberer, M., D. Paulson, M. Sax, M. Steinbacher, J. Dommen, A.S.H. Prevot, R. Fisseha, E.
Weingartner, V. Franhevich, R. Zenob, and U. Balternsperger. 2004. Identification of
polymers as major components of atmospheric organic aerosols. Science, 303, 1659‐
1662.
Karamchandani, P., A. Koo, C. Seigneur. 1998. Reduced gas‐phase kinetic mechanism for
atmospheric plume chemistry. Environ. Sci. Technol., 32, 1709‐1720.
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

223

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
11. References

Koo, B., A.S. Ansari, and S.N. Pandis. 2003. Integrated Approaches to Modeling the Organic
and Inorganic Atmospheric Aerosol Components. Atmos. Environ., 37, 4757‐4768.
Koo, B., G. Yarwood, D.S. Cohan. 2007a. Incorporation of High‐order Decoupled Direct Method
(HDDM) Sensitivity Analysis Capability into CAMx. Prepared for the Texas Commission
on Environmental Quality, Austin, TX. Prepared by ENVIRON International Corporation,
Novato, CA and Rice University, Houston, TX.
Koo, B., A.M. Dunker, G. Yarwood. 2007b. Implementing the Decoupled Direct Method for
Sensitivity Analysis in a Particulate Matter Air Quality Model. Environ. Sci. Technol., 41,
2847‐2854.
Koo, B., G. Yarwood, D.S. Cohan. 2008. Higher‐order Decoupled Direct Method (HDDM) for
Ozone Modeling Sensitivity Analyses and Code Refinements. Prepared for the Texas
Commission on Environmental Quality, Austin, TX. Prepared by ENVIRON International
Corporation, Novato, CA and Rice University, Houston, TX.
Koo, B., G.M. Wilson, R.E. Morris, A.M. Dunker, G. Yarwood. 2009. Comparison of Source
Apportionment and Sensitivity Analysis in a Particulate Matter Air Quality Model.
Environ. Sci. Technol., 43, 6669‐6675.
Koo, B., G. Yarwood and J. Roberts. 2012. An Assessment of Nitryl Chloride Formation
Chemistry and its Importance in Ozone Non‐Attainment Areas in Texas. Final report for
Texas Air Quality Research Program project 10‐015. Prepared by ENVIRON International
Corporation, Novato, CA and NOAA/ESRL, Boulder, CO. Available at
http://aqrp.ceer.utexas.edu/projectinfo%5C10‐015%5C10‐015%20Final%20Report.pdf.
Koo, B., E. Knipping, G. Yarwood. 2014. 1.5‐Dimensional volatility basis set approach for
modeling organic aerosol in CAMx and CMAQ. Atmos. Environ., 95, 158‐164.
Kumar, N., F.W. Lurmann, A.S. Wexler, S. Pandis, and J.H. Seinfeld. 1996. Development and
Application of a Three Dimensional Aerosol Model. Presented at the A&WMA Specialty
Conference on Computing in Environmental Resource Management, Research Triangle
Park, NC, December 2‐4, 1996.
Kumar, N. and A. G. Russell. 1996. Development of a Computationally efficient, Reactive Sub‐
Grid‐Scale Plume Model and the Impact in the Northeastern United States Using
Increasing Levels of Chemical Detail. J. Geophys. Res., 101, 16,737‐16,744.
Landis, M.S. and G. Keeler. 2002. Atmospheric mercury deposition to Lake Michigan during the
Lake Michigan Mass Balance Study. Environ. Sci. Technol., 36, 4518‐4524.
Lane, T.E., N.M. Donahue, S.N. Pandis. 2008. Simulating secondary organic aerosol formation
using the volatility basis‐set approach in a chemical transport model. Atmos. Environ.,
42, 7439‐7451.
Lee, L., P.J. Wooldridge, J.B. Gilman, C. Warneke, J. de Gouw, R.C. Cohen. 2014. Low
temperatures enhance organic nitrate formation: evidence from observations in the
2012 Uintah Basin Winter Ozone Study. Atmos. Chem. Phys. Disc., 14(11), 17401‐17438.
Lin, C.J. and S.O. Pehkonen. 1997. Aqueous‐free radical chemistry of mercury in the presence
of iron oxides and ambient aerosol. Atmos. Environ., 31, 4125‐4137.
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

224

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
11. References

Lin, C.J. and S.O. Pehkonen. 1998. Oxidation of elemental mercury by aqueous chlorine
(HOCl/OCl‐): Implications for tropospheric mercury chemistry. J. Geophys. Res., 103,
28093‐28102.
Lindberg, S., O.R. Bullock, R. Ebinghaus, D. Engstrom, X. Feng, W. Fitzgerald, N. Pirrone, E.
Prestbo, and C. Seigneur. 2007. A synthesis of progress and uncertainties in attributing
the sources of mercury in deposition. Ambio, 36, 19‐33.
Lindqvist, O. and H. Rodhe. 1985. Atmospheric mercury ‐ a review. Tellus, 37B, 136‐159.
Liu, M. and J. Carroll. 1996. A high resolution air pollution model suitable for dispersion studies
in complex terrain. Mon. Wea. Rev., 124, 10, 2396‐2409.
Livneh, B., Y. Xia, K. E. Mitchell, M.B. Ek. 2010. Noah LSM snow model diagnostics and
enhancements. J. Hydromet., 11, doi:10.1175/2009JHM1174.1.
Locatelli, J.D. and P.V. Hobbs. 1974. Fall speeds and masses of solid precipitation particles. J.
Geophys. Res., 79(15), 2185‐2197.
Louis, J.F. 1979. A Parametric Model of Vertical Eddy Fluxes in the Atmosphere. Bound. Lay.
Meteor. 17, 187‐202.
May, A.A., A.A. Presto, C.J. Hennigan, N.T. Nguyen, T.D. Gordon, A.L. Robinson. 2013a. Gas‐
particle partitioning of primary organic aerosol emissions: (1) Gasoline vehicle exhaust.
Atmos. Environ., 77, 128‐139.
May, A.A., A.A. Presto, C.J. Hennigan, N.T. Nguyen, T.D. Gordon, A.L. Robinson. 2013b. Gas‐
Particle Partitioning of Primary Organic Aerosol Emissions: (2) Diesel Vehicles. Environ.
Sci. Technol., 47, 8288‐8296.
May, A.A., E.J.T. Levin, C.J. Hennigan, I. Riipinen, T. Lee, J.L. Collett Jr., J.L. Jimenez, S.M.
Kreidenweis, A.L. Robinson. 2013c. Gas‐particle partitioning of primary organic aerosol
emissions: 3. Biomass burning. J. Geophys. Res., 118, 11327‐11338.
Moore, R. and Tokarczyk, R. 1993. Volatile Biogenic Halocarbons in the Northwest Atlantic.
Global Biogeochem. Cy., 7, 195–210,
Moore, R. and Zafiriou, O. 1994. Photochemical production of methyl iodide in seawater. J.
Geophys. Res., 99, 16415–16420, doi:10.1029/94JD00786.
Morris, R.E., S. Lau, and G. Yarwood. 2003. Development and Application of an Advanced Air
Toxics Hybrid Photochemical Grid Modeling System. Presented at 96th Annual
Conference and Exhibition of the A&WMA, San Diego, California (June 2003).
Munthe, J. 1992. The aqueous oxidation of elemental mercury by ozone. Atmos. Environ., Part
A, 26, 1461‐1468.
Murphy, B.N., S.N. Pandis. 2009. Simulating the Formation of Semivolatile Primary and
Secondary Organic Aerosol in a Regional Chemical Transport Model. Environ. Sci.
Technol., 43, 4722‐4728.
NASA. 1997. Chemical Kinetics and Photochemical Data for Use in Stratospheric Modeling, JPL
Publication 97‐4. Jet Propulsion Laboratory, California Institute of Technology,
Pasadena, California. January 15.
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS
www.camx.com
225

March 2016

CAMx User’s Guide Version 6.3
11. References

NCAR. 2011. The Tropospheric Visible and Ultraviolet (TUV) Radiation Model web page.
National Center for Atmospheric Research, Atmospheric Chemistry Division, Boulder,
Colorado, http://cprm.acd.ucar.edu/Models/TUV/index.shtml.
Nenes, A, C. Pilinis, and S.N. Pandis. 1998. ISORROPIA: A New Thermodynamic Model for
Multiphase Multicomponent Inorganic Aerosols. Aquatic Geochemistry, 4, 123‐152.
Nenes, A., C. Pilinis, and S.N. Pandis. 1999. Continued Development and Testing of a New
Thermodynamic Aerosol Module for Urban and Regional Air Quality Models. Atmos.
Environ. 33, 1553‐1560.
Odman, M. T. and Ingram, C. L. 1993. Multiscale Air Quality Simulation Platform (MAQSIP):
Source Code Documentation and Validation. Technical report, 83 pp., ENV‐96TR002,
MCNC–North Carolina Supercomputing Center, Research Triangle Park, North Carolina,
1996.
Ordóñez, C., J.‐F. Lamarque, S. Tilmes, D.E. Kinnison, E.L. Atlas, D.R. Blake, G. Sousa Santos, G.
Brasseur, and A. Saiz‐Lopez. 2012. Bromine and iodine chemistry in a global chemistry‐
climate model: description and evaluation of very short‐lived oceanic sources. Atmos.
Chem. Phys., 12, 1423‐1447, doi:10.5194/acp‐12‐1423‐2012.
Pal, B., and P.A. Ariya. 2003. Atmospheric reactions of gaseous mercury with ozone and
hydroxyl radical: kinetics and product studies. J. Phys., 107, 189–192.
Pal, B. and P.A. Ariya. 2004. Gas‐phase HO‐initiated reactions of elemental mercury: Kinetics,
product studies and atmospheric implications. Environ. Sci. Technol., 38, 5555‐5566.
Pandis, S.N., A.S. Wexler, and J.H. Seinfeld. 1993. Secondary organic aerosol formation and
transport, II, Predicting the ambient secondary organic aerosol size distribution. Atmos.
Environ., 27A, 2403‐2416.
Parrella, J. P., D.J. Jacob, Q. Liang, Y. Zhang, L. J. Mickley, B. Miller, M. J. Evans et al. 2012.
Tropospheric bromine chemistry: implications for present and pre‐industrial ozone and
mercury. Atmos. Chem. Phys., 12, no. 15: 6723‐6740.
Paulot, F., J.D. Crounse, H.G. Kjaergaard, J.H. Kroll, J.H. Seinfeld, P.O. Wennberg. 2009a.
Isoprene photooxidation: new insights into the production of acids and organic nitrates.
Atmos. Chem. Phys., 9, 1479‐1501.
Paulot, F, J.D. Crounse, H.G. Kjaergaard, A. Kurten, J.M. St.Clair, J.H. Seinfeld, P.O. Wennberg.
2009b. Unexpected Epoxide Formation in the Gas‐Phase Photooxidation of Isoprene.
Science, 325, 730‐733.
Peeters, J., Nguyen, T.L., Vereecken, L. 2009. HOx radical regeneration in the oxidation of
isoprene. Phys. Chem. and Chem. Phys., 11, 5935‐5939
Pehkonen, S.O. and C.J. Lin. 1998. Aqueous photochemistry of divalent mercury with organic
acids. J. Air Waste Manage. Assoc., 48, 144‐150.
Perring, A.E., S.E. Pusede, R.C. Cohen. 2013. An observational perspective on the atmospheric
impacts of alkyl and multifunctional nitrates on ozone and secondary organic aerosol.
Chemical Reviews, 113(8), 5848‐5870.
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

226

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
11. References

Pleim, J. 2007. A combined local and nonlocal closure model for the atmospheric boundary
layer. Part I: Model description and testing. J. Appl. Met. and Clim., 46, 1383‐1395.
Radhakrishnan, K. and A.C. Hindmarsh. 1993. Description and Use of LSODE, the Livermore
Solver for Ordinary Differential Equations. NASA reference Publication 1327. Lawrence
Livermore National Laboratory, Livermore, CA.
Ranjan, M., A.A. Presto, A.A. May, A.L. Robinson. 2012. Temperature Dependence of Gas–
Particle Partitioning of Primary Organic Aerosol Emissions from a Small Diesel Engine.
Aerosol Sci. Technol., 46, 13‐21.
Raofie, F. and P.A. Ariya. 2003. Reactions of BrO with mercury: kinetic studies. J. Phys. IV, 107,
1119–1121.
Robinson, A.L., N.M. Donahue, M.K. Shrivastava, E.A. Weitkamp, A.M. Sage, A.P. Grieshop, T.E.
Lane, J.R. Pierce, S.N. Pandis. 2007. Rethinking organic aerosols: semivolatile emissions
and photochemical aging. Science, 315, 1259‐1262.
Russell, L.M., S.N. Pandis, and J.H. Seinfeld. 1994. Aerosol production and growth in the
marine boundary layer. J. Geophys. Res., 99, 20989‐21003.
Rutter, A.P., and J.J. Schauer. 2007a. The effect of temperature on the gas‐particle partitioning
of reactive mercury in atmospheric aerosols. Atmos. Environ., 41, 8647‐8657.
Rutter, A.P.; Schauer, J.J. 2007b. The impact of aerosol composition on the particle to gas
partitioning of reactive mercury. Environ. Sci. Technol., 41, 3934‐3939.
Ryaboshapko, A., R. Bullock, R. Ebinghaus, I. Ilyin, K. Lohman, J. Munthe, G. Petersen, C.
Seigneur and I. Wängberg. 2002. Comparison of mercury chemistry models. Atmos.
Environ., 36, 3881‐3898.
Sander, S.P., R.R. Friedl, D.M. Golden, M.J. Kurylo, R.E. Huie, V.L. Orkin, G.K. Moortgat, P.H.
Wine, A.R. Ravishankara, C.E. Kolb, M.J. Molina, B.J. Finlayson‐Pitts. 2006. Evaluation
number 15: Chemical kinetics and photochemical data for use in atmospheric studies.
NASA Panel for Data Evaluation, JPL Publication 06‐2, Jet Propulsion Laboratory,
California Insitute of Technology, Pasadena, California.
Sanemasa, I. 1975. The solubility of elemental mercury vapor in water. Bull. Chem. Soc. Jpn.,
48, 1795‐1798.
Sauter, D.P. and P.K. Wang. 1989. An experimental study of the scavenging of aerosol particles
by natural snow crystals. J. Atmos. Sci., 46, 1650‐1655.
Schroeder, W.H. and J. Munthe. 1998. Atmospheric mercury – An overview. Atmos. Environ.,
32, 809‐822.
Scott, B.C. 1978. Parameterization of sulfate removal by precipitation. J. Appl. Meteor., 17,
1375‐1389.
Sehmel, G.A. 1980. Particle and Gas Deposition, a Review. Atmos. Environ., 14, 983‐1011.
Seigneur, C., H. Abeck, G. Chia, M. Reinhard, N.S. Bloom, E. Prestbo and P. Saxena. 1998.
Mercury adsorption to elemental carbon (soot) particles and atmospheric particulate
matter. Atmos. Environ., 32, 2649‐2657.
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS
www.camx.com
227

March 2016

CAMx User’s Guide Version 6.3
11. References

Seigneur, C., P. Karamchandani, K. Lohman, K. Vijayaraghavan and R.‐L. Shia. 2001a. Multiscale
modeling of the atmospheric fate and transport of mercury. J. Geophys. Res., 106,
27795‐27809.
Seigneur, C., P. Karamchandani, K. Lohman and J. Jansen. 2001b. Modeling of mercury in
power plant plumes. 6th International Conference on Mercury as a Global Pollutant, 15‐
19 October 2001, Minamata, Japan.
Seigneur, C., P. Karamchandani, K. Vijayaraghavan, K. Lohman and G. Yelluru. 2003. Scoping
Study for Mercury Deposition in the Upper Midwest. AER Report CP149‐03‐01a,
prepared for the Midwest Regional Planning Organization, Des Plaines, IL.
Seigneur, C., K. Vijayaraghavan, K. Lohman and P. Karamchandani. 2004. Modeling the
atmospheric fate and transport of mercury over North America. Fuel Processing
Technol., 85, 441‐450.
Seigneur, C. and K. Lohman. 2008. Effect of bromine chemistry on the atmospheric mercury
cycle. J. Geophys. Res., 113, D23309, doi:10.1029/2008JD010262.
Seinfeld, J.H., and S.N. Pandis. 1998. Atmospheric Chemistry and Physics, From Air Pollution to
Climate Change. John Wiley and Sons, Inc., NY.
Sillen, G.L. and A.E. Martell, (Eds.). 1964. Stability constants of metal ion complexes, Spec.
Publ. Chem. Soc., 17, 754.
Sillman, S. 1995. The Use of NOy, H2O2, and HNO3 as Indicators for Ozone ‐ NOX‐Hydrocarbon
Sensitivity in Urban Locations. J. Geophys. Res., 100, 14,175‐14,188.
Slinn, S.A. and W.G.N. Slinn. 1980. Predictions for particle deposition on natural waters.
Atmos. Environ., 24, 1013‐1016.
Slinn, W.G.N. 1982. Predictions for particle deposition to vegetative surfaces. Atmos. Environ.,
16, 1785‐1794.
Smagorinsky, J. 1963. General Circulation Experiments with the Primitive Equations: I. The
Basic Experiment. Mon. Wea. Rev., 91, 99‐164.
Smoydzin, L., and R. von Glasow. 2009. Modelling chemistry over the Dead Sea: bromine and
ozone chemistry. Atmospheric Chemistry and Physics, 9(14), 5057‐5072.
Sommar, J., K. Gårdfeldt, D. Strömberg and X. Feng. 2001. A kinetic study of the gas‐phase
reaction between the hydroxyl radical and atomic mercury. Atmos. Environ., 35, 3049‐
3054.
Strader, R., F. Lurmann, and S.N. Pandis. 1999. Evaluation of secondary organic aerosol
formation in winter. Atmos. Environ., 33, 4849‐4863.
Takemura, T., T. Nakajima, O. Dubovik, B.N. Holben, S. Kinne. 2002. Single‐scattering albedo
and radiative forcing of various aerosol species with a global three‐dimensional model.
J. Climate, 15(4), 333‐352.
Tanaka, P.L., D. T. Allen, E. C. McDonald‐Buller, S. Chang, Y. Kimura, C.B.Mullins, G. Yarwood,
and J.D. Neece. 2003. Development of a chlorine mechanism for use in the carbon
bond IV chemistry model. J. Geophys. Res.: Atmos., 108(D4), 1984–2012.
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS
www.camx.com
228

March 2016

CAMx User’s Guide Version 6.3
11. References

Tokos, J.J.S., B. Hall, J.A. Calhoun, and E.M. Prestbo. 1998. Homogeneous gas‐phase reaction of
Hg0 with H2O2, O3, CH3I, and (CH3)2S: Implications for atmospheric Hg cycling. Atmos.
Environ., 32, 823‐827.
Tonnesen, G.S. and R.L. Dennis. 2000. Analysis of radical propagation efficiency to assess
ozone sensitivity to hydrocarbons and NOx. Part 1: Local indicators of odd oxygen
production sensitivity. J. Geophys. Res., 105, 9213‐9225.
Turner, D.B., T. Chico, and A. Catalano. 1986. TUPOS: A Multiple Source Gaussian Dispersion
Algorithm Using On‐Site Turbulence Data. U.S Environmental Protection Agency,
Research Triangle Park, North Carolina (EPA‐600/8‐86/010).
Turner, D.B. and R.H. Schulze. 2007. Practical Guide to Atmospheric Dispersion Modeling. Air
and Waste Management Association.
van Loon, L., E. Mader, and S.L. Scott. 2000. Reduction of the aqueous mercuric ion by sulfite:
UV spectrum of HgSO3 and its intramolecular redox reactions. J. Phys. Chem., 104,
1621‐1626.
van Loon, L.L., E.A. Mader, and S.L. Scott. 2001. Sulfite stabilization and reduction of the
aqueous mercuric ion: kinetic determination of sequential formation constants. J. Phys.
Chem., 105, 3190‐3195.
Vijayaraghavan, K., P. Karamchandani, C. Seigneur, R. Balmori, and S.‐Y. Chen. 2008. Plume‐in‐
grid modeling of atmospheric mercury. J. Geophys. Res., 113, D24305,
doi:10.1029/2008JD010580.
Vogt, R., R. Sander, R. von Glasow, and P.J. Crutzen. 1999. Iodine chemistry and its role in
halogen activation and ozone loss in the marine boundary layer: A model study. J.
Atmos. Chem., 32(3), 375‐395.
Voulgarakis, A., N.H. Savage, O. Wild, G.D. Carver, K.C. Clemitshaw, J.A. Pyle. 2009. Upgrading
photolysis in the p‐TOMCAT CTM: model evaluation and assessment of the role of
clouds. Geo. Mod. Devel., 2, 59‐72.
Wang, Z., J. E. Langstaff, and H.E. Jeffries. 1995. The Application of the Integrated Process Rate
Analysis Method for Investigation of Urban Airshed Model (UAM) Sensitivity to
Speciation in VOC Emissions Data. Air & Waste Management Association annual
meeting, San Antonio, TX.
Wang, Z., and X. Zeng. 2010. Evaluation of snow albedo in land models for weather and
climate studies. J. Appl. Met Clim., 49, doi: 10.1175/2009JAMC2134.1.
Wesely, M.L. 1989. Parameterization of Surface Resistances to Gaseous Dry Deposition in
Regional‐Scale Numerical Models. Atmos. Environ., 23, 1293‐1304.
Wesely, M.L., and B.B. Hicks. 2000. A review of the current status of knowledge on dry
deposition. Atmos. Environ., 34, 2261.
Whitten, G., H.P. Deuel, C.S. Burton, and J.L. Haney. 1996. “Overview of the Implementation of
an Updated Isoprene Chemistry Mechanism in CB4/UAM‐V.” (Revised Memorandum to
OTAG Participants, July 22).
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

229

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
11. References

Whitten, G.Z., and G. Yarwood. 2008. The Ozone Productivity of n‐Propyl Bromide: Part 2—An
Exception to the Maximum Incremental Reactivity Scale. J. Air Waste Manag. Assoc.,
58(7), 891‐901.
Whitten, G.Z., G. Heo, Y. Kimura, E. McDonald‐Buller, D.T. Allen, W.P.I. Carter, G. Yarwood.
2010. A New Condensed Toluene Mechanism for Carbon Bond: CB05‐TU. Atmos.
Environ., 44(40), doi: 10.1016/j.atmosenv.2009.12.029.
Woody, M.C., Baker, K.R., Hayes, P.L., Jimenez, J.L., Koo, B., Pye, H.O.T. 2016. Understanding
sources of organic aerosol during CalNex‐2010 using the CMAQ‐VBS. Atmos. Chem.
Phys., 16, 4081‐4100 (doi:10.5194/acp‐16‐4081‐2016).
Yamartino, R. 2000. Refinement of Horizontal Diffusion in Photochemical Grid Models.
Presented at the American Meteorological Society 11th Joint Conference on the
Applications of Air Pollution Meteorology with the Air and Waste Management
Association, Long Beach, CA, January 9‐13.
Yang, X., R.A. Cox, N.J. Warwick, J.A. Pyle, G.D. Carver, F.M. O'Connor, and N.H. Savage. 2005.
Tropospheric bromine chemistry and its impacts on ozone: A model study. J. Geophys.
Res.: Atmos., 110(D23), DOI: 10.1029/2005JD006244.
Yarwood, G., R.E. Morris, M.A. Yocke, H. Hogo and T. Chico. 1996a. Development of a
Methodology for Source Apportionment of Ozone Concentration Estimates from a
Photochemical Grid Model. Presented at the 89th AWMA Annual Meeting, Nashville
TN, June 23‐28.
Yarwood, G., T.E. Stoeckenius, G. Wilson, R.E. Morris, and M.A. Yocke. 1996b. Development of
a Methodology to Assess Geographic and Temporal Ozone Control Strategies for the
South Coast Air Basin. Prepared for South Coast Air Quality Management District,
Diamond Bar, CA.
Yarwood, G., R.E. Morris, G.M. Wilson. 2004. Particulate Matter Source Apportionment
Technology (PSAT) in the CAMx Photochemical Grid Model. Proceedings of the 27th
NATO/ CCMS International Technical Meeting on Air Pollution Modeling and
Application. Springer Verlag (Available from
http://camx.com/publ/pdfs/yarwood_itm_paper.pdf).
Yarwood. G., G.Z. Whitten, and S. Rao. 2005a. Updates to the Carbon Bond 4 Photochemical
Mechanism. Prepared for the Lake Michigan Air Directors Consortium, Des Plains,
Illinois.
Yarwood. G., S. Rao, M. Yocke, and G.Z. Whitten. 2005b. Updates to the Carbon Bond chemical
mechanism: CB05. Final Report prepared for US EPA. Available at
http://www.camx.com/publ/pdfs/CB05_Final_Report_120805.pdf.
Yarwood, G., J. Jung, G. Z. Whitten, G. Heo, J. Mellberg and E. Estes. 2010. Updates to the
Carbon Bond Mechanism for Version 6 (CB6). Presented at the 9th Annual CMAS
Conference, Chapel Hill, October.
Yarwood, G., H. Gookyoung, W.P.L. Carter, G.Z. Whitten. 2012a. Environmental Chamber
Experiments to Evaluate NOx Sinks and Recycling in Atmospheric Chemical Mechanisms.
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

230

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
11. References

Final Report prepared for the Texas Air Quality Research Program, University of Texas,
Austin, Texas (AQRP Project 10‐042, February 2012).
Yarwood, G., J. Jung, U. Nopmongcol, C. Emery. 2012b. Improving CAMx Performance in
Simulating Ozone Transport from the Gulf of Mexico. Final Report prepared for the
Texas Commission on Environmental Quality, Austin, Texas (September 2012). Prepared
by ENVIRON International Corporation, Novato, CA.
Yarwood, G., T. Sakulyanontvittaya, U. Nopmongcol, B. Koo. 2014. Ozone Depletion by
Bromine and Iodine over the Gulf of Mexico. Final Report prepared for the Texas
Commisision on Environmental Quality, Austin, Texas (November 2014). Prepared by
ENVIRON International Corporation, Novato, CA.
Yarwood, G. and B. Koo. 2015. Improved OSAT, APCA and PSAT Algorithms for CAMx. Final
report prepared for the Texas Commission on Environmental Quality, Austin, Texas
(August, 2015). Prepared by Ramboll Environ, Novato, CA.
Yeh, G.K. and P.J. Ziemann. 2014. Alkyl Nitrate Formation from the Reactions of C8–C14 n‐
Alkanes with OH Radicals in the Presence of NOx: Measured Yields with Essential
Corrections for Gas–Wall Partitioning. J. Phys. Chem., 118(38), 8797‐8806.
Zhang, L., S. Gong, J. Padro, L. Barrie. 2001. A size‐segregated particle dry deposition scheme
for an atmospheric aerosol module. Atmos. Environ., 35, 549‐560.
Zhang, L., J. R. Brook, and R. Vet. 2003. A revised parameterization for gaseous dry deposition
in air‐quality models. Atmos. Chem. Phys., 3, 2067–2082.
Zhang, X., S. Kondragunta, C. Schmidt, and F. Kogan. 2008. Near real time monitoring of
biomass burning particulate emissions (PM2.5) across contiguous United States using
multiple satellite instruments. Atmospheric Environment, 42, 6959‐6972.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

231

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
Appendix A: Mechanism 2 – CB6r2

Appendix A
CAMx Mechanism 2: CB6r2 Gas‐Phase Chemistry
Table A‐1. Reactions and rate constant expressions for the CB6r2 mechanism. k298 is the rate
constant at 298 K and 1 atmosphere using units in molecules/cm3 and 1/s. For photolysis
reactions k298 shows the photolysis rate at a solar zenith angle of 60° and height of 600 m
MSL/AGL. See Table 5‐2 for species names. See Section 3.1 on temperature and pressure
dependencies.
Number
1
2
3
4
5
6

Reactants and Products
NO2 = NO + O
O + O2 + M = O3 + M
O3 + NO = NO2
O + NO + M = NO2 + M
O + NO2 = NO
O + NO2 = NO3

7
8
9
10
11
12
13
14
15
16
17

O + O3 =
O3 = O
O3 = O1D
O1D + M = O + M
O1D + H2O = 2 OH
O3 + OH = HO2
O3 + HO2 = OH
OH + O = HO2
HO2 + O = OH
OH + OH = O
OH + OH = H2O2

18
19

OH + HO2 =
HO2 + HO2 = H2O2

20

HO2 + HO2 + H2O = H2O2

21
22
23
24
25
26
27
28
29
30
31
32

H2O2 = 2 OH
H2O2 + OH = HO2
H2O2 + O = OH + HO2
NO + NO + O2 = 2 NO2
HO2 + NO = OH + NO2
NO2 + O3 = NO3
NO3 = NO2 + O
NO3 = NO
NO3 + NO = 2 NO2
NO3 + NO2 = NO + NO2
NO3 + O = NO2
NO3 + OH = HO2 + NO2

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

Rate Constant Expression
Photolysis
k = 5.68E‐34 (T/300)^‐2.6
k = 1.40E‐12 exp(‐1310/T)
k = 1.00E‐31 (T/300)^‐1.6
k = 5.50E‐12 exp(188/T)
Falloff: F=0.6; n=1
k(0) = 1.30E‐31 (T/300)^‐1.5
k(inf) = 2.30E‐11 (T/300)^0.24
k = 8.00E‐12 exp(‐2060/T)
Photolysis
Photolysis
k = 2.23E‐11 exp(115/T)
k = 2.14E‐10
k = 1.70E‐12 exp(‐940/T)
k = 2.03E‐16 (T/300)^4.57 exp(693/T)
k = 2.40E‐11 exp(110/T)
k = 2.70E‐11 exp(224/T)
k = 6.20E‐14 (T/298)^2.6 exp(945/T)
Falloff: F=0.5; n=1.13
k(0) = 6.90E‐31 (T/300)^‐0.8
k(inf) = 2.60E‐11
k = 4.80E‐11 exp(250/T)
k = k1 + k2 [M]
k1 = 2.20E‐13 exp(600/T)
k2 = 1.90E‐33 exp(980/T)
k = k1 + k2 [M]
k1 = 3.08E‐34 exp(2800/T)
k2 = 2.66E‐54 exp(3180/T)
Photolysis
k = 2.90E‐12 exp(‐160/T)
k = 1.40E‐12 exp(‐2000/T)
k = 3.30E‐39 exp(530/T)
k = 3.45E‐12 exp(270/T)
k = 1.40E‐13 exp(‐2470/T)
Photolysis
Photolysis
k = 1.80E‐11 exp(110/T)
k = 4.50E‐14 exp(‐1260/T)
k = 1.70E‐11
k = 2.00E‐11

232

k298
6.30E‐3
5.78E‐34
1.73E‐14
1.01E‐31
1.03E‐11
2.11E‐12

7.96E‐15
3.33E‐4
8.78E‐6
3.28E‐11
2.14E‐10
7.25E‐14
2.01E‐15
3.47E‐11
5.73E‐11
1.48E‐12
5.25E‐12

1.11E‐10
2.90E‐12

6.53E‐30

3.78E‐6
1.70E‐12
1.70E‐15
1.95E‐38
8.54E‐12
3.52E‐17
1.56E‐1
1.98E‐2
2.60E‐11
6.56E‐16
1.70E‐11
2.00E‐11
www.camx.com

March 2016

CAMx User’s Guide Version 6.3
Appendix A: Mechanism 2 – CB6r2

Number
33
34
35
36

Reactants and Products
NO3 + HO2 = OH + NO2
NO3 + O3 = NO2
NO3 + NO3 = 2 NO2
NO3 + NO2 = N2O5

37

N2O5 = NO3 + NO2

38
39
40

N2O5 = NO2 + NO3
N2O5 + H2O = 2 HNO3
NO + OH = HONO

41
42
43
44
45

NO + NO2 + H2O = 2 HONO
HONO + HONO = NO + NO2
HONO = NO + OH
HONO + OH = NO2
NO2 + OH = HNO3

46

HNO3 + OH = NO3

47
48

HNO3 = OH + NO2
HO2 + NO2 = PNA

49

PNA = HO2 + NO2

50
51
52

PNA = 0.59 HO2 + 0.59 NO2 + 0.41 OH +
0.41 NO3
PNA + OH = NO2
SO2 + OH = SULF + HO2

53
54

C2O3 + NO = NO2 + MEO2 + RO2
C2O3 + NO2 = PAN

55

PAN = NO2 + C2O3

56

PAN = 0.6 NO2 + 0.6 C2O3 + 0.4 NO3 + 0.4
MEO2 + 0.4 RO2

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

Rate Constant Expression
k = 4.00E‐12
k = 1.00E‐17
k = 8.50E‐13 exp(‐2450/T)
Falloff: F=0.35; n=1.33
k(0) = 3.60E‐30 (T/300)^‐4.1
k(inf) = 1.90E‐12 (T/300)^0.2
Falloff: F=0.35; n=1.33
k(0) = 1.30E‐3 (T/300)^‐3.5 exp(‐11000/T)
k(inf) = 9.70E+14 (T/300)^0.1 exp(‐
11080/T)
Photolysis
k = 1.00E‐22
Falloff: F=0.81; n=0.87
k(0) = 7.40E‐31 (T/300)^‐2.4
k(inf) = 3.30E‐11 (T/300)^‐0.3
k = 5.00E‐40
k = 1.00E‐20
Photolysis
k = 2.50E‐12 exp(260/T)
Falloff: F=0.6; n=1
k(0) = 1.80E‐30 (T/300)^‐3
k(inf) = 2.80E‐11
k = k1 + k3 [M] / (1 + k3 [M] / k2)
k1 = 2.40E‐14 exp(460/T)
k2 = 2.70E‐17 exp(2199/T)
k3 = 6.50E‐34 exp(1335/T)

k298
4.00E‐12
1.00E‐17
2.28E‐16
1.24E‐12

4.46E‐2

2.52E‐5
1.00E‐22
9.77E‐12

5.00E‐40
1.00E‐20
1.04E‐3
5.98E‐12
1.06E‐11

1.54E‐13

Photolysis
Falloff: F=0.6; n=1
k(0) = 1.80E‐31 (T/300)^‐3.2
k(inf) = 4.70E‐12
Falloff: F=0.6; n=1
k(0) = 4.10E‐5 exp(‐10650/T)
k(inf) = 4.80E+15 exp(‐11170/T)
Photolysis

2.54E‐7
1.38E‐12

k = 3.20E‐13 exp(690/T)
Falloff: F=0.53; n=1.1
k(0) = 4.50E‐31 (T/300)^‐3.9
k(inf) = 1.30E‐12 (T/300)^‐0.7
k = 7.50E‐12 exp(290/T)
Falloff: F=0.3; n=1.41
k(0) = 2.70E‐28 (T/300)^‐7.1
k(inf) = 1.20E‐11 (T/300)^‐0.9
Falloff: F=0.3; n=1.41
k(0) = 4.90E‐3 exp(‐12100/T)
k(inf) = 5.40E+16 exp(‐13830/T)
Photolysis

3.24E‐12
8.12E‐13

233

8.31E‐2

2.36E‐6

1.98E‐11
9.40E‐12

2.98E‐4

3.47E‐7

www.camx.com

March 2016

Number
57

CAMx User’s Guide Version 6.3
Appendix A: Mechanism 2 – CB6r2

61
62

Reactants and Products
C2O3 + HO2 = 0.41 PACD + 0.15 AACD +
0.15 O3 + 0.44 MEO2 + 0.44 RO2 + 0.44 OH
C2O3 + RO2 = C2O3
C2O3 + C2O3 = 2 MEO2 + 2 RO2
C2O3 + CXO3 = MEO2 + ALD2 + XO2H + 2
RO2
CXO3 + NO = NO2 + ALD2 + XO2H + RO2
CXO3 + NO2 = PANX

63

PANX = NO2 + CXO3

64

PANX = 0.6 NO2 + 0.6 CXO3 + 0.4 NO3 + 0.4
ALD2 + 0.4 XO2H + 0.4 RO2
CXO3 + HO2 = 0.41 PACD + 0.15 AACD +
0.15 O3 + 0.44 ALD2 + 0.44 XO2H + 0.44
RO2 + 0.44 OH
CXO3 + RO2 = 0.8 ALD2 + 0.8 XO2H + 0.8
RO2
CXO3 + CXO3 = 2 ALD2 + 2 XO2H + 2 RO2
RO2 + NO = NO
RO2 + HO2 = HO2
RO2 + RO2 =
MEO2 + NO = FORM + HO2 + NO2
MEO2 + HO2 = 0.9 MEPX + 0.1 FORM
MEO2 + C2O3 = FORM + 0.9 HO2 + 0.9
MEO2 + 0.1 AACD + 0.9 RO2
MEO2 + RO2 = 0.685 FORM + 0.315 MEOH
+ 0.37 HO2 + RO2

58
59
60

65

66
67
68
69
70
71
72
73
74

75
76
77

XO2H + NO = NO2 + HO2
XO2H + HO2 = ROOH
XO2H + C2O3 = 0.8 HO2 + 0.8 MEO2 + 0.2
AACD + 0.8 RO2

78

XO2H + RO2 = 0.6 HO2 + RO2

79

XO2 + NO = NO2

80

XO2 + HO2 = ROOH

81

XO2 + C2O3 = 0.8 MEO2 + 0.2 AACD + 0.8
RO2

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

Rate Constant Expression
k = 5.20E‐13 exp(980/T)

k298
1.39E‐11

k = 8.90E‐13 exp(800/T)
k = 2.90E‐12 exp(500/T)
k = 2.90E‐12 exp(500/T)

1.30E‐11
1.55E‐11
1.55E‐11

k = 6.70E‐12 exp(340/T)
k = k(ref) K
k(ref) = k(54)
K = 1.00E+0
k = k(ref) K
k(ref) = k(55)
K = 1.00E+0
Photolysis

2.10E‐11
9.40E‐12

k = 5.20E‐13 exp(980/T)

1.39E‐11

k = 8.90E‐13 exp(800/T)

1.30E‐11

k = 3.20E‐12 exp(500/T)
k = 2.40E‐12 exp(360/T)
k = 4.80E‐13 exp(800/T)
k = 6.50E‐14 exp(500/T)
k = 2.30E‐12 exp(360/T)
k = 3.80E‐13 exp(780/T)
k = 2.00E‐12 exp(500/T)

1.71E‐11
8.03E‐12
7.03E‐12
3.48E‐13
7.70E‐12
5.21E‐12
1.07E‐11

k = k(ref) K
k(ref) = k(70)
K = 1.00E+0
k = 2.70E‐12 exp(360/T)
k = 6.80E‐13 exp(800/T)
k = k(ref) K
k(ref) = k(58)
K = 1.00E+0
k = k(ref) K
k(ref) = k(70)
K = 1.00E+0
k = k(ref) K
k(ref) = k(75)
K = 1.00E+0
k = k(ref) K
k(ref) = k(76)
K = 1.00E+0
k = k(ref) K
k(ref) = k(58)
K = 1.00E+0

3.48E‐13

234

2.98E‐4

3.47E‐7

9.04E‐12
9.96E‐12
1.30E‐11

3.48E‐13

9.04E‐12

9.96E‐12

1.30E‐11

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
Appendix A: Mechanism 2 – CB6r2

Number
82

Reactants and Products
XO2 + RO2 = RO2

83

XO2N + NO = 0.5 NTR1 + 0.5 NTR2

84

XO2N + HO2 = ROOH

85

XO2N + C2O3 = 0.8 HO2 + 0.8 MEO2 + 0.2
AACD + 0.8 RO2

86

XO2N + RO2 = RO2

87

MEPX + OH = 0.6 MEO2 + 0.6 RO2 + 0.4
FORM + 0.4 OH
MEPX = MEO2 + RO2 + OH
ROOH + OH = 0.54 XO2H + 0.06 XO2N + 0.6
RO2 + 0.4 OH
ROOH = HO2 + OH
NTR1 + OH = NTR2
NTR1 = NO2
FACD + OH = HO2
AACD + OH = MEO2 + RO2
PACD + OH = C2O3
FORM + OH = HO2 + CO
FORM = 2 HO2 + CO
FORM = CO + H2
FORM + O = OH + HO2 + CO
FORM + NO3 = HNO3 + HO2 + CO
FORM + HO2 = HCO3
HCO3 = FORM + HO2
HCO3 + NO = FACD + NO2 + HO2
HCO3 + HO2 = 0.5 MEPX + 0.5 FACD + 0.2
OH + 0.2 HO2
ALD2 + O = C2O3 + OH
ALD2 + OH = C2O3
ALD2 + NO3 = C2O3 + HNO3
ALD2 = MEO2 + RO2 + CO + HO2
ALDX + O = CXO3 + OH
ALDX + OH = CXO3
ALDX + NO3 = CXO3 + HNO3
ALDX = ALD2 + XO2H + RO2 + CO + HO2
GLYD + OH = 0.2 GLY + 0.2 HO2 + 0.8 C2O3
GLYD = 0.74 FORM + 0.89 CO + 1.4 HO2 +
0.15 MEOH + 0.19 OH + 0.11 GLY + 0.11
XO2H + 0.11 RO2
GLYD + NO3 = HNO3 + C2O3

88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114

115

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

Rate Constant Expression
k = k(ref) K
k(ref) = k(70)
K = 1.00E+0
k = k(ref) K
k(ref) = k(75)
K = 1.00E+0
k = k(ref) K
k(ref) = k(76)
K = 1.00E+0
k = k(ref) K
k(ref) = k(58)
K = 1.00E+0
k = k(ref) K
k(ref) = k(70)
K = 1.00E+0
k = 5.30E‐12 exp(190/T)

k298
3.48E‐13

9.04E‐12

9.96E‐12

1.30E‐11

3.48E‐13

1.00E‐11

Photolysis
k = 5.30E‐12 exp(190/T)

2.68E‐6
1.00E‐11

Photolysis
k = 2.00E‐12
Photolysis
k = 4.50E‐13
k = 4.00E‐14 exp(850/T)
k = 5.30E‐12 exp(190/T)
k = 5.40E‐12 exp(135/T)
Photolysis
Photolysis
k = 3.40E‐11 exp(‐1600/T)
k = 5.50E‐16
k = 9.70E‐15 exp(625/T)
k = 2.40E+12 exp(‐7000/T)
k = 5.60E‐12
k = 5.60E‐15 exp(2300/T)

2.68E‐6
2.00E‐12
1.06E‐6
4.50E‐13
6.93E‐13
1.00E‐11
8.49E‐12
1.78E‐5
2.38E‐5
1.58E‐13
5.50E‐16
7.90E‐14
1.51E+2
5.60E‐12
1.26E‐11

k = 1.80E‐11 exp(‐1100/T)
k = 4.70E‐12 exp(345/T)
k = 1.40E‐12 exp(‐1860/T)
Photolysis
k = 1.30E‐11 exp(‐870/T)
k = 4.90E‐12 exp(405/T)
k = 6.30E‐15
Photolysis
k = 8.00E‐12
Photolysis

4.49E‐13
1.50E‐11
2.73E‐15
1.76E‐6
7.02E‐13
1.91E‐11
6.30E‐15
6.96E‐6
8.00E‐12
1.56E‐6

k = 1.40E‐12 exp(‐1860/T)

2.73E‐15

235

www.camx.com

March 2016

Number
116
117
118
119
120
121
122
123

124
125
126
127

128
129
130
131
132

133

134
135
136

137
138

139
140
141

CAMx User’s Guide Version 6.3
Appendix A: Mechanism 2 – CB6r2

Reactants and Products
GLY + OH = 1.8 CO + 0.2 XO2 + 0.2 RO2 +
HO2
GLY = 2 HO2 + 2 CO
GLY + NO3 = HNO3 + 1.5 CO + 0.5 XO2 + 0.5
RO2 + HO2
MGLY = C2O3 + HO2 + CO
MGLY + NO3 = HNO3 + C2O3 + XO2 + RO2
MGLY + OH = C2O3 + CO
H2 + OH = HO2
CO + OH = HO2

CH4 + OH = MEO2 + RO2
ETHA + OH = 0.991 ALD2 + 0.991 XO2H +
0.009 XO2N + RO2
MEOH + OH = FORM + HO2
ETOH + OH = 0.95 ALD2 + 0.9 HO2 + 0.1
XO2H + 0.1 RO2 + 0.078 FORM + 0.011
GLYD
KET = 0.5 ALD2 + 0.5 C2O3 + 0.5 XO2H + 0.5
CXO3 + 0.5 MEO2 + RO2 ‐ 2.5 PAR
ACET = 0.38 CO + 1.38 MEO2 + 1.38 RO2 +
0.62 C2O3
ACET + OH = FORM + C2O3 + XO2 + RO2
PRPA + OH = 0.71 ACET + 0.26 ALDX + 0.26
PAR + 0.97 XO2H + 0.03 XO2N + RO2
PAR + OH = 0.11 ALDX + 0.76 ROR + 0.13
XO2N + 0.11 XO2H + 0.76 XO2 + RO2 ‐ 0.11
PAR
ROR = 0.2 KET + 0.42 ACET + 0.74 ALD2 +
0.37 ALDX + 0.04 XO2N + 0.94 XO2H + 0.98
RO2 + 0.02 ROR ‐ 2.7 PAR
ROR + O2 = KET + HO2
ROR + NO2 = NTR1
ETHY + OH = 0.7 GLY + 0.7 OH + 0.3 FACD +
0.3 CO + 0.3 HO2
ETH + O = FORM + HO2 + CO + 0.7 XO2H +
0.7 RO2 + 0.3 OH
ETH + OH = XO2H + RO2 + 1.56 FORM +
0.22 GLYD
ETH + O3 = FORM + 0.51 CO + 0.16 HO2 +
0.16 OH + 0.37 FACD
ETH + NO3 = 0.5 NO2 + 0.5 NTR1 + 0.5
XO2H + 0.5 XO2 + RO2 + 1.125 FORM
OLE + O = 0.2 ALD2 + 0.3 ALDX + 0.1 HO2 +
0.2 XO2H + 0.2 CO + 0.2 FORM + 0.01 XO2N
+ 0.21 RO2 + 0.2 PAR + 0.1 OH

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

Rate Constant Expression
k = 3.10E‐12 exp(340/T)

k298
9.70E‐12

Photolysis
k = 1.40E‐12 exp(‐1860/T)

5.50E‐5
2.73E‐15

Photolysis
k = 1.40E‐12 exp(‐1860/T)
k = 1.90E‐12 exp(575/T)
k = 7.70E‐12 exp(‐2100/T)
k = k1 + k2 [M]
k1 = 1.44E‐13
k2 = 3.43E‐33
k = 1.85E‐12 exp(‐1690/T)
k = 6.90E‐12 exp(‐1000/T)

1.46E‐4
2.73E‐15
1.31E‐11
6.70E‐15
2.28E‐13

k = 2.85E‐12 exp(‐345/T)
k = 3.00E‐12 exp(20/T)

8.95E‐13
3.21E‐12

Photolysis

2.27E‐7

Photolysis

2.08E‐7

k = 1.41E‐12 exp(‐620.6/T)
k = 7.60E‐12 exp(‐585/T)

1.76E‐13
1.07E‐12

k = 8.10E‐13

8.10E‐13

k = 5.70E+12 exp(‐5780/T)

2.15E+4

k = 1.50E‐14 exp(‐200/T)
k = 8.60E‐12 exp(400/T)
Falloff: F=0.37; n=1.3
k(0) = 5.00E‐30 (T/300)^‐1.5
k(inf) = 1.00E‐12
k = 1.04E‐11 exp(‐792/T)

7.67E‐15
3.29E‐11
7.52E‐13

Falloff: F=0.48; n=1.15
k(0) = 8.60E‐29 (T/300)^‐3.1
k(inf) = 9.00E‐12 (T/300)^‐0.85
k = 9.10E‐15 exp(‐2580/T)

7.84E‐12

1.58E‐18

k = 3.30E‐12 exp(‐2880/T)

2.10E‐16

k = 1.00E‐11 exp(‐280/T)

3.91E‐12

236

6.37E‐15
2.41E‐13

7.29E‐13

www.camx.com

March 2016

Number
142

143

144

145
146
147

148

149
150

151

152
153

154

155
156

157

158

CAMx User’s Guide Version 6.3
Appendix A: Mechanism 2 – CB6r2

Reactants and Products
OLE + OH = 0.781 FORM + 0.488 ALD2 +
0.488 ALDX + 0.976 XO2H + 0.195 XO2 +
0.024 XO2N + 1.195 RO2 ‐ 0.73 PAR
OLE + O3 = 0.295 ALD2 + 0.555 FORM +
0.27 ALDX + 0.15 XO2H + 0.15 RO2 + 0.334
OH + 0.08 HO2 + 0.378 CO + 0.075 GLY +
0.075 MGLY + 0.09 FACD + 0.13 AACD +
0.04 H2O2 ‐ 0.79 PAR

Rate Constant Expression
Falloff: F=0.5; n=1.13
k(0) = 8.00E‐27 (T/300)^‐3.5
k(inf) = 3.00E‐11 (T/300)^‐1
k = 5.50E‐15 exp(‐1880/T)

k298
2.86E‐11

OLE + NO3 = 0.5 NO2 + 0.5 NTR1 + 0.48
XO2 + 0.48 XO2H + 0.04 XO2N + RO2 + 0.5
FORM + 0.25 ALD2 + 0.375 ALDX ‐ 1 PAR
IOLE + O = 1.24 ALD2 + 0.66 ALDX + 0.1
XO2H + 0.1 RO2 + 0.1 CO + 0.1 PAR
IOLE + OH = 1.3 ALD2 + 0.7 ALDX + XO2H +
RO2
IOLE + O3 = 0.732 ALD2 + 0.442 ALDX +
0.128 FORM + 0.245 CO + 0.5 OH + 0.3
XO2H + 0.3 RO2 + 0.24 GLY + 0.06 MGLY +
0.29 PAR + 0.08 AACD + 0.08 H2O2

k = 4.60E‐13 exp(‐1155/T)

9.54E‐15

k = 2.30E‐11

2.30E‐11

k = 1.05E‐11 exp(519/T)

5.99E‐11

k = 4.70E‐15 exp(‐1013/T)

1.57E‐16

k = 3.70E‐13

3.70E‐13

k = 2.70E‐11 exp(390/T)
k = 3.00E‐11

9.99E‐11
3.00E‐11

k = 2.39E‐12 exp(365/T)

8.13E‐12

k = 7.43E‐13 exp(700/T)

7.78E‐12

k = k(ref) K
k(ref) = k(58)
K = 1.00E+0
k = k(ref) K
k(ref) = k(70)
K = 1.00E+0
k = 3.30E+9 exp(‐8300/T)
k = 1.03E‐14 exp(‐1995/T)

1.30E‐11

2.64E‐3
1.27E‐17

k = 3.03E‐12 exp(‐448/T)

6.74E‐13

k = 5.58E‐12 exp(511/T)

3.10E‐11

IOLE + NO3 = 0.5 NO2 + 0.5 NTR1 + 0.48
XO2 + 0.48 XO2H + 0.04 XO2N + RO2 + 0.5
ALD2 + 0.625 ALDX + PAR
ISOP + OH = ISO2 + RO2
ISOP + O = 0.75 ISPD + 0.5 FORM + 0.25
XO2 + 0.25 RO2 + 0.25 HO2 + 0.25 CXO3 +
0.25 PAR
ISO2 + NO = 0.1 INTR + 0.9 NO2 + 0.673
FORM + 0.9 ISPD + 0.818 HO2 + 0.082
XO2H + 0.082 RO2
ISO2 + HO2 = 0.88 ISPX + 0.12 OH + 0.12
HO2 + 0.12 FORM + 0.12 ISPD
ISO2 + C2O3 = 0.598 FORM + 1 ISPD +
0.728 HO2 + 0.072 XO2H + 0.8 MEO2 + 0.2
AACD + 0.872 RO2
ISO2 + RO2 = 0.598 FORM + 1 ISPD + 0.728
HO2 + 0.072 XO2H + 0.072 RO2
ISO2 = HO2 + HPLD
ISOP + O3 = 0.6 FORM + 0.65 ISPD + 0.15
ALDX + 0.2 CXO3 + 0.35 PAR + 0.266 OH +
0.2 XO2 + 0.2 RO2 + 0.066 HO2 + 0.066 CO
ISOP + NO3 = 0.35 NO2 + 0.65 NTR2 + 0.64
XO2H + 0.33 XO2 + 0.03 XO2N + RO2 + 0.35
FORM + 0.35 ISPD
ISPD + OH = 0.022 XO2N + 0.521 XO2 +
0.115 MGLY + 0.115 MEO2 + 0.269 GLYD +
0.269 C2O3 + 0.457 OPO3 + 0.117 PAR +
0.137 ACET + 0.137 CO + 0.137 HO2 + 0.658
RO2

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

237

1.00E‐17

3.48E‐13

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
Appendix A: Mechanism 2 – CB6r2

Number
159

Reactants and Products
ISPD + O3 = 0.04 ALD2 + 0.231 FORM +
0.531 MGLY + 0.17 GLY + 0.17 ACET + 0.543
CO + 0.461 OH + 0.15 FACD + 0.398 HO2 +
0.143 C2O3

Rate Constant Expression
k = 3.88E‐15 exp(‐1770/T)

k298
1.02E‐17

160

ISPD + NO3 = 0.717 HNO3 + 0.142 NTR2 +
0.142 NO2 + 0.142 XO2 + 0.142 XO2H +
0.113 GLYD + 0.113 MGLY + 0.717 PAR +
0.717 CXO3 + 0.284 RO2

k = 4.10E‐12 exp(‐1860/T)

7.98E‐15

161

ISPD = 0.76 HO2 + 0.34 XO2H + 0.16 XO2 +
0.34 MEO2 + 0.208 C2O3 + 0.26 FORM +
0.24 OLE + 0.24 PAR + 0.17 ACET + 0.128
GLYD + 0.84 RO2

Photolysis

1.60E‐5

162

ISPX + OH = 0.904 EPOX + 0.933 OH + 0.067
ISO2 + 0.067 RO2 + 0.029 IOLE + 0.029
ALDX
HPLD = OH + ISPD
HPLD + NO3 = HNO3 + ISPD
EPOX + OH = EPX2 + RO2
EPX2 + HO2 = 0.275 GLYD + 0.275 GLY +
0.275 MGLY + 1.125 OH + 0.825 HO2 +
0.375 FORM + 0.074 FACD + 0.251 CO +
2.175 PAR
EPX2 + NO = 0.275 GLYD + 0.275 GLY +
0.275 MGLY + 0.125 OH + 0.825 HO2 +
0.375 FORM + NO2 + 0.251 CO + 2.175 PAR
EPX2 + C2O3 = 0.22 GLYD + 0.22 GLY + 0.22
MGLY + 0.1 OH + 0.66 HO2 + 0.3 FORM +
0.2 CO + 1.74 PAR + 0.8 MEO2 + 0.2 AACD +
0.8 RO2

k = 2.23E‐11 exp(372/T)

7.77E‐11

Photolysis
k = 6.00E‐12 exp(‐1860/T)
k = 5.78E‐11 exp(‐400/T)
k = 7.43E‐13 exp(700/T)

4.41E‐4
1.17E‐14
1.51E‐11
7.78E‐12

k = 2.39E‐12 exp(365/T)

8.13E‐12

k = k(ref) K
k(ref) = k(58)
K = 1.00E+0

1.30E‐11

163
164
165
166

167

168

169

EPX2 + RO2 = 0.275 GLYD + 0.275 GLY +
k = k(ref) K
0.275 MGLY + 0.125 OH + 0.825 HO2 +
k(ref) = k(70)
0.375 FORM + 0.251 CO + 2.175 PAR + RO2
K = 1.00E+0
170
INTR + OH = 0.63 XO2 + 0.37 XO2H + RO2 + k = 3.10E‐11
0.444 NO2 + 0.185 NO3 + 0.104 INTR +
0.592 FORM + 0.331 GLYD + 0.185 FACD +
2.7 PAR + 0.098 OLE + 0.078 ALDX + 0.266
NTR2
171
TERP + O = 0.15 ALDX + 5.12 PAR
k = 3.60E‐11
172
TERP + OH = 0.75 XO2H + 0.5 XO2 + 0.25
k = 1.50E‐11 exp(449/T)
XO2N + 1.5 RO2 + 0.28 FORM + 1.66 PAR +
0.47 ALDX
173
TERP + O3 = 0.57 OH + 0.07 XO2H + 0.69
k = 1.20E‐15 exp(‐821/T)
XO2 + 0.18 XO2N + 0.94 RO2 + 0.24 FORM
+ 0.001 CO + 7 PAR + 0.21 ALDX + 0.39
CXO3
174
TERP + NO3 = 0.47 NO2 + 0.28 XO2H + 0.75 k = 3.70E‐12 exp(175/T)
XO2 + 0.25 XO2N + 1.28 RO2 + 0.47 ALDX +
0.53 NTR2
175
BENZ + OH = 0.53 CRES + 0.352 BZO2 +
k = 2.30E‐12 exp(‐190/T)
0.352 RO2 + 0.118 OPEN + 0.118 OH + 0.53
HO2
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS
238

3.48E‐13

3.10E‐11

3.60E‐11
6.77E‐11

7.63E‐17

6.66E‐12

1.22E‐12

www.camx.com

March 2016

Number
176
177

CAMx User’s Guide Version 6.3
Appendix A: Mechanism 2 – CB6r2

Reactants and Products
BZO2 + NO = 0.918 NO2 + 0.082 NTR2 +
0.918 GLY + 0.918 OPEN + 0.918 HO2
BZO2 + C2O3 = GLY + OPEN + HO2 + MEO2
+ RO2

178
179

BZO2 + HO2 =
BZO2 + RO2 = GLY + OPEN + HO2 + RO2

180

TOL + OH = 0.18 CRES + 0.65 TO2 + 0.72
RO2 + 0.1 OPEN + 0.1 OH + 0.07 XO2H +
0.18 HO2
TO2 + NO = 0.86 NO2 + 0.14 NTR2 + 0.417
GLY + 0.443 MGLY + 0.66 OPEN + 0.2 XOPN
+ 0.86 HO2
TO2 + C2O3 = 0.48 GLY + 0.52 MGLY + 0.77
OPEN + 0.23 XOPN + HO2 + MEO2 + RO2

181

182

183
184

TO2 + HO2 =
TO2 + RO2 = 0.48 GLY + 0.52 MGLY + 0.77
OPEN + 0.23 XOPN + HO2 + RO2

185

XYL + OH = 0.155 CRES + 0.544 XLO2 +
0.602 RO2 + 0.244 XOPN + 0.244 OH +
0.058 XO2H + 0.155 HO2
XLO2 + NO = 0.86 NO2 + 0.14 NTR2 + 0.221
GLY + 0.675 MGLY + 0.3 OPEN + 0.56 XOPN
+ 0.86 HO2
XLO2 + HO2 =
XLO2 + C2O3 = 0.26 GLY + 0.77 MGLY +
0.35 OPEN + 0.65 XOPN + HO2 + MEO2 +
RO2
XLO2 + RO2 = 0.26 GLY + 0.77 MGLY + 0.35
OPEN + 0.65 XOPN + HO2 + RO2

186

187
188

189

190

191

192
193
194
195
196
197

CRES + OH = 0.025 GLY + 0.025 OPEN +
HO2 + 0.2 CRO + 0.732 CAT1 + 0.02 XO2N +
0.02 RO2
CRES + NO3 = 0.3 CRO + HNO3 + 0.48 XO2 +
0.12 XO2H + 0.24 GLY + 0.24 MGLY + 0.48
OPO3 + 0.1 XO2N + 0.7 RO2
CRO + NO2 = CRON
CRO + HO2 = CRES
CRON + OH = NTR2 + 0.5 CRO
CRON + NO3 = NTR2 + 0.5 CRO + HNO3
CRON = HONO + HO2 + FORM + OPEN
XOPN = 0.4 GLY + XO2H + 0.7 HO2 + 0.7 CO
+ 0.3 C2O3

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

Rate Constant Expression
k = 2.70E‐12 exp(360/T)

k298
9.04E‐12

k = k(ref) K
k(ref) = k(58)
K = 1.00E+0
k = 1.90E‐13 exp(1300/T)
k = k(ref) K
k(ref) = k(70)
K = 1.00E+0
k = 1.80E‐12 exp(340/T)

1.30E‐11

k = 2.70E‐12 exp(360/T)

9.04E‐12

k = k(ref) K
k(ref) = k(58)
K = 1.00E+0
k = 1.90E‐13 exp(1300/T)
k = k(ref) K
k(ref) = k(70)
K = 1.00E+0
k = 1.85E‐11

1.30E‐11

1.85E‐11

k = 2.70E‐12 exp(360/T)

9.04E‐12

k = 1.90E‐13 exp(1300/T)
k = k(ref) K
k(ref) = k(58)
K = 1.00E+0
k = k(ref) K
k(ref) = k(70)
K = 1.00E+0
k = 1.70E‐12 exp(950/T)

1.49E‐11
1.30E‐11

4.12E‐11

k = 1.40E‐11

1.40E‐11

k = 2.10E‐12
k = 5.50E‐12
k = 1.53E‐12
k = 3.80E‐12
Photolysis
Photolysis

2.10E‐12
5.50E‐12
1.53E‐12
3.80E‐12
9.45E‐5
5.04E‐4

239

1.49E‐11
3.48E‐13

5.63E‐12

1.49E‐11
3.48E‐13

3.48E‐13

www.camx.com

March 2016

Number
198

CAMx User’s Guide Version 6.3
Appendix A: Mechanism 2 – CB6r2

208

Reactants and Products
XOPN + OH = MGLY + 0.4 GLY + 2 XO2H + 2
RO2
XOPN + O3 = 1.2 MGLY + 0.5 OH + 0.6 C2O3
+ 0.1 ALD2 + 0.5 CO + 0.3 XO2H + 0.3 RO2
XOPN + NO3 = 0.5 NO2 + 0.5 NTR2 + 0.45
XO2H + 0.45 XO2 + 0.1 XO2N + RO2 + 0.25
OPEN + 0.25 MGLY
OPEN = OPO3 + HO2 + CO
OPEN + OH = 0.6 OPO3 + 0.4 XO2H + 0.4
RO2 + 0.4 GLY
OPEN + O3 = 1.4 GLY + 0.24 MGLY + 0.5 OH
+ 0.12 C2O3 + 0.08 FORM + 0.02 ALD2 +
1.98 CO + 0.56 HO2
OPEN + NO3 = OPO3 + HNO3
CAT1 + OH = 0.14 FORM + 0.2 HO2 + 0.5
CRO
CAT1 + NO3 = CRO + HNO3
OPO3 + NO = NO2 + 0.5 GLY + 0.5 CO + 0.8
HO2 + 0.2 CXO3
OPO3 + NO2 = OPAN

209

OPAN = OPO3 + NO2

210

OPO3 + HO2 = 0.41 PACD + 0.15 AACD +
0.15 O3 + 0.44 ALDX + 0.44 XO2H + 0.44
RO2 + 0.44 OH
OPO3 + C2O3 = MEO2 + XO2 + ALDX + 2
RO2

199
200

201
202
203

204
205
206
207

211

212

OPO3 + RO2 = 0.8 XO2H + 0.8 ALDX + 1.8
RO2 + 0.2 AACD

213

OPAN + OH = 0.5 NO2 + 0.5 GLY + CO + 0.5
NTR2
PANX + OH = ALD2 + NO2
NTR2 = HNO3
ECH4 + OH = MEO2 + RO2

214
215
216

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

Rate Constant Expression
k = 9.00E‐11

k298
9.00E‐11

k = 1.08E‐16 exp(‐500/T)

2.02E‐17

k = 3.00E‐12

3.00E‐12

Photolysis
k = 4.40E‐11

5.04E‐4
4.40E‐11

k = 5.40E‐17 exp(‐500/T)

1.01E‐17

k = 3.80E‐12
k = 5.00E‐11

3.80E‐12
5.00E‐11

k = 1.70E‐10
k = 1.00E‐11

1.70E‐10
1.00E‐11

k = k(ref) K
k(ref) = k(54)
K = 1.00E+0
k = k(ref) K
k(ref) = k(55)
K = 1.00E+0
k = k(ref) K
k(ref) = k(57)
K = 1.00E+0
k = k(ref) K
k(ref) = k(59)
K = 1.00E+0
k = k(ref) K
k(ref) = k(58)
K = 1.00E+0
k = 3.60E‐11

9.40E‐12

3.60E‐11

k = 3.00E‐12
k = 2.30E‐5
k = 1.85E‐12 exp(‐1690/T)

3.00E‐12
2.30E‐5
6.37E‐15

240

2.98E‐4

1.39E‐11

1.55E‐11

1.30E‐11

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
Appendix B: Mechanism 3 – CB6r2 with Halogen Chemistry

Appendix B
CAMx Mechanism 3: CB6r2 with Halogen Chemistry
Table B‐1. Listing of the CB6r2 halogen mechanism (see Table A‐1 for a complete listing of
CB6r2). k298 is the rate constant at 298 K and 1 atmosphere using units in molecules/cm3 and
1/s. For photolysis reactions k298 shows the photolysis rate at a solar zenith angle of 60° and
height of 600 m MSL/AGL. See Table B‐2 for species names. See Section 3.1 on temperature
and pressure dependencies.
Number
1
2
3
4
5
6
7

8

9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29

30
31
32

Reactants and Products
CL2 = 2 CL
HOCL = CL + OH
CL + O3 = CLO
CLO + CLO = 0.3 CL2 + 1.4 CL
CLO + NO = CL + NO2
CLO + HO2 = HOCL
CLO + NO2 = CLN3

CLN3 = CLO + NO2

CLN3 = CLO + NO2
CLN3 = CL + NO3
CLN3 + H2O = HOCL + HNO3
OH + HCL = CL
OH + FMCL = CL + CO
FMCL = CL + CO + HO2
CLO + MEO2 = CL + FORM + HO2
CL + CH4 = HCL + MEO2
CL + PAR = HCL
CL + ETHA = HCL + 0.991 ALD2 + 0.991
XO2H + 0.009 XO2N + RO2
CL + PRPA = HCL + ACET + 0.97 XO2H +
0.03 XO2N + RO2
CL + ISOP = FMCL + ISPD + 0.96 XO2H +
0.04 XO2N + RO2
HCL + N2O5 = CLN2 + HNO3
CLN2 = CL + NO2
BR2 = 2 BR
HOBR = BR + OH
BR2 + OH = HOBR + BR
HBR + OH = BR
BR + O3 = BRO
BR + HO2 = HBR
BR + NO2 = BRN2

BR + NO3 = BRO + NO2
BRO = BR + O
BRO + HO2 = HOBR

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

Rate Constant Expression
Photolysis
Photolysis
k = 2.30E‐11 exp(‐200/T)
k = 1.63E‐14
k = 6.40E‐12 exp(290/T)
k = 2.70E‐12 exp(220/T)
Falloff: F=0.6; n=1
k(0) = 1.80E‐31 (T/300)^‐3.4
k(inf) = 1.50E‐11 (T/300)^‐1.9
Falloff: F=0.6; n=1
k(0) = 4.48E‐5 (T/300)^‐1 exp(‐12530/T)
k(inf) = 3.71E+15 (T/300)^3.5 exp(‐
12530/T)
Photolysis
Photolysis
k = 2.50E‐22
k = 6.58E‐13 (T/300)^1.2 exp(58/T)
k = 3.67E‐11 exp(‐1419/T)
Photolysis
k = 4.10E‐13 exp(‐800/T)
k = 6.60E‐12 exp(‐1240/T)
k = 5.00E‐11
k = 8.30E‐11 exp(‐100/T)

4.97E‐6
4.67E‐4
2.50E‐22
7.93E‐13
3.14E‐13
6.10E‐8
2.80E‐14
1.03E‐13
5.00E‐11
5.93E‐11

k = 1.40E‐10

1.40E‐10

k = 4.30E‐10

4.30E‐10

k = 6.00E‐13
Photolysis
Photolysis
Photolysis
k = 5.40E‐11 exp(180/T)
k = 5.50E‐12 exp(‐200/T)
k = 1.60E‐11 exp(780/T)
k = 4.80E‐12 exp(310/T)
Falloff: F=0.6; n=1
k(0) = 4.20E‐31 (T/300)^‐2.4
k(inf) = 2.70E‐11
k = 1.60E‐11
Photolysis
k = 4.50E‐12 exp(460/T)

6.00E‐13
2.86E‐4
2.79E‐2
1.51E‐3
9.88E‐11
2.81E‐12
2.19E‐10
1.36E‐11
4.89E‐12

241

k298
1.56E‐3
1.34E‐4
1.18E‐11
1.63E‐14
1.69E‐11
5.65E‐12
2.34E‐12

3.11E‐4

1.60E‐11
2.05E‐2
2.11E‐11
www.camx.com

March 2016

CAMx User’s Guide Version 6.3
Appendix B: Mechanism 3 – CB6r2 with Halogen Chemistry

Number
33
34
35
36
37

Reactants and Products
BRO + OH = BR + HO2
BRO + BRO = 2 BR
BRO + BRO = BR2
BRO + NO = BR + NO2
BRO + NO2 = BRN3

38
39
40
41
42
43

BRN2 = BR + NO2
BRN3 = BR + NO3
BRN3 + H2O = HOBR + HNO3
FMBR + OH = BR + CO
FMBR = BR + CO + HO2
BRO + MEO2 = 0.75 HOBR + 0.25 BR +
FORM
BR + FORM = HBR + CO + HO2
BR + ALD2 = HBR + C2O3
BR + OLE = FMBR + ALD2 + XO2H ‐ 1 PAR
+ RO2
BR + ISOP = FMBR + ISPD + 0.96 XO2H +
0.04 XO2N + RO2
I2 = 2 I
HOI = I + OH
I2 + OH = I + HOI
I2 + NO3 = I + INO3
HI + OH = I
I + O3 = IO
I + HO2 = HI
I + NO2 = INO2

44
45
46
47
48
49
50
51
52
53
54
55

56
57
58
59
60

IO = I + O
IO + IO = 0.4 I + 0.4 OIO + 0.6 I2O2
IO + HO2 = HOI
IO + NO = I + NO2
IO + NO2 = INO3

61
62
63

HOI + OH = IO
OIO = I
OIO + OH = HIO3

64
65
66
67
68
69
70
71
72

OIO + IO = IXOY
OIO + OIO = IXOY
OIO + NO = IO + NO2
I2O2 = I + OIO
I2O2 + O3 = IXOY
INO2 = I + NO2
INO2 + INO2 = I2 + 2 NO2
INO3 = I + NO3
INO3 + H2O = HOI + HNO3

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

Rate Constant Expression
k = 1.70E‐11 exp(250/T)
k = 2.40E‐12 exp(40/T)
k = 2.80E‐14 exp(860/T)
k = 8.80E‐12 exp(260/T)
Falloff: F=0.6; n=1
k(0) = 5.20E‐31 (T/300)^‐3.2
k(inf) = 6.90E‐12
Photolysis
Photolysis
k = 2.50E‐22
k = 5.00E‐12
Photolysis
k = 4.10E‐13 exp(‐800/T)

k298
3.93E‐11
2.74E‐12
5.02E‐13
2.11E‐11
2.81E‐12

k = 7.70E‐11 exp(‐580/T)
k = 1.80E‐11 exp(‐460/T)
k = 3.60E‐12

1.10E‐11
3.84E‐12
3.60E‐12

k = 5.00E‐12

5.00E‐12

Photolysis
Photolysis
k = 2.10E‐10
k = 1.50E‐12
k = 1.60E‐11 exp(440/T)
k = 2.10E‐11 exp(‐830/T)
k = 1.50E‐11 exp(‐1090/T)
Falloff: F=0.63; n=1
k(0) = 3.00E‐31 (T/300)^‐1
k(inf) = 6.60E‐11
Photolysis
k = 5.40E‐11 exp(180/T)
k = 1.40E‐11 exp(540/T)
k = 7.15E‐12 exp(300/T)
Falloff: F=0.4; n=1
k(0) = 7.70E‐31 (T/300)^‐5
k(inf) = 1.60E‐11
k = 5.00E‐12
Photolysis
Falloff: F=0.3; n=1
k(0) = 1.50E‐27 (T/300)^‐3.93
k(inf) = 5.50E‐10 exp(46/T)
k = 1.00E‐10
k = 1.50E‐10
k = 1.10E‐12 exp(542/T)
k = 1.00E+1
k = 1.00E‐12
Photolysis
k = 4.70E‐13 exp(‐1670/T)
Photolysis
k = 2.50E‐22

1.30E‐1
6.36E‐2
2.10E‐10
1.50E‐12
7.00E‐11
1.30E‐12
3.87E‐13
5.24E‐12

242

3.21E‐3
9.76E‐4
2.50E‐22
5.00E‐12
4.15E‐6
2.80E‐14

1.18E‐1
9.88E‐11
8.57E‐11
1.96E‐11
3.55E‐12

5.00E‐12
1.28E‐1
4.72E‐10

1.00E‐10
1.50E‐10
6.78E‐12
1.00E+1
1.00E‐12
3.21E‐3
1.73E‐15
1.25E‐2
2.50E‐22
www.camx.com

March 2016

Number
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88

Reactants and Products
CLO + BRO = CL + BR
CLO + IO = CL + I
BRO + IO = BR + I
CH3I = I + MEO2
MI2 = 2 I + FORM
MIB = I + BR + FORM
MIC = I + CL + FORM
MB3 = 3 BR + HO2 + CO
MB3 + OH = 3 BR + CO
MB2 + OH = 2 BR + HO2 + CO
MBC + OH = BR + MEO2
MBC2 + OH = BR + MEO2
MB2C + OH = BR + MEO2
IALK = I + ALDX + XO2H + RO2
SSCL + HNO3 = HCL + SSN3
SSBR + HOBR = BR2

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

CAMx User’s Guide Version 6.3
Appendix B: Mechanism 3 – CB6r2 with Halogen Chemistry
Rate Constant Expression
k = 4.70E‐12 exp(320/T)
k = 4.70E‐12 exp(280/T)
k = 1.50E‐11 exp(510/T)
Photolysis
Photolysis
Photolysis
Photolysis
Photolysis
k = 1.35E‐12 exp(‐600/T)
k = 2.00E‐12 exp(‐840/T)
k = 2.35E‐12 exp(‐1300/T)
k = 9.00E‐13 exp(‐600/T)
k = 9.00E‐13 exp(‐600/T)
Photolysis
k = 1.00E‐12
k = 1.00E‐12

243

k298
1.38E‐11
1.20E‐11
8.31E‐11
3.19E‐6
4.69E‐3
2.53E‐4
7.48E‐5
4.64E‐7
1.80E‐13
1.19E‐13
3.00E‐14
1.20E‐13
1.20E‐13
5.88E‐7
1.00E‐12
1.00E‐12

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
Appendix B: Mechanism 3 – CB6r2 with Halogen Chemistry

Table B‐2. Chemical species included in CB6r2h.
Species
CL2
CL
CLO
HCL
HOCL
CLN2
CLN3
FMCL
BR2
BR
BRO
HBR
HOBR
BRN2
BRN3
FMBR
I2
I
IO
OIO
I2O2
IXOY
HI
HOI
HIO3
INO2
INO3
CH3I
MI2
MIB
MIC
MBC
MB2
MBC2
MB2C
MB3
IALK
SSCL
SSBR
SSN3

Description
Molecular chlorine
Chlorine atom
Chlorine monoxide
Hydrogen chloride
Hypochlorous acid
Nitryl chloride: ClNO2
Chlorine nitrate: ClONO2
Formyl chloride: HC(O)Cl
Molecular bromine
Bromine atom
Bromine monoxide
Hydrogen bromide
Hypobromous acid
Nitryl broride: BrNO2
Bromine nitrate: BrONO2
Formyl bromide: HC(O)Br
Molecular iodine
Iodine atom
Iodine monoxide
Iodine dioxide
Diiodine dioxide
Condensable iodine oxides (> I2O2)
Hydrogen iodide
Hypoiodous acid
Iodic acid: HONO2
Nitryl iodide: INO2
Iodine nitrate: IONO2
Iodomethane
Diiodomethane: CH2I2
Bromoiodomethane: CH2BrI
Chloroiodomethane: CH2ClI
Chlorobromomethane: CH2ClBr
Dibromomethane: CH2Br2
Dichlorobromomethane: CHCl2Br
Chlorodibromomethane: CHClBr2
Bromoform CHBr3
Alkyl iodides
Pseudo gas‐phase species for sea salt chloride
Pseudo gas‐phase species for sea salt bromide
Pseudo gas‐phase species for sea salt nitrate

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

244

C

H

1
1

1

1

Constituents
O
N
Cl
2
1
1
1
1
1
1
2
1
3
1
1
1

1

1

1
2
3
1

1
2
2
3
1
1
1

1
1
1
1
1
1
1
1
1
3
0
0
0

I

2
1
1
1
1
1
1
1

1
1
1

Br

1
3
2
3

3
2
2
2
2
2
3
3
1
7

1
1
1
2
1

2
1
1
1
2
2
1
1
1
1
1
1
2
1
1

1
2
1
2
3
1

1
1
3

1

Mol. Wt.
70.9
35.5
51.4
36.5
52.4
81.4
97.4
64.5
159.8
79.9
95.9
80.9
96.9
125.9
141.9
108.9
253.8
126.9
142.9
158.9
285.8
301.8
127.9
143.9
175.9
172.9
188.9
141.9
267.8
220.8
176.4
129.4
173.8
165.8
210.3
252.7
170.0
35.5
79.9
62.0

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
Appendix C: Mechanism 6 – CB05

Appendix C
CAMx Mechanism 6: CB05 Gas‐Phase Chemistry
Table C‐1. Reactions and rate constant expressions for the CB05 mechanism. k298 is the rate
constant at 298 K and 1 atmosphere using units in molecules/cm3 and 1/s. See Table 5‐2 for
species names. See Section 3.1 on temperature and pressure dependencies.
Number
1
2
3
4
5

Reactants and Products
NO2 = NO + O
O + O2 + M = O3 + M
O3 + NO = NO2
O + NO2 = NO
O + NO2 = NO3

6

O + NO = NO2

7
8
9
10
11
12
13
14
15
16
17
18

NO2 + O3 = NO3
O3 = O
O3 = O1D
O1D + M = O + M
O1D + H2O = 2 OH
O3 + OH = HO2
O3 + HO2 = OH
NO3 = NO2 + O
NO3 = NO
NO3 + NO = 2 NO2
NO3 + NO2 = NO + NO2
NO3 + NO2 = N2O5

19
20
21

N2O5 + H2O = 2 HNO3
N2O5 + H2O + H2O = 2 HNO3
N2O5 = NO3 + NO2

22
23
24

NO + NO + O2 = 2 NO2
NO + NO2 + H2O = 2 HONO
NO + OH = HONO

25
26
27
28

HONO = NO + OH
OH + HONO = NO2
HONO + HONO = NO + NO2
NO2 + OH = HNO3

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

Rate Constant Expression
Photolysis
k = 6.00E‐34 (T/300)^‐2.4
k = 3.00E‐12 exp(‐1500/T)
k = 5.60E‐12 exp(180/T)
Falloff: F=0.6; n=1
k(0) = 2.50E‐31 (T/300)^‐1.8 exp(‐2/T)
k(inf) = 2.20E‐11 (T/300)^‐0.7 exp(‐3/T)
Falloff: F=0.6; n=1
k(0) = 9.00E‐32 (T/300)^‐1.5
k(inf) = 3.00E‐11
k = 1.20E‐13 exp(‐2450/T)
Photolysis
Photolysis
k = 2.10E‐11 exp(102/T)
k = 2.20E‐10
k = 1.70E‐12 exp(‐940/T)
k = 1.00E‐14 exp(‐490/T)
Photolysis
Photolysis
k = 1.50E‐11 exp(170/T)
k = 4.50E‐14 exp(‐1260/T)
Falloff: F=0.6; n=1
k(0) = 2.00E‐30 (T/300)^‐4.4
k(inf) = 1.40E‐12 (T/300)^‐0.7
k = 2.50E‐22
k = 1.80E‐39
Falloff: F=0.45; n=1
k(0) = 1.00E‐3 (T/300)^‐3.5 exp(‐11000/T)
k(inf) = 9.70E+14 (T/300)^0.1 exp(‐
11080/T)
k = 3.30E‐39 exp(530/T)
k = 5.00E‐40
Falloff: F=0.6; n=1
k(0) = 7.00E‐31 (T/300)^‐2.6
k(inf) = 3.60E‐11 (T/300)^‐0.1
Photolysis
k = 1.80E‐11 exp(‐390/T)
k = 1.00E‐20
Falloff: F=0.6; n=1
k(0) = 2.00E‐30 (T/300)^‐3
k(inf) = 2.50E‐11

245

k298
5.89E‐3
6.10E‐34
1.95E‐14
1.02E‐11
3.26E‐12

1.66E‐12

3.23E‐17
3.34E‐4
8.95E‐6
2.96E‐11
2.20E‐10
7.25E‐14
1.93E‐15
1.51E‐1
1.64E‐2
2.65E‐11
6.56E‐16
1.18E‐12

2.50E‐22
1.80E‐39
5.28E‐2

1.95E‐38
5.00E‐40
7.41E‐12

1.05E‐3
4.86E‐12
1.00E‐20
1.05E‐11

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
Appendix C: Mechanism 6 – CB05

Number
29

Reactants and Products
OH + HNO3 = NO3

Rate Constant Expression
k = k1 + k3 [M] / (1 + k3 [M] / k2)
k1 = 2.40E‐14 exp(460/T)
k2 = 2.70E‐17 exp(2199/T)
k3 = 6.50E‐34 exp(1335/T)

k298
1.54E‐13

30
31

HO2 + NO = OH + NO2
HO2 + NO2 = PNA

8.10E‐12
1.77E‐12

32

PNA = HO2 + NO2

33
34

OH + PNA = NO2
HO2 + HO2 = H2O2

35

HO2 + HO2 + H2O = H2O2

36
37
38
39
40
41
42

H2O2 = 2 OH
OH + H2O2 = HO2
O1D + H2 = OH + HO2
OH + H2 = HO2
OH + O = HO2
OH + OH = O
OH + OH = H2O2

43
44
45
46
47
48
49
50
51

OH + HO2 =
HO2 + O = OH
H2O2 + O = OH + HO2
NO3 + O = NO2
NO3 + OH = HO2 + NO2
NO3 + HO2 = HNO3
NO3 + O3 = NO2
NO3 + NO3 = 2 NO2
PNA = 0.61 HO2 + 0.61 NO2 + 0.39 OH +
0.39 NO3
HNO3 = OH + NO2
N2O5 = NO2 + NO3
XO2 + NO = NO2
XO2N + NO = NTR
XO2 + HO2 = ROOH
XO2N + HO2 = ROOH
XO2 + XO2 =
XO2N + XO2N =
XO2 + XO2N =
NTR + OH = HNO3 + HO2 + 0.33 FORM +
0.33 ALD2 + 0.33 ALDX ‐ 0.66 PAR

k = 3.50E‐12 exp(250/T)
Falloff: F=0.6; n=1
k(0) = 3.00E‐31 (T/300)^‐3.2
k(inf) = 4.70E‐12
Falloff: F=0.6; n=1
k(0) = 4.10E‐5 exp(‐10650/T)
k(inf) = 4.80E+15 exp(‐11170/T)
k = 1.30E‐12 exp(380/T)
k = k1 + k2 [M]
k1 = 2.30E‐13 exp(600/T)
k2 = 1.70E‐33 exp(1000/T)
k = k1 + k2 [M]
k1 = 3.22E‐34 exp(2800/T)
k2 = 2.38E‐54 exp(3200/T)
Photolysis
k = 2.90E‐12 exp(‐160/T)
k = 1.10E‐10
k = 5.50E‐12 exp(‐2000/T)
k = 2.20E‐11 exp(120/T)
k = 4.20E‐12 exp(‐240/T)
Falloff: F=0.6; n=1
k(0) = 6.90E‐31 (T/300)^‐1
k(inf) = 2.60E‐11
k = 4.80E‐11 exp(250/T)
k = 3.00E‐11 exp(200/T)
k = 1.40E‐12 exp(‐2000/T)
k = 1.00E‐11
k = 2.20E‐11
k = 3.50E‐12
k = 1.00E‐17
k = 8.50E‐13 exp(‐2450/T)
Photolysis

52
53
54
55
56
57
58
59
60
61

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

Photolysis
Photolysis
k = 2.60E‐12 exp(365/T)
k = 2.60E‐12 exp(365/T)
k = 7.50E‐13 exp(700/T)
k = 7.50E‐13 exp(700/T)
k = 6.80E‐14
k = 6.80E‐14
k = 6.80E‐14
k = 5.90E‐13 exp(‐360/T)

246

8.31E‐2

4.65E‐12
2.92E‐12

6.58E‐30

3.78E‐6
1.70E‐12
1.10E‐10
6.69E‐15
3.29E‐11
1.88E‐12
6.29E‐12

1.11E‐10
5.87E‐11
1.70E‐15
1.00E‐11
2.20E‐11
3.50E‐12
1.00E‐17
2.28E‐16
2.53E‐6
2.55E‐7
2.52E‐5
8.85E‐12
8.85E‐12
7.86E‐12
7.86E‐12
6.80E‐14
6.80E‐14
6.80E‐14
1.76E‐13

www.camx.com

March 2016

Number
62

CAMx User’s Guide Version 6.3
Appendix C: Mechanism 6 – CB05

63

Reactants and Products
NTR = NO2 + HO2 + 0.33 FORM + 0.33 ALD2
+ 0.33 ALDX ‐ 0.66 PAR
SO2 + OH = SULF + HO2

64
65
66

ROOH + OH = XO2 + 0.5 ALD2 + 0.5 ALDX
ROOH = OH + HO2 + 0.5 ALD2 + 0.5 ALDX
OH + CO = HO2

67
68
69
70

72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89

OH + CH4 = MEO2
MEO2 + NO = FORM + HO2 + NO2
MEO2 + HO2 = MEPX
MEO2 + MEO2 = 1.37 FORM + 0.74 HO2 +
0.63 MEOH
MEPX + OH = 0.7 MEO2 + 0.3 XO2 + 0.3
HO2
MEPX = FORM + HO2 + OH
MEOH + OH = FORM + HO2
FORM + OH = HO2 + CO
FORM = 2 HO2 + CO
FORM = CO
FORM + O = OH + HO2 + CO
FORM + NO3 = HNO3 + HO2 + CO
FORM + HO2 = HCO3
HCO3 = FORM + HO2
HCO3 + NO = FACD + NO2 + HO2
HCO3 + HO2 = MEPX
FACD + OH = HO2
ALD2 + O = C2O3 + OH
ALD2 + OH = C2O3
ALD2 + NO3 = C2O3 + HNO3
ALD2 = MEO2 + CO + HO2
C2O3 + NO = MEO2 + NO2
C2O3 + NO2 = PAN

90

PAN = C2O3 + NO2

91
92

PAN = C2O3 + NO2
C2O3 + HO2 = 0.8 PACD + 0.2 AACD + 0.2
O3
C2O3 + MEO2 = 0.9 MEO2 + 0.9 HO2 +
FORM + 0.1 AACD
C2O3 + XO2 = 0.9 MEO2 + 0.1 AACD
C2O3 + C2O3 = 2 MEO2
PACD + OH = C2O3
PACD = MEO2 + OH

71

93
94
95
96
97

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

Rate Constant Expression
Photolysis

k298
1.06E‐6

Falloff: F=0.6; n=1
k(0) = 3.00E‐31 (T/300)^‐3.3
k(inf) = 1.50E‐12
k = 3.01E‐12 exp(190/T)
Photolysis
k = k1 + k2 [M]
k1 = 1.44E‐13
k2 = 3.43E‐33
k = 2.45E‐12 exp(‐1775/T)
k = 2.80E‐12 exp(300/T)
k = 4.10E‐13 exp(750/T)
k = 9.50E‐14 exp(390/T)

8.89E‐13

k = 3.80E‐12 exp(200/T)

7.43E‐12

Photolysis
k = 7.30E‐12 exp(‐620/T)
k = 9.00E‐12
Photolysis
Photolysis
k = 3.40E‐11 exp(‐1600/T)
k = 5.80E‐16
k = 9.70E‐15 exp(625/T)
k = 2.40E+12 exp(‐7000/T)
k = 5.60E‐12
k = 5.60E‐15 exp(2300/T)
k = 4.00E‐13
k = 1.80E‐11 exp(‐1100/T)
k = 5.60E‐12 exp(270/T)
k = 1.40E‐12 exp(‐1900/T)
Photolysis
k = 8.10E‐12 exp(270/T)
Falloff: F=0.3; n=1
k(0) = 2.70E‐28 (T/300)^‐7.1
k(inf) = 1.20E‐11 (T/300)^‐0.9
Falloff: F=0.3; n=1
k(0) = 4.90E‐3 exp(‐12100/T)
k(inf) = 5.40E+16 exp(‐13830/T)
Photolysis
k = 4.30E‐13 exp(1040/T)

2.72E‐6
9.12E‐13
9.00E‐12
1.40E‐5
2.43E‐5
1.58E‐13
5.80E‐16
7.90E‐14
1.51E+2
5.60E‐12
1.26E‐11
4.00E‐13
4.49E‐13
1.39E‐11
2.38E‐15
1.76E‐6
2.00E‐11
1.05E‐11

k = 2.00E‐12 exp(500/T)

1.07E‐11

k = 4.40E‐13 exp(1070/T)
k = 2.90E‐12 exp(500/T)
k = 4.00E‐13 exp(200/T)
Photolysis

1.60E‐11
1.55E‐11
7.83E‐13
0.00E+0

247

5.69E‐12
2.72E‐6
2.28E‐13

6.34E‐15
7.66E‐12
5.08E‐12
3.52E‐13

3.31E‐4

3.47E‐7
1.41E‐11

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
Appendix C: Mechanism 6 – CB05

Number
98
99
100
101
102
103
104

Reactants and Products
AACD + OH = MEO2
ALDX + O = CXO3 + OH
ALDX + OH = CXO3
ALDX + NO3 = CXO3 + HNO3
ALDX = MEO2 + CO + HO2
CXO3 + NO = ALD2 + NO2 + HO2 + XO2
CXO3 + NO2 = PANX

105

PANX = CXO3 + NO2

106
107
108

PANX = CXO3 + NO2
PANX + OH = ALD2 + NO2
CXO3 + HO2 = 0.8 PACD + 0.2 AACD + 0.2
O3
CXO3 + MEO2 = 0.9 ALD2 + 0.9 XO2 + HO2
+ 0.1 AACD + 0.1 FORM
CXO3 + XO2 = 0.9 ALD2 + 0.1 AACD
CXO3 + CXO3 = 2 ALD2 + 2 XO2 + 2 HO2
CXO3 + C2O3 = MEO2 + XO2 + HO2 + ALD2
OH + ETHA = 0.991 ALD2 + 0.991 XO2 +
0.009 XO2N + HO2
OH + ETOH = HO2 + 0.9 ALD2 + 0.05 ALDX +
0.1 FORM + 0.1 XO2
PAR + OH = 0.87 XO2 + 0.13 XO2N + 0.11
HO2 + 0.06 ALD2 ‐ 0.11 PAR + 0.76 ROR +
0.05 ALDX
ROR = 0.96 XO2 + 0.6 ALD2 + 0.94 HO2 ‐ 2.1
PAR + 0.04 XO2N + 0.02 ROR + 0.5 ALDX
ROR = HO2
ROR + NO2 = NTR
O + OLE = 0.2 ALD2 + 0.3 ALDX + 0.3 HO2 +
0.2 XO2 + 0.2 CO + 0.2 FORM + 0.01 XO2N
+ 0.2 PAR + 0.1 OH
OH + OLE = 0.8 FORM + 0.33 ALD2 + 0.62
ALDX + 0.8 XO2 + 0.95 HO2 ‐ 0.7 PAR
O3 + OLE = 0.18 ALD2 + 0.74 FORM + 0.32
ALDX + 0.22 XO2 + 0.1 OH + 0.33 CO + 0.44
HO2 ‐ 1 PAR
NO3 + OLE = NO2 + FORM + 0.91 XO2 +
0.09 XO2N + 0.56 ALDX + 0.35 ALD2 ‐ 1 PAR
O + ETH = FORM + 1.7 HO2 + CO + 0.7 XO2
+ 0.3 OH
OH + ETH = XO2 + 1.56 FORM + 0.22 ALDX
+ HO2

109
110
111
112
113
114
115

116
117
118
119

120
121

122
123
124

125

O3 + ETH = FORM + 0.63 CO + 0.13 HO2 +
0.13 OH + 0.37 FACD

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

Rate Constant Expression
k = 4.00E‐13 exp(200/T)
k = 1.30E‐11 exp(‐870/T)
k = 5.10E‐12 exp(405/T)
k = 6.50E‐15
Photolysis
k = 6.70E‐12 exp(340/T)
Falloff: F=0.3; n=1
k(0) = 2.70E‐28 (T/300)^‐7.1
k(inf) = 1.20E‐11 (T/300)^‐0.9
Falloff: F=0.3; n=1
k(0) = 4.90E‐3 exp(‐12100/T)
k(inf) = 5.40E+16 exp(‐13830/T)
Photolysis
k = 3.00E‐13
k = 4.30E‐13 exp(1040/T)

k298
7.83E‐13
7.02E‐13
1.99E‐11
6.50E‐15
6.96E‐6
2.10E‐11
1.05E‐11

k = 2.00E‐12 exp(500/T)

1.07E‐11

k = 4.40E‐13 exp(1070/T)
k = 2.90E‐12 exp(500/T)
k = 2.90E‐12 exp(500/T)
k = 8.70E‐12 exp(‐1070/T)

1.60E‐11
1.55E‐11
1.55E‐11
2.40E‐13

k = 6.90E‐12 exp(‐230/T)

3.19E‐12

k = 8.10E‐13

8.10E‐13

k = 1.00E+15 exp(‐8000/T)

2.19E+3

k = 1.60E+3
k = 1.50E‐11
k = 1.00E‐11 exp(‐280/T)

1.60E+3
1.50E‐11
3.91E‐12

k = 3.20E‐11

3.20E‐11

k = 6.50E‐15 exp(‐1900/T)

1.11E‐17

k = 7.00E‐13 exp(‐2160/T)

4.98E‐16

k = 1.04E‐11 exp(‐792/T)

7.29E‐13

Falloff: F=0.6; n=1
k(0) = 1.00E‐28 (T/300)^‐0.8
k(inf) = 8.80E‐12
k = 1.20E‐14 exp(‐2630/T)

8.15E‐12

248

3.31E‐4

3.47E‐7
3.00E‐13
1.41E‐11

1.76E‐18
www.camx.com

March 2016

Number
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140

141
142
143
144
145
146

147
148
149

150

CAMx User’s Guide Version 6.3
Appendix C: Mechanism 6 – CB05

Reactants and Products
NO3 + ETH = NO2 + XO2 + 2 FORM
IOLE + O = 1.24 ALD2 + 0.66 ALDX + 0.1
HO2 + 0.1 XO2 + 0.1 CO + 0.1 PAR
IOLE + OH = 1.3 ALD2 + 0.7 ALDX + HO2 +
XO2
IOLE + O3 = 0.65 ALD2 + 0.35 ALDX + 0.25
FORM + 0.25 CO + 0.5 O + 0.5 OH + 0.5 HO2
IOLE + NO3 = 1.18 ALD2 + 0.64 ALDX + HO2
+ NO2
TOL + OH = 0.44 HO2 + 0.08 XO2 + 0.36
CRES + 0.56 TO2
TO2 + NO = 0.9 NO2 + 0.9 HO2 + 0.9 OPEN
+ 0.1 NTR
TO2 = CRES + HO2
OH + CRES = 0.4 CRO + 0.6 XO2 + 0.6 HO2 +
0.3 OPEN
CRES + NO3 = CRO + HNO3
CRO + NO2 = NTR
CRO + HO2 = CRES
OPEN = C2O3 + HO2 + CO
OPEN + OH = XO2 + 2 CO + 2 HO2 + C2O3 +
FORM
OPEN + O3 = 0.03 ALDX + 0.62 C2O3 + 0.7
FORM + 0.03 XO2 + 0.69 CO + 0.08 OH +
0.76 HO2 + 0.2 MGLY
OH + XYL = 0.7 HO2 + 0.5 XO2 + 0.2 CRES +
0.8 MGLY + 1.1 PAR + 0.3 TO2
OH + MGLY = XO2 + C2O3
MGLY = C2O3 + HO2 + CO
O + ISOP = 0.75 ISPD + 0.5 FORM + 0.25
XO2 + 0.25 HO2 + 0.25 CXO3 + 0.25 PAR
OH + ISOP = 0.912 ISPD + 0.629 FORM +
0.991 XO2 + 0.912 HO2 + 0.088 XO2N
O3 + ISOP = 0.65 ISPD + 0.6 FORM + 0.2
XO2 + 0.066 HO2 + 0.266 OH + 0.2 CXO3 +
0.15 ALDX + 0.35 PAR + 0.066 CO
NO3 + ISOP = 0.2 ISPD + 0.8 NTR + XO2 +
0.8 HO2 + 0.2 NO2 + 0.8 ALDX + 2.4 PAR
NO2 + ISOP = 0.2 ISPD + 0.8 NTR + XO2 +
0.8 HO2 + 0.2 NO + 0.8 ALDX + 2.4 PAR
OH + ISPD = 1.565 PAR + 0.167 FORM +
0.713 XO2 + 0.503 HO2 + 0.334 CO + 0.168
MGLY + 0.252 ALD2 + 0.21 C2O3 + 0.25
CXO3 + 0.12 ALDX
O3 + ISPD = 0.114 C2O3 + 0.15 FORM +
0.85 MGLY + 0.154 HO2 + 0.268 OH + 0.064
XO2 + 0.02 ALD2 + 0.36 PAR + 0.225 CO

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

Rate Constant Expression
k = 3.30E‐12 exp(‐2880/T)
k = 2.30E‐11

k298
2.10E‐16
2.30E‐11

k = 1.00E‐11 exp(550/T)

6.33E‐11

k = 8.40E‐15 exp(‐1100/T)

2.09E‐16

k = 9.60E‐13 exp(‐270/T)

3.88E‐13

k = 1.80E‐12 exp(355/T)

5.92E‐12

k = 8.10E‐12

8.10E‐12

k = 4.20E+0
k = 4.10E‐11

4.20E+0
4.10E‐11

k = 2.20E‐11
k = 1.40E‐11
k = 5.50E‐12
Photolysis
k = 3.00E‐11

2.20E‐11
1.40E‐11
5.50E‐12
1.26E‐4
3.00E‐11

k = 5.40E‐17 exp(‐500/T)

1.01E‐17

k = 1.70E‐11 exp(116/T)

2.51E‐11

k = 1.70E‐11
Photolysis
k = 3.60E‐11

1.70E‐11
1.54E‐4
3.60E‐11

k = 2.54E‐11 exp(407.6/T)

9.97E‐11

k = 7.86E‐15 exp(‐1912/T)

1.29E‐17

k = 3.03E‐12 exp(‐448/T)

6.74E‐13

k = 1.50E‐19

1.50E‐19

k = 3.36E‐11

3.36E‐11

k = 7.10E‐18

7.10E‐18

249

www.camx.com

March 2016

Number
151

152

153
154
155

156

CAMx User’s Guide Version 6.3
Appendix C: Mechanism 6 – CB05

Reactants and Products
NO3 + ISPD = 0.357 ALDX + 0.282 FORM +
1.282 PAR + 0.925 HO2 + 0.643 CO + 0.85
NTR + 0.075 CXO3 + 0.075 XO2 + 0.15
HNO3
ISPD = 0.333 CO + 0.067 ALD2 + 0.9 FORM
+ 0.832 PAR + 1.033 HO2 + 0.7 XO2 + 0.967
C2O3
TERP + O = 0.15 ALDX + 5.12 PAR
TERP + OH = 0.75 HO2 + 1.25 XO2 + 0.25
XO2N + 0.28 FORM + 1.66 PAR + 0.47 ALDX
TERP + O3 = 0.57 OH + 0.07 HO2 + 0.76
XO2 + 0.18 XO2N + 0.24 FORM + 0.001 CO
+ 7 PAR + 0.21 ALDX + 0.39 CXO3
TERP + NO3 = 0.47 NO2 + 0.28 HO2 + 1.03
XO2 + 0.25 XO2N + 0.47 ALDX + 0.53 NTR

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

Rate Constant Expression
k = 1.00E‐15

k298
1.00E‐15

Photolysis

1.11E‐6

k = 3.60E‐11
k = 1.50E‐11 exp(449/T)

3.60E‐11
6.77E‐11

k = 1.20E‐15 exp(‐821/T)

7.63E‐17

k = 3.70E‐12 exp(175/T)

6.66E‐12

250

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
Appendix D: Mechanism 5 – SAPRC07TC

Appendix D
CAMx Mechanism 5: SAPRC07TC Gas‐Phase Chemistry
Table D‐1. Reactions and rate constants for the SAPRC07TC mechanism. k300 is the rate
constant at 300 K and 1 atmosphere using units in molecules/cm‐3 and 1/s. See Table D‐2 for
species names. See Section 3.1 on temperature and pressure dependencies.
Number
1
2
3
4

Reactants and Products
NO2 = NO + O3P
O3P + O2 + M = O3
O3P + O3 =
O3P + NO = NO2

5
6

O3P + NO2 = NO
O3P + NO2 = NO3

7
8
9
10
11

O3 + NO = NO2
O3 + NO2 = NO3
NO + NO3 = 2. NO2
NO + NO + O2 = 2. NO2
NO2 + NO3 = N2O5

12

N2O5 = NO2 + NO3

13
14
15
16
17
18
19
20
21
22

N2O5 + H2O = 2. HNO3
N2O5 + H2O + H2O = 2. HNO3
NO2 + NO3 = NO + NO2
NO3 = NO
NO3 = NO2 + O3P
O3 = O1D
O3 = O3P
O1D + H2O = 2. OH
O1D + M = O3P
OH + NO = HONO

23
24
25

HONO = OH + NO
OH + HONO = NO2
OH + NO2 = HNO3

26
27

OH + NO3 = HO2 + NO2
OH + HNO3 = NO3

28

HNO3 = OH + NO2

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

Rate Constant Expression
Photolysis
k = 5.68E‐34 (T/300)^‐2.6
k = 8.00E‐12 exp(‐2060/T)
Falloff: F=0.6; n=1
k(0) = 9.00E‐32 (T/300)^‐1.5
k(inf) = 3.00E‐11
k = 5.50E‐12 exp(188/T)
Falloff: F=0.6; n=1
k(0) = 2.50E‐31 (T/300)^‐1.8
k(inf) = 2.20E‐11 (T/300)^‐0.7
k = 3.00E‐12 exp(‐1500/T)
k = 1.40E‐13 exp(‐2470/T)
k = 1.80E‐11 exp(110/T)
k = 3.30E‐39 exp(530/T)
Falloff: F=0.35; n=1.33
k(0) = 3.60E‐30 (T/300)^‐4.1
k(inf) = 1.90E‐12 (T/300)^0.2
Falloff: F=0.35; n=1.33
k(0) = 1.30E‐3 (T/300)^‐3.5 exp(‐11000/T)
k(inf) = 9.70E+14 (T/300)^0.1 exp(‐11080/T)
k = 1.00E‐22
k = 0.00E+0
k = 4.50E‐14 exp(‐1260/T)
Photolysis
Photolysis
Photolysis
Photolysis
k = 1.63E‐10 exp(60/T)
k = 2.38E‐11 exp(96/T)
Falloff: F=0.6; n=1
k(0) = 7.00E‐31 (T/300)^‐2.6
k(inf) = 3.60E‐11 (T/300)^‐0.1
Photolysis
k = 2.50E‐12 exp(260/T)
Falloff: F=0.6; n=1
k(0) = 1.80E‐30 (T/300)^‐3
k(inf) = 2.80E‐11
k = 2.00E‐11
k = k1 + k3 [M] / (1 + k3 [M] / k2)
k1 = 2.40E‐14 exp(460/T)
k2 = 2.70E‐17 exp(2199/T)
k3 = 6.50E‐34 exp(1335/T)
Photolysis

251

k300
6.37E‐3
5.68E‐34
8.34E‐15
1.64E‐12

1.03E‐11
3.24E‐12

2.02E‐14
3.72E‐17
2.60E‐11
1.93E‐38
1.24E‐12

5.69E‐2

1.00E‐22
0.00E+0
6.75E‐16
1.98E‐2
1.56E‐1
9.47E‐6
3.40E‐4
1.99E‐10
3.28E‐11
7.31E‐12

9.88E‐4
5.95E‐12
1.05E‐11

2.00E‐11
1.51E‐13

2.55E‐7

www.camx.com

March 2016

Number Reactants and Products
29
OH + CO = HO2 + CO2

30
31
32

OH + O3 = HO2
HO2 + NO = OH + NO2
HO2 + NO2 = PNA

33

PNA = HO2 + NO2

34
35
36
37

PNA = 0.61 HO2 + 0.61 NO2 + 0.39 OH + 0.39
NO3
PNA + OH = NO2
HO2 + O3 = OH
HO2 + HO2 = H2O2

38

HO2 + HO2 + H2O = H2O2

39
40
41
42
43
44

NO3 + HO2 = 0.8 OH + 0.8 NO2 + 0.2 HNO3
NO3 + NO3 = 2. NO2
H2O2 = 2. OH
H2O2 + OH = HO2
OH + HO2 =
OH + SO2 = HO2 + SULF

45
46
47
48
49
50
51
52
53
54
55

OH + H2 = HO2
MEO2 + NO = NO2 + HCHO + HO2
MEO2 + HO2 = COOH
MEO2 + HO2 = HCHO
MEO2 + NO3 = HCHO + HO2 + NO2
MEO2 + MEO2 = MEOH + HCHO
MEO2 + MEO2 = 2. HCHO + 2. HO2
RO2C + NO = NO2
RO2C + HO2 =
RO2C + NO3 = NO2
RO2C + MEO2 = 0.5 HO2 + 0.75 HCHO + 0.25
MEOH
RO2C + RO2C =
RO2X + NO = XN
RO2X + HO2 =
RO2X + NO3 = NO2
RO2X + MEO2 = 0.5 HO2 + 0.75 HCHO + 0.25
MEOH
RO2X + RO2C =
RO2X + RO2X =

56
57
58
59
60
61
62

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

CAMx User’s Guide Version 6.3
Appendix D: Mechanism 5 – SAPRC07TC
Rate Constant Expression
k = k1 + k2 [M]
k1 = 1.44E‐13
k2 = 3.43E‐33
k = 1.70E‐12 exp(‐940/T)
k = 3.60E‐12 exp(270/T)
Falloff: F=0.6; n=1
k(0) = 2.00E‐31 (T/300)^‐3.4
k(inf) = 2.90E‐12 (T/300)^‐1.1
Falloff: F=0.6; n=1
k(0) = 3.72E‐5 (T/300)^‐2.4 exp(‐10650/T)
k(inf) = 5.42E+15 (T/300)^‐2.3 exp(‐11170/T)
Photolysis
k = 1.30E‐12 exp(380/T)
k = 2.03E‐16 (T/300)^4.57 exp(693/T)
k = k1 + k2 [M]
k1 = 2.20E‐13 exp(600/T)
k2 = 1.90E‐33 exp(980/T)
k = k1 + k2 [M]
k1 = 3.08E‐34 exp(2800/T)
k2 = 2.66E‐54 exp(3180/T)
k = 4.00E‐12
k = 8.50E‐13 exp(‐2450/T)
Photolysis
k = 1.80E‐12
k = 4.80E‐11 exp(250/T)
Falloff: F=0.6; n=1
k(0) = 3.30E‐31 (T/300)^‐4.3
k(inf) = 1.60E‐12
k = 7.70E‐12 exp(‐2100/T)
k = 2.30E‐12 exp(360/T)
k = 3.46E‐13 (T/300)^0.36 exp(780/T)
k = 3.34E‐14 (T/300)^‐3.53 exp(780/T)
k = 1.30E‐12
k = 6.39E‐14 (T/300)^‐1.8 exp(365/T)
k = 7.40E‐13 exp(‐520/T)
k = 2.60E‐12 exp(380/T)
k = 3.80E‐13 exp(900/T)
k = 2.30E‐12
k = 2.00E‐13

k300
2.28E‐13

7.41E‐14
8.85E‐12
1.12E‐12

1.07E‐1

3.17E‐6
4.61E‐12
2.05E‐15
2.84E‐12

6.09E‐30

4.00E‐12
2.41E‐16
3.78E‐6
1.80E‐12
1.10E‐10
9.49E‐13

7.02E‐15
7.64E‐12
4.66E‐12
4.50E‐13
1.30E‐12
2.16E‐13
1.31E‐13
9.23E‐12
7.63E‐12
2.30E‐12
2.00E‐13

k = 3.50E‐14
k = k(52)
k = k(53)
k = k(54)
k = k(55)

3.50E‐14
9.23E‐12
7.63E‐12
2.30E‐12
2.00E‐13

k = k(56)
k = k(56)

3.50E‐14
3.50E‐14

252

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
Appendix D: Mechanism 5 – SAPRC07TC

Number Reactants and Products
63
MCO3 + NO2 = PAN

64

PAN = MCO3 + NO2

65

PAN = 0.6 MCO3 + 0.6 NO2 + 0.4 MEO2 + 0.4
CO2 + 0.4 NO3
MCO3 + NO = MEO2 + CO2 + NO2
MCO3 + HO2 = 0.7 CO3H + 0.3 AACD + 0.3 O3
MCO3 + NO3 = MEO2 + CO2 + NO2
MCO3 + MEO2 = 0.1 AACD + HCHO + 0.9 HO2
+ 0.9 MEO2 + 0.9 CO2
MCO3 + RO2C = MEO2 + CO2
MCO3 + RO2X = MEO2 + CO2
MCO3 + MCO3 = 2. MEO2 + 2. CO2
RCO3 + NO2 = PAN2
PAN2 = RCO3 + NO2
PAN2 = 0.6 RCO3 + 0.6 NO2 + 0.4 RO2C + 0.4
XHO2 + 0.4 YRPX + 0.4 XCCH + 0.4 CO2 + 0.4
NO3
RCO3 + NO = NO2 + RO2C + XHO2 + YRPX +
XCCH + CO2
RCO3 + HO2 = 0.75 RO3H + 0.25 PACD + 0.25
O3
RCO3 + NO3 = NO2 + RO2C + XHO2 + YRPX +
XCCH + CO2
RCO3 + MEO2 = HCHO + HO2 + RO2C + XHO2
+ XCCH + YRPX + CO2
RCO3 + RO2C = RO2C + XHO2 + XCCH + YRPX +
CO2
RCO3 + RO2X = RO2C + XHO2 + XCCH + YRPX +
CO2
RCO3 + MCO3 = 2. CO2 + MEO2 + RO2C +
XHO2 + YRPX + XCCH
RCO3 + RCO3 = 2. RO2C + 2. XHO2 + 2. XCCH +
2. YRPX + 2. CO2
BZC3 + NO2 = PBZN
PBZN = BZC3 + NO2
PBZN = 0.6 BZC3 + 0.6 NO2 + 0.4 CO2 + 0.4
BZO + 0.4 RO2C + 0.4 NO3
BZC3 + NO = NO2 + CO2 + BZO + RO2C
BZC3 + HO2 = 0.75 RO3H + 0.25 PACD + 0.25
O3 + 4. XC
BZC3 + NO3 = NO2 + CO2 + BZO + RO2C
BZC3 + MEO2 = HCHO + HO2 + RO2C + BZO +
CO2
BZC3 + RO2C = RO2C + BZO + CO2
BZC3 + RO2X = RO2C + BZO + CO2
BZC3 + MCO3 = 2. CO2 + MEO2 + BZO + RO2C

66
67
68
69
70
71
72
73
74
75

76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

Rate Constant Expression
Falloff: F=0.3; n=1.41
k(0) = 2.70E‐28 (T/300)^‐7.1
k(inf) = 1.21E‐11 (T/300)^‐0.9
Falloff: F=0.3; n=1.41
k(0) = 4.90E‐3 exp(‐12100/T)
k(inf) = 4.00E+16 exp(‐13600/T)
Photolysis

k300
9.38E‐12

6.27E‐4

3.50E‐7

k = 7.50E‐12 exp(290/T)
k = 5.20E‐13 exp(980/T)
k = k(54)
k = 2.00E‐12 exp(500/T)

1.97E‐11
1.36E‐11
2.30E‐12
1.06E‐11

k = 4.40E‐13 exp(1070/T)
k = k(70)
k = 2.90E‐12 exp(500/T)
k = 1.21E‐11 (T/300)^‐1.07
k = 8.30E+16 exp(‐13940/T)
Photolysis

1.56E‐11
1.56E‐11
1.54E‐11
1.21E‐11
5.48E‐4
3.50E‐7

k = 6.70E‐12 exp(340/T)

2.08E‐11

k = k(67)

1.36E‐11

k = k(54)

2.30E‐12

k = k(69)

1.06E‐11

k = k(70)

1.56E‐11

k = k(70)

1.56E‐11

k = k(72)

1.54E‐11

k = k(72)

1.54E‐11

k = 1.37E‐11
k = 7.90E+16 exp(‐14000/T)
Photolysis

1.37E‐11
4.27E‐4
3.50E‐7

k = k(76)
k = k(67)

2.08E‐11
1.36E‐11

k = k(54)
k = k(69)

2.30E‐12
1.06E‐11

k = k(70)
k = k(70)
k = k(72)

1.56E‐11
1.56E‐11
1.54E‐11

253

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
Appendix D: Mechanism 5 – SAPRC07TC

Number Reactants and Products
94
BZC3 + RCO3 = 2. CO2 + 2. RO2C + XHO2 +
YRPX + XCCH + BZO
95
BZC3 + BZC3 = 2. BZO + 2. RO2C + 2. CO2
96
MAC3 + NO2 = MPAN
97
MPAN = MAC3 + NO2
98
MPAN = 0.6 MAC3 + 0.6 NO2 + 0.4 CO2 + 0.4
HCHO + 0.4 MCO3 + 0.4 NO3
99
MAC3 + NO = NO2 + CO2 + HCHO + MCO3
100
MAC3 + HO2 = 0.75 RO3H + 0.25 PACD + 0.25
O3 + XC
101
MAC3 + NO3 = NO2 + CO2 + HCHO + MCO3
102
MAC3 + MEO2 = 2. HCHO + HO2 + CO2 +
MCO3
103
MAC3 + RO2C = CO2 + HCHO + MCO3
104
MAC3 + RO2X = CO2 + HCHO + MCO3
105
MAC3 + MCO3 = 2. CO2 + MEO2 + HCHO +
MCO3
106
MAC3 + RCO3 = HCHO + MCO3 + RO2C +
XHO2 + YRPX + XCCH + 2. CO2
107
MAC3 + BZC3 = HCHO + MCO3 + BZO + RO2C
+ 2. CO2
108
MAC3 + MAC3 = 2. HCHO + 2. MCO3 + 2. CO2
109
TBUO + NO2 = RNO3 ‐ 2. XC
110
TBUO = ACET + MEO2
111
BZO + NO2 = NPHE
112
BZO + HO2 = CRES ‐ 1. XC
113
BZO = CRES + RO2C + XHO2 ‐ 1. XC
114
XHO2 + NO = NO + HO2
115
XHO2 + HO2 = HO2
116
XHO2 + NO3 = NO3 + HO2
117
XHO2 + MEO2 = MEO2 + 0.5 HO2
118
XHO2 + RO2C = RO2C + 0.5 HO2
119
XHO2 + RO2X = RO2X + 0.5 HO2
120
XHO2 + MCO3 = MCO3 + HO2
121
XHO2 + RCO3 = RCO3 + HO2
122
XHO2 + BZC3 = BZC3 + HO2
123
XHO2 + MAC3 = MAC3 + HO2
124
XOH + NO = NO + OH
125
XOH + HO2 = HO2
126
XOH + NO3 = NO3 + OH
127
XOH + MEO2 = MEO2 + 0.5 OH
128
XOH + RO2C = RO2C + 0.5 OH
129
XOH + RO2X = RO2X + 0.5 OH
130
XOH + MCO3 = MCO3 + OH
131
XOH + RCO3 = RCO3 + OH
132
XOH + BZC3 = BZC3 + OH
133
XOH + MAC3 = MAC3 + OH
134
XNO2 + NO = NO + NO2
135
XNO2 + HO2 = HO2 + XN
136
XNO2 + NO3 = NO3 + NO2
137
XNO2 + MEO2 = MEO2 + 0.5 NO2 + 0.5 XN
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

Rate Constant Expression
k = k(72)

k300
1.54E‐11

k = k(72)
k = k(73)
k = 1.60E+16 exp(‐13486/T)
Photolysis

1.54E‐11
1.21E‐11
4.80E‐4
3.50E‐7

k = k(76)
k = k(67)

2.08E‐11
1.36E‐11

k = k(54)
k = k(69)

2.30E‐12
1.06E‐11

k = k(70)
k = k(70)
k = k(72)

1.56E‐11
1.56E‐11
1.54E‐11

k = k(72)

1.54E‐11

k = k(72)

1.54E‐11

k = k(72)
k = 2.40E‐11
k = 7.50E+14 exp(‐8152/T)
k = 2.30E‐11 exp(150/T)
k = k(53)
k = 1.00E‐3
k = k(52)
k = k(53)
k = k(54)
k = k(55)
k = k(56)
k = k(56)
k = k(70)
k = k(70)
k = k(70)
k = k(70)
k = k(52)
k = k(53)
k = k(54)
k = k(55)
k = k(56)
k = k(56)
k = k(70)
k = k(70)
k = k(70)
k = k(70)
k = k(52)
k = k(53)
k = k(54)
k = k(55)

1.54E‐11
2.40E‐11
1.19E+3
3.79E‐11
7.63E‐12
1.00E‐3
9.23E‐12
7.63E‐12
2.30E‐12
2.00E‐13
3.50E‐14
3.50E‐14
1.56E‐11
1.56E‐11
1.56E‐11
1.56E‐11
9.23E‐12
7.63E‐12
2.30E‐12
2.00E‐13
3.50E‐14
3.50E‐14
1.56E‐11
1.56E‐11
1.56E‐11
1.56E‐11
9.23E‐12
7.63E‐12
2.30E‐12
2.00E‐13

254

www.camx.com

March 2016

Number
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188

CAMx User’s Guide Version 6.3
Appendix D: Mechanism 5 – SAPRC07TC

Reactants and Products
Rate Constant Expression
XNO2 + RO2C = RO2C + 0.5 NO2 + 0.5 XN
k = k(56)
XNO2 + RO2X = RO2X + 0.5 NO2 + 0.5 XN
k = k(56)
XNO2 + MCO3 = MCO3 + NO2
k = k(70)
XNO2 + RCO3 = RCO3 + NO2
k = k(70)
XNO2 + BZC3 = BZC3 + NO2
k = k(70)
XNO2 + MAC3 = MAC3 + NO2
k = k(70)
XMEO + NO = NO + MEO2
k = k(52)
XMEO + HO2 = HO2 + XC
k = k(53)
XMEO + NO3 = NO3 + MEO2
k = k(54)
XMEO + MEO2 = 1.5 MEO2 + 0.5 XC
k = k(55)
XMEO + RO2C = RO2C + 0.5 MEO2 + 0.5 XC
k = k(56)
XMEO + RO2X = RO2X + 0.5 MEO2 + 0.5 XC
k = k(56)
XMEO + MCO3 = MCO3 + MEO2
k = k(70)
XMEO + RCO3 = RCO3 + MEO2
k = k(70)
XMEO + BZC3 = BZC3 + MEO2
k = k(70)
XMEO + MAC3 = MAC3 + MEO2
k = k(70)
XMC3 + NO = NO + MCO3
k = k(52)
XMC3 + HO2 = HO2 + 2. XC
k = k(53)
XMC3 + NO3 = NO3 + MCO3
k = k(54)
XMC3 + MEO2 = MEO2 + 0.5 MCO3 + XC
k = k(55)
XMC3 + RO2C = RO2C + 0.5 MCO3 + XC
k = k(56)
XMC3 + RO2X = RO2X + 0.5 MCO3 + XC
k = k(56)
XMC3 + MCO3 = 2. MCO3
k = k(70)
XMC3 + RCO3 = RCO3 + MCO3
k = k(70)
XMC3 + BZC3 = BZC3 + MCO3
k = k(70)
XMC3 + MAC3 = MAC3 + MCO3
k = k(70)
XRC3 + NO = NO + RCO3
k = k(52)
XRC3 + HO2 = HO2 + 3. XC
k = k(53)
XRC3 + NO3 = NO3 + RCO3
k = k(54)
XRC3 + MEO2 = MEO2 + 0.5 RCO3 + 1.5 XC
k = k(55)
XRC3 + RO2C = RO2C + 0.5 RCO3 + 1.5 XC
k = k(56)
XRC3 + RO2X = RO2X + 0.5 RCO3 + 1.5 XC
k = k(56)
XRC3 + MCO3 = MCO3 + RCO3
k = k(70)
XRC3 + RCO3 = 2. RCO3
k = k(70)
XRC3 + BZC3 = BZC3 + RCO3
k = k(70)
XRC3 + MAC3 = MAC3 + RCO3
k = k(70)
XMA3 + NO = NO + MAC3
k = k(52)
XMA3 + HO2 = HO2 + 4. XC
k = k(53)
XMA3 + NO3 = NO3 + MAC3
k = k(54)
XMA3 + MEO2 = MEO2 + 0.5 MAC3 + 2. XC
k = k(55)
XMA3 + RO2C = RO2C + 0.5 MAC3 + 2. XC
k = k(56)
XMA3 + RO2X = RO2X + 0.5 MAC3 + 2. XC
k = k(56)
XMA3 + MCO3 = MCO3 + MAC3
k = k(70)
XMA3 + RCO3 = RCO3 + MAC3
k = k(70)
XMA3 + BZC3 = BZC3 + MAC3
k = k(70)
XMA3 + MAC3 = 2. MAC3
k = k(70)
XTBU + NO = NO + TBUO
k = k(52)
XTBU + HO2 = HO2 + 4. XC
k = k(53)
XTBU + NO3 = NO3 + TBUO
k = k(54)
XTBU + MEO2 = MEO2 + 0.5 TBUO + 2. XC
k = k(55)
XTBU + RO2C = RO2C + 0.5 TBUO + 2. XC
k = k(56)
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS
255

k300
3.50E‐14
3.50E‐14
1.56E‐11
1.56E‐11
1.56E‐11
1.56E‐11
9.23E‐12
7.63E‐12
2.30E‐12
2.00E‐13
3.50E‐14
3.50E‐14
1.56E‐11
1.56E‐11
1.56E‐11
1.56E‐11
9.23E‐12
7.63E‐12
2.30E‐12
2.00E‐13
3.50E‐14
3.50E‐14
1.56E‐11
1.56E‐11
1.56E‐11
1.56E‐11
9.23E‐12
7.63E‐12
2.30E‐12
2.00E‐13
3.50E‐14
3.50E‐14
1.56E‐11
1.56E‐11
1.56E‐11
1.56E‐11
9.23E‐12
7.63E‐12
2.30E‐12
2.00E‐13
3.50E‐14
3.50E‐14
1.56E‐11
1.56E‐11
1.56E‐11
1.56E‐11
9.23E‐12
7.63E‐12
2.30E‐12
2.00E‐13
3.50E‐14
www.camx.com

March 2016

Number
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211

212
213
214
215
216

217
218
219
220

221

222
223
224

225

CAMx User’s Guide Version 6.3
Appendix D: Mechanism 5 – SAPRC07TC

Reactants and Products
XTBU + RO2X = RO2X + 0.5 TBUO + 2. XC
XTBU + MCO3 = MCO3 + TBUO
XTBU + RCO3 = RCO3 + TBUO
XTBU + BZC3 = BZC3 + TBUO
XTBU + MAC3 = MAC3 + TBUO
XCO + NO = NO + CO
XCO + HO2 = HO2 + XC
XCO + NO3 = NO3 + CO
XCO + MEO2 = MEO2 + 0.5 CO + 0.5 XC
XCO + RO2C = RO2C + 0.5 CO + 0.5 XC
XCO + RO2X = RO2X + 0.5 CO + 0.5 XC
XCO + MCO3 = MCO3 + CO
XCO + RCO3 = RCO3 + CO
XCO + BZC3 = BZC3 + CO
XCO + MAC3 = MAC3 + CO
HCHO = 2. HO2 + CO
HCHO = CO
HCHO + OH = HO2 + CO
HCHO + NO3 = HNO3 + HO2 + CO
CCHO + OH = MCO3
CCHO = CO + HO2 + MEO2
CCHO + NO3 = HNO3 + MCO3
RCHO + OH = 0.965 RCO3 + 0.035 RO2C +
0.035 XHO2 + 0.035 XCO + 0.035 XCCH +
0.035 YRPX
RCHO = RO2C + XHO2 + YRPX + XCCH + CO +
HO2
RCHO + NO3 = HNO3 + RCO3
ACET + OH = RO2C + XMC3 + XHCH + YRPX
ACET = 0.62 MCO3 + 1.38 MEO2 + 0.38 CO
MEK + OH = 0.967 RO2C + 0.039 RO2X + 0.039
ZRN3 + 0.376 XHO2 + 0.51 XMC3 + 0.074
XRC3 + 0.088 XHCH + 0.504 XCCH + 0.376
XRCH + YRPX + 0.3 XC
MEK = MCO3 + RO2C + XHO2 + XCCH + YRPX
MEOH + OH = HCHO + HO2
FACD + OH = HO2 + CO2
AACD + OH = 0.509 MEO2 + 0.491 RO2C +
0.509 CO2 + 0.491 XHO2 + 0.491 XMGL +
0.491 YRPX ‐ 0.491 XC
PACD + OH = RO2C + XHO2 + 0.143 CO2 +
0.142 XCCH + 0.4 XRCH + 0.457 XBAC + YRPX ‐
0.455 XC
COOH + OH = 0.3 HCHO + 0.3 OH + 0.7 MEO2
COOH = HCHO + HO2 + OH
ROOH + OH = 0.744 OH + 0.251 RO2C + 0.004
RO2X + 0.004 ZRN3 + 0.744 RCHO + 0.239
XHO2 + 0.012 XOH + 0.012 XHCH + 0.012
XCCH + 0.205 XRCH + 0.034 XPD2 + 0.256
YRPX ‐ 0.115 XC
ROOH = RCHO + HO2 + OH

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

Rate Constant Expression
k = k(56)
k = k(70)
k = k(70)
k = k(70)
k = k(70)
k = k(52)
k = k(53)
k = k(54)
k = k(55)
k = k(56)
k = k(56)
k = k(70)
k = k(70)
k = k(70)
k = k(70)
Photolysis
Photolysis
k = 5.40E‐12 exp(135/T)
k = 2.00E‐12 exp(‐2431/T)
k = 4.40E‐12 exp(365/T)
Photolysis
k = 1.40E‐12 exp(‐1860/T)
k = 5.10E‐12 exp(405/T)

k300
3.50E‐14
1.56E‐11
1.56E‐11
1.56E‐11
1.56E‐11
9.23E‐12
7.63E‐12
2.30E‐12
2.00E‐13
3.50E‐14
3.50E‐14
1.56E‐11
1.56E‐11
1.56E‐11
1.56E‐11
1.78E‐5
2.38E‐5
8.47E‐12
6.05E‐16
1.49E‐11
1.77E‐6
2.84E‐15
1.97E‐11

Photolysis

6.95E‐6

k = 1.40E‐12 exp(‐1601/T)
k = 4.56E‐14 (T/300)^3.65 exp(429/T)
Photolysis
k = 1.30E‐12 (T/300)^2 exp(‐25/T)

6.74E‐15
1.91E‐13
1.04E‐7
1.20E‐12

Photolysis
k = 2.85E‐12 exp(‐345/T)
k = 4.50E‐13
k = 4.20E‐14 exp(855/T)

8.13E‐7
9.02E‐13
4.50E‐13
7.26E‐13

k = 1.20E‐12

1.20E‐12

k = 3.80E‐12 exp(200/T)
Photolysis
k = 2.50E‐11

7.40E‐12
2.72E‐6
2.50E‐11

Photolysis

2.72E‐6

256

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
Appendix D: Mechanism 5 – SAPRC07TC

Number Reactants and Products
Rate Constant Expression
226
R6PX + OH = 0.84 OH + 0.222 RO2C + 0.029
k = 5.60E‐11
RO2X + 0.029 ZRN3 + 0.84 PRD2 + 0.09 XHO2
+ 0.041 XOH + 0.02 XCCH + 0.075 XRCH +
0.084 XPD2 + 0.16 YRPX + 0.02 XC
227
R6PX = OH + 0.142 HO2 + 0.782 RO2C + 0.077 Photolysis
RO2X + 0.077 ZRN3 + 0.085 RCHO + 0.142
PRD2 + 0.782 XHO2 + 0.026 XCCH + 0.058
XRCH + 0.698 XPD2 + 0.858 Y6PX + 0.017 XC
228
RAPX + OH = 0.139 OH + 0.148 HO2 + 0.589
k = 1.41E‐10
RO2C + 0.124 RO2X + 0.124 ZRN3 + 0.074
PRD2 + 0.147 MGLY + 0.139 IPRD + 0.565
XHO2 + 0.024 XOH + 0.448 XRCH + 0.026 XGLY
+ 0.03 XMEK + 0.252 XMGL + 0.073 XAF1 +
0.073 XAF2 + 0.713 Y6PX + 2.674 XC
229
RAPX = OH + HO2 + 0.5 GLY + 0.5 MGLY + 0.5 Photolysis
AFG1 + 0.5 AFG2 + 0.5 XC
230
GLY = 2. CO + 2. HO2
Photolysis
231
GLY = HCHO + CO
Photolysis
232
GLY + OH = 0.63 HO2 + 1.26 CO + 0.37 RCO3 ‐ k = 1.10E‐11
0.37 XC
233
GLY + NO3 = HNO3 + 0.63 HO2 + 1.26 CO +
k = 2.80E‐12 exp(‐2376/T)
0.37 RCO3 ‐ 0.37 XC
234
MGLY = HO2 + CO + MCO3
Photolysis
235
MGLY + OH = CO + MCO3
k = 1.50E‐11
236
MGLY + NO3 = HNO3 + CO + MCO3
k = 1.40E‐12 exp(‐1895/T)
237
BACL = 2. MCO3
Photolysis
238
CRES + OH = 0.2 BZO + 0.8 RO2C + 0.8 XHO2 + k = 1.70E‐12 exp(950/T)
0.8 Y6PX + 0.25 XMGL + 5.05 XC
239
CRES + NO3 = HNO3 + BZO + XC
k = 1.40E‐11
240
NPHE + OH = BZO + XN
k = 3.50E‐12
241
NPHE = HONO + 6. XC
Photolysis
242
NPHE = 6. XC + XN
Photolysis
243
BALD + OH = BZC3
k = 1.20E‐11
244
BALD = 7. XC
Photolysis
245
BALD + NO3 = HNO3 + BZC3
k = 1.34E‐12 exp(‐1860/T)
246
AFG1 + OH = 0.217 MAC3 + 0.723 RO2C + 0.06 k = 7.40E‐11
RO2X + 0.06 ZRN3 + 0.521 XHO2 + 0.201
XMC3 + 0.334 XCO + 0.407 XRCH + 0.129
XMEK + 0.107 XGLY + 0.267 XMGL + 0.783
Y6PX + 0.284 XC
247
AFG1 + O3 = 0.826 OH + 0.522 HO2 + 0.652
k = 9.66E‐18
RO2C + 0.522 CO + 0.174 CO2 + 0.432 GLY +
0.568 MGLY + 0.652 XRC3 + 0.652 XHCH +
0.652 Y6PX ‐ 0.872 XC
248
AFG1 = 1.023 HO2 + 0.173 MEO2 + 0.305
Photolysis
MCO3 + 0.5 MAC3 + 0.695 CO + 0.195 GLY +
0.305 MGLY + 0.217 XC
249
AFG2 + OH = 0.217 MAC3 + 0.723 RO2C + 0.06 k = 7.40E‐11
RO2X + 0.06 ZRN3 + 0.521 XHO2 + 0.201
XMC3 + 0.334 XCO + 0.407 XRCH + 0.129
XMEK + 0.107 XGLY + 0.267 XMGL + 0.783
Y6PX + 0.284 XC
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS
257

k300
5.60E‐11

2.72E‐6

1.41E‐10

2.72E‐6
7.88E‐5
2.23E‐5
1.10E‐11
1.02E‐15
1.39E‐4
1.50E‐11
2.53E‐15
2.45E‐4
4.03E‐11
1.40E‐11
3.50E‐12
9.55E‐6
9.55E‐5
1.20E‐11
2.48E‐5
2.72E‐15
7.40E‐11

9.66E‐18

3.07E‐3

7.40E‐11

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
Appendix D: Mechanism 5 – SAPRC07TC

Number Reactants and Products
250
AFG2 + O3 = 0.826 OH + 0.522 HO2 + 0.652
RO2C + 0.522 CO + 0.174 CO2 + 0.432 GLY +
0.568 MGLY + 0.652 XRC3 + 0.652 XHCH +
0.652 Y6PX ‐ 0.872 XC
251
AFG2 = PRD2 ‐ 1. XC
252
AFG3 + OH = 0.206 MAC3 + 0.733 RO2C +
0.117 RO2X + 0.117 ZRN3 + 0.561 XHO2 +
0.117 XMC3 + 0.114 XCO + 0.274 XGLY + 0.153
XMGL + 0.019 XBAC + 0.195 XAF1 + 0.195
XAF2 + 0.231 XIPR + 0.794 Y6PX + 0.938 XC
253
AFG3 + O3 = 0.471 OH + 0.554 HO2 + 0.013
MCO3 + 0.258 RO2C + 0.007 RO2X + 0.007
ZRN3 + 0.58 CO + 0.19 CO2 + 0.366 GLY +
0.184 MGLY + 0.35 AFG1 + 0.35 AFG2 + 0.139
AFG3 + 0.003 MACR + 0.004 MVK + 0.003
IPRD + 0.095 XHO2 + 0.163 XRC3 + 0.163
XHCH + 0.095 XMGL + 0.264 Y6PX ‐ 0.575 XC
254
MACR + OH = 0.5 MAC3 + 0.5 RO2C + 0.5
XHO2 + 0.416 XCO + 0.084 XHCH + 0.416
XMEK + 0.084 XMGL + 0.5 YRPX ‐ 0.416 XC
255
MACR + O3 = 0.208 OH + 0.108 HO2 + 0.1
RO2C + 0.45 CO + 0.117 CO2 + 0.1 HCHO + 0.9
MGLY + 0.333 FACD + 0.1 XRC3 + 0.1 XHCH +
0.1 YRPX ‐ 0.1 XC
256
MACR + NO3 = 0.5 MAC3 + 0.5 RO2C + 0.5
HNO3 + 0.5 XHO2 + 0.5 XCO + 0.5 YRPX + 1.5
XC + 0.5 XN
257
MACR + O3P = RCHO + XC
258
MACR = 0.33 OH + 0.67 HO2 + 0.34 MCO3 +
0.33 MAC3 + 0.33 RO2C + 0.67 CO + 0.34
HCHO + 0.33 XMC3 + 0.33 XHCH + 0.33 YRPX
259
MVK + OH = 0.975 RO2C + 0.025 RO2X + 0.025
ZRN3 + 0.3 XHO2 + 0.675 XMC3 + 0.3 XHCH +
0.675 XGLD + 0.3 XMGL + YRPX ‐ 0.05 XC
260
MVK + O3 = 0.164 OH + 0.064 HO2 + 0.05
RO2C + 0.05 XHO2 + 0.475 CO + 0.124 CO2 +
0.05 HCHO + 0.95 MGLY + 0.351 FACD + 0.05
XRC3 + 0.05 XHCH + 0.05 YRPX ‐ 0.05 XC
261
MVK + O3P = 0.45 RCHO + 0.55 MEK + 0.45 XC
262
MVK = 0.4 MEO2 + 0.6 CO + 0.6 PRD2 + 0.4
MAC3 ‐ 2.2 XC
263
IPRD + OH = 0.289 MAC3 + 0.67 RO2C + 0.67
XHO2 + 0.041 RO2X + 0.041 ZRN3 + 0.336 XCO
+ 0.055 XHCH + 0.129 XGLD + 0.013 XRCH +
0.15 XMEK + 0.332 XPD2 + 0.15 XGLY + 0.174
XMGL ‐ 0.504 XC + 0.711 Y6PX
264
IPRD + O3 = 0.285 OH + 0.4 HO2 + 0.048 RO2C
+ 0.048 XRC3 + 0.498 CO + 0.14 CO2 + 0.124
HCHO + 0.21 MEK + 0.023 GLY + 0.742 MGLY +
0.1 FACD + 0.372 PACD + 0.047 XGLD + 0.001
XHCH + 0.048 Y6PX ‐ 0.329 XC
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

Rate Constant Expression
k = 9.66E‐18

k300
9.66E‐18

Photolysis
k = 9.35E‐11

3.07E‐3
9.35E‐11

k = 1.43E‐17

1.43E‐17

k = 8.00E‐12 exp(380/T)

2.84E‐11

k = 1.40E‐15 exp(‐2100/T)

1.28E‐18

k = 1.50E‐12 exp(‐1815/T)

3.54E‐15

k = 6.34E‐12
Photolysis

6.34E‐12
1.39E‐6

k = 2.60E‐12 exp(610/T)

1.99E‐11

k = 8.50E‐16 exp(‐1520/T)

5.36E‐18

k = 4.32E‐12
Photolysis

4.32E‐12
5.25E‐7

k = 6.19E‐11

6.19E‐11

k = 4.18E‐18

4.18E‐18

258

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
Appendix D: Mechanism 5 – SAPRC07TC

Number Reactants and Products
265
IPRD + NO3 = 0.15 MAC3 + 0.15 HNO3 + 0.799
RO2C + 0.799 XHO2 + 0.051 RO2X + 0.051
ZRN3 + 0.572 XCO + 0.227 XHCH + 0.218 XRCH
+ 0.008 XMGL + 0.572 XRN3 + 0.85 Y6PX +
0.278 XN ‐ 0.815 XC
266
IPRD = 1.233 HO2 + 0.467 MCO3 + 0.3 RCO3 +
1.233 CO + 0.3 HCHO + 0.467 GLYD + 0.233
MEK ‐ 0.233 XC
267
PRD2 + OH = 0.472 HO2 + 0.379 XHO2 + 0.029
XMC3 + 0.049 XRC3 + 0.473 RO2C + 0.071
RO2X + 0.071 ZRN3 + 0.002 HCHO + 0.211
XHCH + 0.001 CCHO + 0.083 XCCH + 0.143
RCHO + 0.402 XRCH + 0.115 XMEK + 0.329
PRD2 + 0.007 XPD2 + 0.528 Y6PX + 0.877 XC
268
PRD2 = 0.913 XHO2 + 0.4 MCO3 + 0.6 RCO3 +
1.59 RO2C + 0.087 RO2X + 0.087 ZRN3 + 0.303
XHCH + 0.163 XCCH + 0.78 XRCH + Y6PX ‐
0.091 XC
269
RNO3 + OH = 0.189 HO2 + 0.305 XHO2 + 0.019
NO2 + 0.313 XNO2 + 0.976 RO2C + 0.175
RO2X + 0.175 ZRN3 + 0.011 XHCH + 0.429
XCCH + 0.001 RCHO + 0.036 XRCH + 0.004
XACE + 0.01 MEK + 0.17 XMEK + 0.008 PRD2 +
0.031 XPD2 + 0.189 RNO3 + 0.305 XRN3 +
0.157 YRPX + 0.636 Y6PX + 0.174 XN + 0.04 XC
270
RNO3 = 0.344 HO2 + 0.554 XHO2 + NO2 +
0.721 RO2C + 0.102 RO2X + 0.102 ZRN3 +
0.074 HCHO + 0.061 XHCH + 0.214 CCHO +
0.23 XCCH + 0.074 RCHO + 0.063 XRCH +
0.008 XACE + 0.124 MEK + 0.083 XMEK + 0.19
PRD2 + 0.261 XPD2 + 0.066 YRPX + 0.591 Y6PX
+ 0.396 XC
271
GLYD + OH = MCO3
272
GLYD = CO + 2. HO2 + HCHO
273
GLYD + NO3 = HNO3 + MCO3
274
ACRO + OH = 0.25 XHO2 + 0.75 MAC3 + 0.25
RO2C + 0.167 XCO + 0.083 XHCH + 0.167 XCCH
+ 0.083 XGLY + 0.25 YRPX ‐ 0.75 XC
275
ACRO + O3 = 0.83 HO2 + 0.33 OH + 1.005 CO +
0.31 CO2 + 0.5 HCHO + 0.185 FACD + 0.5 GLY
276
ACRO + NO3 = 0.031 XHO2 + 0.967 MAC3 +
0.031 RO2C + 0.002 RO2X + 0.002 ZRN3 +
0.967 HNO3 + 0.031 XCO + 0.031 XRN3 +
0.033 YRPX + 0.002 XN ‐ 1.097 XC
277
ACRO + O3P = RCHO
278
ACRO = 1.066 HO2 + 0.178 OH + 0.234 MEO2
+ 0.33 MAC3 + 1.188 CO + 0.102 CO2 + 0.34
HCHO + 0.05 AACD ‐ 0.284 XC
279
CO3H + OH = 0.98 MCO3 + 0.02 RO2C + 0.02
CO2 + 0.02 XOH + 0.02 XHCH + 0.02 YRPX
280
CO3H = MEO2 + CO2 + OH
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

Rate Constant Expression
k = 1.00E‐13

k300
1.00E‐13

Photolysis

1.39E‐6

k = 1.55E‐11

1.55E‐11

Photolysis

2.26E‐8

k = 7.20E‐12

7.20E‐12

Photolysis

1.20E‐6

k = k(208)
Photolysis
k = k(210)
k = 1.99E‐11

1.49E‐11
2.75E‐6
2.84E‐15
1.99E‐11

k = 1.40E‐15 exp(‐2528/T)

3.07E‐19

k = 1.18E‐15

1.18E‐15

k = 2.37E‐12
Photolysis

2.37E‐12
1.28E‐6

k = 5.28E‐12

5.28E‐12

Photolysis

3.60E‐7

259

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
Appendix D: Mechanism 5 – SAPRC07TC

Number Reactants and Products
Rate Constant Expression
281
RO3H + OH = 0.806 RCO3 + 0.194 RO2C +
k = 6.42E‐12
0.194 YRPX + 0.11 CO2 + 0.11 XOH + 0.11
XCCH + 0.084 XHO2 + 0.084 XRCH
282
RO3H = XHO2 + XCCH + YRPX + CO2 + OH
Photolysis
283
XHCH + NO = NO + HCHO
k = k(52)
284
XHCH + HO2 = HO2 + XC
k = k(53)
285
XHCH + NO3 = NO3 + HCHO
k = k(54)
286
XHCH + MEO2 = MEO2 + 0.5 HCHO + 0.5 XC
k = k(55)
287
XHCH + RO2C = RO2C + 0.5 HCHO + 0.5 XC
k = k(56)
288
XHCH + RO2X = RO2X + 0.5 HCHO + 0.5 XC
k = k(56)
289
XHCH + MCO3 = MCO3 + HCHO
k = k(70)
290
XHCH + RCO3 = RCO3 + HCHO
k = k(70)
291
XHCH + BZC3 = BZC3 + HCHO
k = k(70)
292
XHCH + MAC3 = MAC3 + HCHO
k = k(70)
293
XCCH + NO = NO + CCHO
k = k(52)
294
XCCH + HO2 = HO2 + 2. XC
k = k(53)
295
XCCH + NO3 = NO3 + CCHO
k = k(54)
296
XCCH + MEO2 = MEO2 + 0.5 CCHO + XC
k = k(55)
297
XCCH + RO2C = RO2C + 0.5 CCHO + XC
k = k(56)
298
XCCH + RO2X = RO2X + 0.5 CCHO + XC
k = k(56)
299
XCCH + MCO3 = MCO3 + CCHO
k = k(70)
300
XCCH + RCO3 = RCO3 + CCHO
k = k(70)
301
XCCH + BZC3 = BZC3 + CCHO
k = k(70)
302
XCCH + MAC3 = MAC3 + CCHO
k = k(70)
303
XRCH + NO = NO + RCHO
k = k(52)
304
XRCH + HO2 = HO2 + 3. XC
k = k(53)
305
XRCH + NO3 = NO3 + RCHO
k = k(54)
306
XRCH + MEO2 = MEO2 + 0.5 RCHO + 1.5 XC
k = k(55)
307
XRCH + RO2C = RO2C + 0.5 RCHO + 1.5 XC
k = k(56)
308
XRCH + RO2X = RO2X + 0.5 RCHO + 1.5 XC
k = k(56)
309
XRCH + MCO3 = MCO3 + RCHO
k = k(70)
310
XRCH + RCO3 = RCO3 + RCHO
k = k(70)
311
XRCH + BZC3 = BZC3 + RCHO
k = k(70)
312
XRCH + MAC3 = MAC3 + RCHO
k = k(70)
313
XACE + NO = NO + ACET
k = k(52)
314
XACE + HO2 = HO2 + 3. XC
k = k(53)
315
XACE + NO3 = NO3 + ACET
k = k(54)
316
XACE + MEO2 = MEO2 + 0.5 ACET + 1.5 XC
k = k(55)
317
XACE + RO2C = RO2C + 0.5 ACET + 1.5 XC
k = k(56)
318
XACE + RO2X = RO2X + 0.5 ACET + 1.5 XC
k = k(56)
319
XACE + MCO3 = MCO3 + ACET
k = k(70)
320
XACE + RCO3 = RCO3 + ACET
k = k(70)
321
XACE + BZC3 = BZC3 + ACET
k = k(70)
322
XACE + MAC3 = MAC3 + ACET
k = k(70)
323
XMEK + NO = NO + MEK
k = k(52)
324
XMEK + HO2 = HO2 + 4. XC
k = k(53)
325
XMEK + NO3 = NO3 + MEK
k = k(54)
326
XMEK + MEO2 = MEO2 + 0.5 MEK + 2. XC
k = k(55)
327
XMEK + RO2C = RO2C + 0.5 MEK + 2. XC
k = k(56)
328
XMEK + RO2X = RO2X + 0.5 MEK + 2. XC
k = k(56)
329
XMEK + MCO3 = MCO3 + MEK
k = k(70)
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS
260

k300
6.42E‐12

3.60E‐7
9.23E‐12
7.63E‐12
2.30E‐12
2.00E‐13
3.50E‐14
3.50E‐14
1.56E‐11
1.56E‐11
1.56E‐11
1.56E‐11
9.23E‐12
7.63E‐12
2.30E‐12
2.00E‐13
3.50E‐14
3.50E‐14
1.56E‐11
1.56E‐11
1.56E‐11
1.56E‐11
9.23E‐12
7.63E‐12
2.30E‐12
2.00E‐13
3.50E‐14
3.50E‐14
1.56E‐11
1.56E‐11
1.56E‐11
1.56E‐11
9.23E‐12
7.63E‐12
2.30E‐12
2.00E‐13
3.50E‐14
3.50E‐14
1.56E‐11
1.56E‐11
1.56E‐11
1.56E‐11
9.23E‐12
7.63E‐12
2.30E‐12
2.00E‐13
3.50E‐14
3.50E‐14
1.56E‐11
www.camx.com

March 2016

Number
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380

CAMx User’s Guide Version 6.3
Appendix D: Mechanism 5 – SAPRC07TC

Reactants and Products
Rate Constant Expression
XMEK + RCO3 = RCO3 + MEK
k = k(70)
XMEK + BZC3 = BZC3 + MEK
k = k(70)
XMEK + MAC3 = MAC3 + MEK
k = k(70)
XPD2 + NO = NO + PRD2
k = k(52)
XPD2 + HO2 = HO2 + 6. XC
k = k(53)
XPD2 + NO3 = NO3 + PRD2
k = k(54)
XPD2 + MEO2 = MEO2 + 0.5 PRD2 + 3. XC
k = k(55)
XPD2 + RO2C = RO2C + 0.5 PRD2 + 3. XC
k = k(56)
XPD2 + RO2X = RO2X + 0.5 PRD2 + 3. XC
k = k(56)
XPD2 + MCO3 = MCO3 + PRD2
k = k(70)
XPD2 + RCO3 = RCO3 + PRD2
k = k(70)
XPD2 + BZC3 = BZC3 + PRD2
k = k(70)
XPD2 + MAC3 = MAC3 + PRD2
k = k(70)
XGLY + NO = NO + GLY
k = k(52)
XGLY + HO2 = HO2 + 2. XC
k = k(53)
XGLY + NO3 = NO3 + GLY
k = k(54)
XGLY + MEO2 = MEO2 + 0.5 GLY + XC
k = k(55)
XGLY + RO2C = RO2C + 0.5 GLY + XC
k = k(56)
XGLY + RO2X = RO2X + 0.5 GLY + XC
k = k(56)
XGLY + MCO3 = MCO3 + GLY
k = k(70)
XGLY + RCO3 = RCO3 + GLY
k = k(70)
XGLY + BZC3 = BZC3 + GLY
k = k(70)
XGLY + MAC3 = MAC3 + GLY
k = k(70)
XMGL + NO = NO + MGLY
k = k(52)
XMGL + HO2 = HO2 + 3. XC
k = k(53)
XMGL + NO3 = NO3 + MGLY
k = k(54)
XMGL + MEO2 = MEO2 + 0.5 MGLY + 1.5 XC
k = k(55)
XMGL + RO2C = RO2C + 0.5 MGLY + 1.5 XC
k = k(56)
XMGL + RO2X = RO2X + 0.5 MGLY + 1.5 XC
k = k(56)
XMGL + MCO3 = MCO3 + MGLY
k = k(70)
XMGL + RCO3 = RCO3 + MGLY
k = k(70)
XMGL + BZC3 = BZC3 + MGLY
k = k(70)
XMGL + MAC3 = MAC3 + MGLY
k = k(70)
XBAC + NO = NO + BACL
k = k(52)
XBAC + HO2 = HO2 + 4. XC
k = k(53)
XBAC + NO3 = NO3 + BACL
k = k(54)
XBAC + MEO2 = MEO2 + 0.5 BACL + 2. XC
k = k(55)
XBAC + RO2C = RO2C + 0.5 BACL + 2. XC
k = k(56)
XBAC + RO2X = RO2X + 0.5 BACL + 2. XC
k = k(56)
XBAC + MCO3 = MCO3 + BACL
k = k(70)
XBAC + RCO3 = RCO3 + BACL
k = k(70)
XBAC + BZC3 = BZC3 + BACL
k = k(70)
XBAC + MAC3 = MAC3 + BACL
k = k(70)
XBAL + NO = NO + BALD
k = k(52)
XBAL + HO2 = HO2 + 7. XC
k = k(53)
XBAL + NO3 = NO3 + BALD
k = k(54)
XBAL + MEO2 = MEO2 + 0.5 BALD + 3.5 XC
k = k(55)
XBAL + RO2C = RO2C + 0.5 BALD + 3.5 XC
k = k(56)
XBAL + RO2X = RO2X + 0.5 BALD + 3.5 XC
k = k(56)
XBAL + MCO3 = MCO3 + BALD
k = k(70)
XBAL + RCO3 = RCO3 + BALD
k = k(70)
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS
261

k300
1.56E‐11
1.56E‐11
1.56E‐11
9.23E‐12
7.63E‐12
2.30E‐12
2.00E‐13
3.50E‐14
3.50E‐14
1.56E‐11
1.56E‐11
1.56E‐11
1.56E‐11
9.23E‐12
7.63E‐12
2.30E‐12
2.00E‐13
3.50E‐14
3.50E‐14
1.56E‐11
1.56E‐11
1.56E‐11
1.56E‐11
9.23E‐12
7.63E‐12
2.30E‐12
2.00E‐13
3.50E‐14
3.50E‐14
1.56E‐11
1.56E‐11
1.56E‐11
1.56E‐11
9.23E‐12
7.63E‐12
2.30E‐12
2.00E‐13
3.50E‐14
3.50E‐14
1.56E‐11
1.56E‐11
1.56E‐11
1.56E‐11
9.23E‐12
7.63E‐12
2.30E‐12
2.00E‐13
3.50E‐14
3.50E‐14
1.56E‐11
1.56E‐11
www.camx.com

March 2016

Number
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431

CAMx User’s Guide Version 6.3
Appendix D: Mechanism 5 – SAPRC07TC

Reactants and Products
Rate Constant Expression
XBAL + BZC3 = BZC3 + BALD
k = k(70)
XBAL + MAC3 = MAC3 + BALD
k = k(70)
XAF1 + NO = NO + AFG1
k = k(52)
XAF1 + HO2 = HO2 + 5. XC
k = k(53)
XAF1 + NO3 = NO3 + AFG1
k = k(54)
XAF1 + MEO2 = MEO2 + 0.5 AFG1 + 2.5 XC
k = k(55)
XAF1 + RO2C = RO2C + 0.5 AFG1 + 2.5 XC
k = k(56)
XAF1 + RO2X = RO2X + 0.5 AFG1 + 2.5 XC
k = k(56)
XAF1 + MCO3 = MCO3 + AFG1
k = k(70)
XAF1 + RCO3 = RCO3 + AFG1
k = k(70)
XAF1 + BZC3 = BZC3 + AFG1
k = k(70)
XAF1 + MAC3 = MAC3 + AFG1
k = k(70)
XAF2 + NO = NO + AFG2
k = k(52)
XAF2 + HO2 = HO2 + 5. XC
k = k(53)
XAF2 + NO3 = NO3 + AFG2
k = k(54)
XAF2 + MEO2 = MEO2 + 0.5 AFG2 + 2.5 XC
k = k(55)
XAF2 + RO2C = RO2C + 0.5 AFG2 + 2.5 XC
k = k(56)
XAF2 + RO2X = RO2X + 0.5 AFG2 + 2.5 XC
k = k(56)
XAF2 + MCO3 = MCO3 + AFG2
k = k(70)
XAF2 + RCO3 = RCO3 + AFG2
k = k(70)
XAF2 + BZC3 = BZC3 + AFG2
k = k(70)
XAF2 + MAC3 = MAC3 + AFG2
k = k(70)
XAF3 + NO = NO + AFG3
k = k(52)
XAF3 + HO2 = HO2 + 7. XC
k = k(53)
XAF3 + NO3 = NO3 + AFG3
k = k(54)
XAF3 + MEO2 = MEO2 + 0.5 AFG3 + 3.5 XC
k = k(55)
XAF3 + RO2C = RO2C + 0.5 AFG3 + 3.5 XC
k = k(56)
XAF3 + RO2X = RO2X + 0.5 AFG3 + 3.5 XC
k = k(56)
XAF3 + MCO3 = MCO3 + AFG3
k = k(70)
XAF3 + RCO3 = RCO3 + AFG3
k = k(70)
XAF3 + BZC3 = BZC3 + AFG3
k = k(70)
XAF3 + MAC3 = MAC3 + AFG3
k = k(70)
XMAC + NO = NO + MACR
k = k(52)
XMAC + HO2 = HO2 + 4. XC
k = k(53)
XMAC + NO3 = NO3 + MACR
k = k(54)
XMAC + MEO2 = MEO2 + 0.5 MACR + 2. XC
k = k(55)
XMAC + RO2C = RO2C + 0.5 MACR + 2. XC
k = k(56)
XMAC + RO2X = RO2X + 0.5 MACR + 2. XC
k = k(56)
XMAC + MCO3 = MCO3 + MACR
k = k(70)
XMAC + RCO3 = RCO3 + MACR
k = k(70)
XMAC + BZC3 = BZC3 + MACR
k = k(70)
XMAC + MAC3 = MAC3 + MACR
k = k(70)
XMVK + NO = NO + MVK
k = k(52)
XMVK + HO2 = HO2 + 4. XC
k = k(53)
XMVK + NO3 = NO3 + MVK
k = k(54)
XMVK + MEO2 = MEO2 + 0.5 MVK + 2. XC
k = k(55)
XMVK + RO2C = RO2C + 0.5 MVK + 2. XC
k = k(56)
XMVK + RO2X = RO2X + 0.5 MVK + 2. XC
k = k(56)
XMVK + MCO3 = MCO3 + MVK
k = k(70)
XMVK + RCO3 = RCO3 + MVK
k = k(70)
XMVK + BZC3 = BZC3 + MVK
k = k(70)
COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS
262

k300
1.56E‐11
1.56E‐11
9.23E‐12
7.63E‐12
2.30E‐12
2.00E‐13
3.50E‐14
3.50E‐14
1.56E‐11
1.56E‐11
1.56E‐11
1.56E‐11
9.23E‐12
7.63E‐12
2.30E‐12
2.00E‐13
3.50E‐14
3.50E‐14
1.56E‐11
1.56E‐11
1.56E‐11
1.56E‐11
9.23E‐12
7.63E‐12
2.30E‐12
2.00E‐13
3.50E‐14
3.50E‐14
1.56E‐11
1.56E‐11
1.56E‐11
1.56E‐11
9.23E‐12
7.63E‐12
2.30E‐12
2.00E‐13
3.50E‐14
3.50E‐14
1.56E‐11
1.56E‐11
1.56E‐11
1.56E‐11
9.23E‐12
7.63E‐12
2.30E‐12
2.00E‐13
3.50E‐14
3.50E‐14
1.56E‐11
1.56E‐11
1.56E‐11
www.camx.com

March 2016

Number
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479

CAMx User’s Guide Version 6.3
Appendix D: Mechanism 5 – SAPRC07TC

Reactants and Products
XMVK + MAC3 = MAC3 + MVK
XIPR + NO = NO + IPRD
XIPR + HO2 = HO2 + 5. XC
XIPR + NO3 = NO3 + IPRD
XIPR + MEO2 = MEO2 + 0.5 IPRD + 2.5 XC
XIPR + RO2C = RO2C + 0.5 IPRD + 2.5 XC
XIPR + RO2X = RO2X + 0.5 IPRD + 2.5 XC
XIPR + MCO3 = MCO3 + IPRD
XIPR + RCO3 = RCO3 + IPRD
XIPR + BZC3 = BZC3 + IPRD
XIPR + MAC3 = MAC3 + IPRD
XRN3 + NO = NO + RNO3
XRN3 + HO2 = HO2 + 6. XC + XN
XRN3 + NO3 = NO3 + RNO3
XRN3 + MEO2 = MEO2 + 0.5 RNO3 + 0.5 XN +
3. XC
XRN3 + RO2C = RO2C + 0.5 RNO3 + 0.5 XN + 3.
XC
XRN3 + RO2X = RO2X + 0.5 RNO3 + 0.5 XN + 3.
XC
XRN3 + MCO3 = MCO3 + RNO3
XRN3 + RCO3 = RCO3 + RNO3
XRN3 + BZC3 = BZC3 + RNO3
XRN3 + MAC3 = MAC3 + RNO3
YRPX + NO = NO
YRPX + HO2 = HO2 + ROOH ‐ 3. XC
YRPX + NO3 = NO3
YRPX + MEO2 = MEO2 + 0.5 MEK ‐ 2. XC
YRPX + RO2C = RO2C + 0.5 MEK ‐ 2. XC
YRPX + RO2X = RO2X + 0.5 MEK ‐ 2. XC
YRPX + MCO3 = MCO3
YRPX + RCO3 = RCO3
YRPX + BZC3 = BZC3
YRPX + MAC3 = MAC3
Y6PX + NO = NO
Y6PX + HO2 = HO2 + R6PX ‐ 6. XC
Y6PX + NO3 = NO3
Y6PX + MEO2 = MEO2 + 0.5 PRD2 ‐ 3. XC
Y6PX + RO2C = RO2C + 0.5 PRD2 ‐ 3. XC
Y6PX + RO2X = RO2X + 0.5 PRD2 ‐ 3. XC
Y6PX + MCO3 = MCO3
Y6PX + RCO3 = RCO3
Y6PX + BZC3 = BZC3
Y6PX + MAC3 = MAC3
YAPX + NO = NO
YAPX + HO2 = HO2 + RAPX ‐ 8. XC
YAPX + NO3 = NO3
YAPX + MEO2 = MEO2 + 0.5 PRD2 ‐ 3. XC
YAPX + RO2C = RO2C + 0.5 PRD2 ‐ 3. XC
YAPX + RO2X = RO2X + 0.5 PRD2 ‐ 3. XC
YAPX + MCO3 = MCO3

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

Rate Constant Expression
k = k(70)
k = k(52)
k = k(53)
k = k(54)
k = k(55)
k = k(56)
k = k(56)
k = k(70)
k = k(70)
k = k(70)
k = k(70)
k = k(52)
k = k(53)
k = k(54)
k = k(55)

k300
1.56E‐11
9.23E‐12
7.63E‐12
2.30E‐12
2.00E‐13
3.50E‐14
3.50E‐14
1.56E‐11
1.56E‐11
1.56E‐11
1.56E‐11
9.23E‐12
7.63E‐12
2.30E‐12
2.00E‐13

k = k(56)

3.50E‐14

k = k(56)

3.50E‐14

k=
k=
k=
k=
k=
k=
k=
k=
k=
k=
k=
k=
k=
k=
k=
k=
k=
k=
k=
k=
k=
k=
k=
k=
k=
k=
k=
k=
k=
k=
k=

1.56E‐11
1.56E‐11
1.56E‐11
1.56E‐11
9.23E‐12
7.63E‐12
2.30E‐12
2.00E‐13
3.50E‐14
3.50E‐14
1.56E‐11
1.56E‐11
1.56E‐11
1.56E‐11
9.23E‐12
7.63E‐12
2.30E‐12
2.00E‐13
3.50E‐14
3.50E‐14
1.56E‐11
1.56E‐11
1.56E‐11
1.56E‐11
9.23E‐12
7.63E‐12
2.30E‐12
2.00E‐13
3.50E‐14
3.50E‐14
1.56E‐11

263

k(70)
k(70)
k(70)
k(70)
k(52)
k(53)
k(54)
k(55)
k(56)
k(56)
k(70)
k(70)
k(70)
k(70)
k(52)
k(53)
k(54)
k(55)
k(56)
k(56)
k(70)
k(70)
k(70)
k(70)
k(52)
k(53)
k(54)
k(55)
k(56)
k(56)
k(70)

www.camx.com

March 2016

Number
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514

515
516
517

518

CAMx User’s Guide Version 6.3
Appendix D: Mechanism 5 – SAPRC07TC

Reactants and Products
YAPX + RCO3 = RCO3
YAPX + BZC3 = BZC3
YAPX + MAC3 = MAC3
ZRN3 + NO = NO + RNO3 ‐ 1. XN
ZRN3 + HO2 = HO2 + 6. XC
ZRN3 + NO3 = NO3 + PRD2 + HO2
ZRN3 + MEO2 = MEO2 + 0.5 PRD2 + 0.5 HO2 +
3. XC
ZRN3 + RO2C = RO2C + 0.5 PRD2 + 0.5 HO2 +
3. XC
ZRN3 + RO2X = RO2X + 0.5 PRD2 + 0.5 HO2 +
3. XC
ZRN3 + MCO3 = MCO3 + PRD2 + HO2
ZRN3 + RCO3 = RCO3 + PRD2 + HO2
ZRN3 + BZC3 = BZC3 + PRD2 + HO2
ZRN3 + MAC3 = MAC3 + PRD2 + HO2
XGLD + NO = NO + GLYD
XGLD + HO2 = HO2 + 2. XC
XGLD + NO3 = NO3 + GLYD
XGLD + MEO2 = MEO2 + 0.5 GLYD + XC
XGLD + RO2C = RO2C + 0.5 GLYD + XC
XGLD + RO2X = RO2X + 0.5 GLYD + XC
XGLD + MCO3 = MCO3 + GLYD
XGLD + RCO3 = RCO3 + GLYD
XGLD + BZC3 = BZC3 + GLYD
XGLD + MAC3 = MAC3 + GLYD
XACR + NO = NO + ACRO
XACR + HO2 = HO2 + 3. XC
XACR + NO3 = NO3 + ACRO
XACR + MEO2 = MEO2 + 0.5 ACRO + 1.5 XC
XACR + RO2C = RO2C + 0.5 ACRO + 1.5 XC
XACR + RO2X = RO2X + 0.5 ACRO + 1.5 XC
XACR + MCO3 = MCO3 + ACRO
XACR + RCO3 = RCO3 + ACRO
XACR + BZC3 = BZC3 + ACRO
XACR + MAC3 = MAC3 + ACRO
CH4 + OH = MEO2
ETHE + OH = XHO2 + RO2C + 1.61 XHCH +
0.195 XGLD + YRPX
ETHE + O3 = 0.16 HO2 + 0.16 OH + 0.51 CO +
0.12 CO2 + HCHO + 0.37 FACD
ETHE + NO3 = XHO2 + RO2C + XRCH + YRPX +
XN ‐ 1. XC
ETHE + O3P = 0.8 HO2 + 0.29 XHO2 + 0.51
MEO2 + 0.29 RO2C + 0.51 CO + 0.278 XCO +
0.278 XHCH + 0.1 CCHO + 0.012 XGLY + 0.29
YRPX + 0.2 XC
PRPE + OH = 0.984 XHO2 + 0.984 RO2C +
0.016 RO2X + 0.016 ZRN3 + 0.984 XHCH +
0.984 XCCH + YRPX ‐ 0.048 XC

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

Rate Constant Expression
k = k(70)
k = k(70)
k = k(70)
k = k(52)
k = k(53)
k = k(54)
k = k(55)

k300
1.56E‐11
1.56E‐11
1.56E‐11
9.23E‐12
7.63E‐12
2.30E‐12
2.00E‐13

k = k(56)

3.50E‐14

k = k(56)

3.50E‐14

k = k(70)
k = k(70)
k = k(70)
k = k(70)
k = k(52)
k = k(53)
k = k(54)
k = k(55)
k = k(56)
k = k(56)
k = k(70)
k = k(70)
k = k(70)
k = k(70)
k = k(52)
k = k(53)
k = k(54)
k = k(55)
k = k(56)
k = k(56)
k = k(70)
k = k(70)
k = k(70)
k = k(70)
k = 1.85E‐12 exp(‐1690/T)
Falloff: F=0.6; n=1
k(0) = 1.00E‐28 (T/300)^‐4.5
k(inf) = 8.80E‐12 (T/300)^‐0.85
k = 9.14E‐15 exp(‐2580/T)

1.56E‐11
1.56E‐11
1.56E‐11
1.56E‐11
9.23E‐12
7.63E‐12
2.30E‐12
2.00E‐13
3.50E‐14
3.50E‐14
1.56E‐11
1.56E‐11
1.56E‐11
1.56E‐11
9.23E‐12
7.63E‐12
2.30E‐12
2.00E‐13
3.50E‐14
3.50E‐14
1.56E‐11
1.56E‐11
1.56E‐11
1.56E‐11
6.62E‐15
8.15E‐12

k = 3.30E‐12 (T/300)^2 exp(‐2880/T)

2.24E‐16

k = 1.07E‐11 exp(‐800/T)

7.43E‐13

k = 4.85E‐12 exp(504/T)

2.60E‐11

264

1.68E‐18

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
Appendix D: Mechanism 5 – SAPRC07TC

Number Reactants and Products
519
PRPE + O3 = 0.165 HO2 + 0.35 OH + 0.355
MEO2 + 0.525 CO + 0.215 CO2 + 0.5 HCHO +
0.5 CCHO + 0.185 FACD + 0.075 AACD + 0.07
XC
520
PRPE + NO3 = 0.949 XHO2 + 0.949 RO2C +
0.051 RO2X + 0.051 ZRN3 + YRPX + XN + 2.694
XC
521
PRPE + O3P = 0.45 RCHO + 0.55 MEK ‐ 0.55 XC
522
BD13 + OH = 0.951 XHO2 + 1.189 RO2C +
0.049 RO2X + 0.049 ZRN3 + 0.708 XHCH + 0.48
XACR + 0.471 XIPR + YRPX ‐ 0.797 XC
523
BD13 + O3 = 0.08 HO2 + 0.08 OH + 0.255 CO +
0.185 CO2 + 0.5 HCHO + 0.185 FACD + 0.5
ACRO + 0.375 MVK + 0.125 PRD2 ‐ 0.875 XC
524
BD13 + NO3 = 0.815 XHO2 + 0.12 XNO2 +
1.055 RO2C + 0.065 RO2X + 0.065 ZRN3 +
0.115 XHCH + 0.46 XMVK + 0.12 XIPR + 0.355
XRN3 + YRPX + 0.525 XN ‐ 1.075 XC
525
BD13 + O3P = 0.25 HO2 + 0.117 XHO2 + 0.118
XMA3 + 0.235 RO2C + 0.015 RO2X + 0.015
ZRN3 + 0.115 XCO + 0.115 XACR + 0.001 XAF1
+ 0.001 XAF2 + 0.75 PRD2 + 0.25 YRPX ‐ 1.532
XC
526
ISOP + OH = 0.907 XHO2 + 0.986 RO2C + 0.093
RO2X + 0.093 ZRN3 + 0.624 XHCH + 0.23
XMAC + 0.32 XMVK + 0.357 XIPR + Y6PX ‐
0.167 XC
527
ISOP + O3 = 0.066 HO2 + 0.266 OH + 0.192
XMA3 + 0.192 RO2C + 0.008 RO2X + 0.008
ZRN3 + 0.275 CO + 0.122 CO2 + 0.4 HCHO +
0.192 XHCH + 0.204 FACD + 0.39 MACR + 0.16
MVK + 0.15 IPRD + 0.1 PRD2 + 0.2 Y6PX ‐
0.559 XC
528
ISOP + NO3 = 0.749 XHO2 + 0.187 XNO2 +
0.936 RO2C + 0.064 RO2X + 0.064 ZRN3 +
0.936 XIPR + Y6PX + 0.813 XN ‐ 0.064 XC
529
ISOP + O3P = 0.25 MEO2 + 0.24 XMA3 + 0.24
RO2C + 0.01 RO2X + 0.01 ZRN3 + 0.24 XHCH +
0.75 PRD2 + 0.25 Y6PX ‐ 1.01 XC
530
APIN + OH = 0.799 XHO2 + 0.004 XRC3 + 1.042
RO2C + 0.197 RO2X + 0.197 ZRN3 + 0.002 XCO
+ 0.022 XHCH + 0.776 XRCH + 0.034 XACE +
0.02 XMGL + 0.023 XBAC + Y6PX + 6.2 XC
531
APIN + O3 = 0.009 HO2 + 0.102 XHO2 + 0.728
OH + 0.001 XMC3 + 0.297 XRC3 + 1.511 RO2C
+ 0.337 RO2X + 0.337 ZRN3 + 0.029 CO +
0.051 XCO + 0.017 CO2 + 0.344 XHCH + 0.24
XRCH + 0.345 XACE + 0.008 MEK + 0.002 XGLY
+ 0.081 XBAC + 0.255 PRD2 + 0.737 Y6PX +
2.999 XC

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

Rate Constant Expression
k = 5.51E‐15 exp(‐1878/T)

k300
1.05E‐17

k = 4.59E‐13 exp(‐1156/T)

9.73E‐15

k = 1.02E‐11 exp(‐280/T)
k = 1.48E‐11 exp(448/T)

4.01E‐12
6.59E‐11

k = 1.34E‐14 exp(‐2283/T)

6.64E‐18

k = 1.00E‐13

1.00E‐13

k = 2.26E‐11 exp(‐40/T)

1.98E‐11

k = 2.54E‐11 exp(410/T)

9.96E‐11

k = 7.86E‐15 exp(‐1912/T)

1.34E‐17

k = 3.03E‐12 exp(‐448/T)

6.81E‐13

k = 3.50E‐11

3.50E‐11

k = 1.21E‐11 exp(436/T)

5.18E‐11

k = 5.00E‐16 exp(‐530/T)

8.55E‐17

265

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
Appendix D: Mechanism 5 – SAPRC07TC

Number Reactants and Products
532
APIN + NO3 = 0.056 XHO2 + 0.643 XNO2 +
0.007 XRC3 + 1.05 RO2C + 0.293 RO2X + 0.293
ZRN3 + 0.005 XCO + 0.007 XHCH + 0.684 XRCH
+ 0.069 XACE + 0.002 XMGL + 0.056 XRN3 +
Y6PX + 0.301 XN + 5.608 XC
533
APIN + O3P = PRD2 + 4. XC
534
ACYE + OH = 0.3 HO2 + 0.7 OH + 0.3 CO + 0.3
FACD + 0.7 GLY
535
536

537

538

539

540

541

542

543
544

ACYE + O3 = 1.5 HO2 + 0.5 OH + 1.5 CO + 0.5
CO2
BENZ + OH = 0.57 HO2 + 0.29 XHO2 + 0.116
OH + 0.29 RO2C + 0.024 RO2X + 0.024 ZRN3 +
0.29 XGLY + 0.57 CRES + 0.029 XAF1 + 0.261
XAF2 + 0.116 AFG3 + 0.314 YAPX ‐ 0.976 XC
TOLU + OH = 0.181 HO2 + 0.454 XHO2 + 0.312
OH + 0.454 RO2C + 0.054 RO2X + 0.054 ZRN3
+ 0.238 XGLY + 0.151 XMGL + 0.181 CRES +
0.065 XBAL + 0.195 XAF1 + 0.195 XAF2 + 0.312
AFG3 + 0.073 Y6PX + 0.435 YAPX ‐ 0.109 XC
MXYL + OH = 0.159 HO2 + 0.52 XHO2 + 0.239
OH + 0.52 RO2C + 0.082 RO2X + 0.082 ZRN3 +
0.1 XGLY + 0.38 XMGL + 0.159 CRES + 0.041
XBAL + 0.336 XAF1 + 0.144 XAF2 + 0.239 AFG3
+ 0.047 Y6PX + 0.555 YAPX + 0.695 XC
OXYL + OH = 0.161 HO2 + 0.554 XHO2 + 0.198
OH + 0.554 RO2C + 0.087 RO2X + 0.087 ZRN3
+ 0.084 XGLY + 0.238 XMGL + 0.185 XBAC +
0.161 CRES + 0.047 XBAL + 0.253 XAF1 + 0.253
XAF2 + 0.198 AFG3 + 0.055 Y6PX + 0.586 YAPX
+ 0.484 XC
PXYL + OH = 0.159 HO2 + 0.487 XHO2 + 0.278
OH + 0.487 RO2C + 0.076 RO2X + 0.076 ZRN3
+ 0.286 XGLY + 0.112 XMGL + 0.159 CRES +
0.088 XBAL + 0.045 XAF1 + 0.067 XAF2 + 0.278
AFG3 + 0.286 XAF3 + 0.102 Y6PX + 0.461 YAPX
+ 0.399 XC
B124 + OH = 0.022 HO2 + 0.627 XHO2 + 0.23
OH + 0.627 RO2C + 0.121 RO2X + 0.121 ZRN3
+ 0.074 XGLY + 0.405 XMGL + 0.112 XBAC +
0.022 CRES + 0.036 XBAL + 0.088 XAF1 + 0.352
XAF2 + 0.23 AFG3 + 0.151 XAF3 + 0.043 Y6PX
+ 0.705 YAPX + 1.19 XC
ETOH + OH = 0.95 HO2 + 0.05 XHO2 + 0.05
RO2C + 0.081 XHCH + 0.95 CCHO + 0.01 XGLD
+ 0.05 YRPX ‐ 0.001 XC
ALK1 + OH = XHO2 + RO2C + XCCH + YRPX
ALK2 + OH = 0.965 XHO2 + 0.965 RO2C +
0.035 RO2X + 0.035 ZRN3 + 0.261 XRCH +
0.704 XACE + YRPX ‐ 0.105 XC

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

Rate Constant Expression
k = 1.19E‐12 exp(490/T)

k300
6.09E‐12

k = 3.20E‐11
Falloff: F=0.6; n=1
k(0) = 5.50E‐30 (T/300)^‐2
k(inf) = 8.30E‐13
k = 1.00E‐14 exp(‐4100/T)

3.20E‐11
7.56E‐13

1.16E‐20

k = 2.33E‐12 exp(‐193/T)

1.22E‐12

k = 1.81E‐12 exp(338/T)

5.58E‐12

k = 2.31E‐11

2.31E‐11

k = 1.36E‐11

1.36E‐11

k = 1.43E‐11

1.43E‐11

k = 3.25E‐11

3.25E‐11

k = 5.49E‐13 (T/300)^2 exp(530/T)

3.21E‐12

k = 1.34E‐12 (T/300)^2 exp(‐499/T)
k = 1.49E‐12 (T/300)^2 exp(‐87/T)

2.54E‐13
1.11E‐12

266

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
Appendix D: Mechanism 5 – SAPRC07TC

Number Reactants and Products
545
ALK3 + OH = 0.695 XHO2 + 0.236 XTBU +
1.253 RO2C + 0.07 RO2X + 0.07 ZRN3 + 0.026
XHCH + 0.445 XCCH + 0.122 XRCH + 0.024
XACE + 0.332 XMEK + 0.983 YRPX + 0.017
Y6PX ‐ 0.046 XC
546
ALK4 + OH = 0.83 XHO2 + 0.01 XMEO + 0.011
XMC3 + 1.763 RO2C + 0.149 RO2X + 0.149
ZRN3 + 0.002 XCO + 0.029 XHCH + 0.438 XCCH
+ 0.236 XRCH + 0.426 XACE + 0.106 XMEK +
0.146 XPD2 + Y6PX ‐ 0.119 XC
547
ALK5 + OH = 0.647 XHO2 + 1.605 RO2C +
0.353 RO2X + 0.353 ZRN3 + 0.04 XHCH + 0.106
XCCH + 0.209 XRCH + 0.071 XACE + 0.086
XMEK + 0.407 XPD2 + Y6PX + 2.004 XC
548
OLE1 + OH = 0.871 XHO2 + 0.001 XMEO +
1.202 RO2C + 0.128 RO2X + 0.128 ZRN3 +
0.582 XHCH + 0.01 XCCH + 0.007 XGLD + 0.666
XRCH + 0.007 XACE + 0.036 XACR + 0.001
XMAC + 0.012 XMVK + 0.009 XIPR + 0.168
XPD2 + 0.169 YRPX + 0.831 Y6PX + 0.383 XC
549
OLE1 + O3 = 0.095 HO2 + 0.057 XHO2 + 0.128
OH + 0.09 RO2C + 0.005 RO2X + 0.005 ZRN3 +
0.303 CO + 0.088 CO2 + 0.5 HCHO + 0.011
XCCH + 0.5 RCHO + 0.044 XRCH + 0.003 XACE
+ 0.009 MEK + 0.185 FACD + 0.159 PACD +
0.268 PRD2 + 0.011 YRPX + 0.052 Y6PX + 0.11
XC
550
OLE1 + NO3 = 0.772 XHO2 + 1.463 RO2C +
0.228 RO2X + 0.228 ZRN3 + 0.013 XCCH +
0.003 XRCH + 0.034 XACE + 0.774 XRN3 +
0.169 YRPX + 0.831 Y6PX + 0.226 XN ‐ 1.149
XC
551
OLE1 + O3P = 0.45 RCHO + 0.39 MEK + 0.16
PRD2 + 1.13 XC
552
OLE2 + OH = 0.912 XHO2 + 0.953 RO2C +
0.088 RO2X + 0.088 ZRN3 + 0.179 XHCH +
0.835 XCCH + 0.51 XRCH + 0.144 XACE + 0.08
XMEK + 0.002 XMVK + 0.012 XIPR + 0.023
XPD2 + 0.319 YRPX + 0.681 Y6PX + 0.135 XC
553
OLE2 + O3 = 0.094 HO2 + 0.041 XHO2 + 0.443
OH + 0.307 MEO2 + 0.156 XMC3 + 0.008 XRC3
+ 0.212 RO2C + 0.003 RO2X + 0.003 ZRN3 +
0.299 CO + 0.161 CO2 + 0.131 HCHO + 0.114
XHCH + 0.453 CCHO + 0.071 XCCH + 0.333
RCHO + 0.019 XRCH + 0.051 ACET + 0.033
MEK + 0.001 XMEK + 0.024 FACD + 0.065
AACD + 0.235 PACD + 0.037 PRD2 + 0.073
YRPX + 0.136 Y6PX + 0.16 XC

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

Rate Constant Expression
k = 1.51E‐12 exp(126/T)

k300
2.30E‐12

k = 3.75E‐12 exp(44/T)

4.34E‐12

k = 2.70E‐12 exp(374/T)

9.39E‐12

k = 6.72E‐12 exp(501/T)

3.57E‐11

k = 3.19E‐15 exp(‐1701/T)

1.10E‐17

k = 5.37E‐13 exp(‐1047/T)

1.64E‐14

k = 1.61E‐11 exp(‐326/T)

5.43E‐12

k = 1.26E‐11 exp(488/T)

6.41E‐11

k = 8.59E‐15 exp(‐1255/T)

1.31E‐16

267

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
Appendix D: Mechanism 5 – SAPRC07TC

Number Reactants and Products
554
OLE2 + NO3 = 0.4 XHO2 + 0.426 XNO2 + 0.035
XMEO + 1.193 RO2C + 0.14 RO2X + 0.14 ZRN3
+ 0.072 XHCH + 0.579 XCCH + 0.163 XRCH +
0.116 XACE + 0.002 XMEK + 0.32 XRN3 +
0.319 YRPX + 0.681 Y6PX + 0.254 XN + 0.13 XC
555
OLE2 + O3P = 0.079 RCHO + 0.751 MEK + 0.17
PRD2 + 0.739 XC
556
ARO1 + OH = 0.123 HO2 + 0.566 XHO2 + 0.202
OH + 0.566 RO2C + 0.11 RO2X + 0.11 ZRN3 +
0.158 XGLY + 0.1 XMGL + 0.123 CRES + 0.072
XAF1 + 0.185 XAF2 + 0.202 AFG3 + 0.309 XPD2
+ 0.369 Y6PX + 0.31 XC
557
ARO2 + OH = 0.077 HO2 + 0.617 XHO2 + 0.178
OH + 0.617 RO2C + 0.128 RO2X + 0.128 ZRN3
+ 0.088 XGLY + 0.312 XMGL + 0.134 XBAC +
0.077 CRES + 0.026 XBAL + 0.221 XAF1 + 0.247
XAF2 + 0.178 AFG3 + 0.068 XAF3 + 0.057 XPD2
+ 0.101 Y6PX + 1.459 XC
558
TERP + OH = 0.734 XHO2 + 0.064 XRC3 + 1.211
RO2C + 0.201 RO2X + 0.201 ZRN3 + 0.001 XCO
+ 0.411 XHCH + 0.385 XRCH + 0.037 XACE +
0.007 XMEK + 0.003 XMGL + 0.009 XBAC +
0.003 XMVK + 0.002 XIPR + 0.409 XPD2 + Y6PX
+ 4.375 XC
559
TERP + O3 = 0.078 HO2 + 0.046 XHO2 + 0.499
OH + 0.202 XMC3 + 0.059 XRC3 + 0.49 RO2C +
0.121 RO2X + 0.121 ZRN3 + 0.249 CO + 0.063
CO2 + 0.127 HCHO + 0.033 XHCH + 0.208
XRCH + 0.057 XACE + 0.002 MEK + 0.172 FACD
+ 0.068 PACD + 0.003 XMGL + 0.039 XBAC +
0.002 XMAC + 0.001 XIPR + 0.502 PRD2 +
0.428 Y6PX + 3.852 XC
560
TERP + NO3 = 0.227 XHO2 + 0.287 XNO2 +
0.026 XRC3 + 1.786 RO2C + 0.46 RO2X + 0.46
ZRN3 + 0.012 XCO + 0.023 XHCH + 0.002 XGLD
+ 0.403 XRCH + 0.239 XACE + 0.005 XMAC +
0.001 XMVK + 0.004 XIPR + 0.228 XRN3 +
Y6PX + 0.485 XN + 3.785 XC
561
TERP + O3P = 0.237 RCHO + 0.763 PRD2 +
4.711 XC
562
SESQ + OH = 0.734 XHO2 + 0.064 XRC3 +
1.211 RO2C + 0.201 RO2X + 0.201 ZRN3 +
0.001 XCO + 0.411 XHCH + 0.385 XRCH +
0.037 XACE + 0.007 XMEK + 0.003 XMGL +
0.009 XBAC + 0.003 XMVK + 0.002 XIPR +
0.409 XPD2 + Y6PX + 9.375 XC

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

Rate Constant Expression
k = 2.31E‐13 exp(382/T)

k300
8.25E‐13

k = 1.43E‐11 exp(111/T)

2.07E‐11

k = 7.84E‐12

7.84E‐12

k = 3.09E‐11

3.09E‐11

k = 2.27E‐11 exp(435/T)

9.68E‐11

k = 8.28E‐16 exp(‐785/T)

6.05E‐17

k = 1.33E‐12 exp(490/T)

6.81E‐12

k = 4.02E‐11

4.02E‐11

k = k(558)

9.68E‐11

268

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
Appendix D: Mechanism 5 – SAPRC07TC

Number Reactants and Products
563
SESQ + O3 = 0.078 HO2 + 0.046 XHO2 + 0.499
OH + 0.202 XMC3 + 0.059 XRC3 + 0.49 RO2C +
0.121 RO2X + 0.121 ZRN3 + 0.249 CO + 0.063
CO2 + 0.127 HCHO + 0.033 XHCH + 0.208
XRCH + 0.057 XACE + 0.002 MEK + 0.172 FACD
+ 0.068 PACD + 0.003 XMGL + 0.039 XBAC +
0.002 XMAC + 0.001 XIPR + 0.502 PRD2 +
0.428 Y6PX + 8.852 XC
564
SESQ + NO3 = 0.227 XHO2 + 0.287 XNO2 +
0.026 XRC3 + 1.786 RO2C + 0.46 RO2X + 0.46
ZRN3 + 0.012 XCO + 0.023 XHCH + 0.002 XCCH
+ 0.403 XRCH + 0.239 XACE + 0.005 XMAC +
0.001 XMVK + 0.004 XIPR + 0.228 XRN3 +
Y6PX + 0.485 XN + 8.785 XC
565
SESQ + O3P = 0.237 RCHO + 0.763 PRD2 +
9.711 XC

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

Rate Constant Expression
k = k(559)

k300
6.05E‐17

k = k(560)

6.81E‐12

k = k(561)

4.02E‐11

269

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
Appendix D: Mechanism 5 – SAPRC07TC

Table D‐2. Explicit species in the SAPRC07TC mechanism.
Species Name

BD13
AACD
ACET
ACRO
ACYE
AFG1

AFG2

AFG3
ALK1

ALK2

ALK3
ALK4
ALK5
APIN
ARO1
ARO2
B124
BACL
BALD
BENZ
BZC3
BZO
CCHO
CO3H
CH4
CO
CO2
COOH
CRES
ETHE
ETOH
FACD
GLY
H2
H2O
HCHO
HNO3
PNA

Description

1,3‐butadiene
Acetic acid
Acetone
Acrolein
Acetylene
Lumped photoreactive monounsaturated dicarbonyl
aromatic fragmentation products that photolyze to form
radicals
Lumped photoreactive monounsaturated dicarbonyl
aromatic fragmentation products that photolyze to form
non‐radical products
Lumped diunsaturatred dicarbonyl aromatic fragmentation
product.
Alkanes and other non‐aromatic compounds that react only
with OH, and have kOH between 2 and 5E2 ppm‐1 min‐1.
(Primarily ethane)
Alkanes and other non‐aromatic compounds that react only
with OH, and have kOH between 5E2 and 2.5E3 ppm‐1 min‐1.
(Primarily propane and acetylene)
Alkanes and other non‐aromatic compounds that react only
with OH, and have kOH between 2.5E3 and 5E3 ppm‐1 min‐1.
Alkanes and other non‐aromatic compounds that react only
with OH, and have kOH between 5E3 and 1E4 ppm‐1 min‐1.
Alkanes and other non‐aromatic compounds that react only
with OH, and have kOH greater than 1E4 ppm‐1 min‐1.
‐pinene
Aromatics with kOH < 2E4 ppm‐1 min‐1.
Aromatics with kOH > 2E4 ppm‐1 min‐1.
1,2,4‐trimethyl benzene
Biacetyl
Aromatic aldehydes (e.g., benzaldehyde)
Benzene
Peroxyacyl radical formed from Aromatic Aldehydes
Phenoxy Radicals
Acetaldehyde
Peroxyacetic acid
Methane
Carbon Monoxide
Carbon Dioxide
Methyl Hydroperoxide
Phenols and Cresols
Ethene
Ethanol
Formic Acid
Glyoxal
Hydrogen
Water
Formaldehyde
Nitric Acid
Peroxynitric Acid

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

270

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
Appendix D: Mechanism 5 – SAPRC07TC
Species Name

HO2
H2O2
GLYD
HONO
IPRD
ISOP
M
MAC3
MACR
MPAN
MCO3
MEK

MEO2
MEOH
MGLY
MVK
MXYL
N2O5
NO
NO2
NO3
NPHE
O1D
O2
O3
O3P
OH
OLE1
OLE2
OXYL
PACD
PAN
PAN2
PBZN
PRD2

PRPE
PXYL
R6PX

RAPX

Description

Hydroperoxide Radicals
Hydrogen Peroxide
Glycolaldehyde
Nitrous Acid
Lumped isoprene product species
Isoprene
Atmospheric pressure
Peroxyacyl radicals formed from methacrolein and other
acroleins.
Methacrolein
PAN analogue formed from Methacrolein
Acetyl Peroxy Radicals
Ketones and other non‐aldehyde oxygenated products
which react with OH radicals faster than 5E‐13 but slower
than 5E‐12 cm3 molec‐2 sec‐1. (Based on mechanism for
methyl ethyl ketone).
Methyl Peroxy Radicals
Methanol
Methyl Glyoxal
Methyl Vinyl Ketone
m‐xylene
Nitrogen Pentoxide
Nitric Oxide
Nitrogen Dioxide
Nitrate Radical
Nitrophenols
Excited Oxygen Atoms
Oxygen
Ozone
Ground State Oxygen Atoms
Hydroxyl Radicals
Alkenes (other than ethene) with kOH < 7E4 ppm‐1 min‐1.
Alkenes with kOH > 7E4 ppm‐1 min‐1.
o‐xylene
Propanoic acid
Peroxy Acetyl Nitrate
PPN and other higher alkyl PAN analogues
PAN analogues formed from Aromatic Aldehydes
Ketones and other non‐aldehyde oxygenated products
which react with OH radicals faster than 5E‐12 cm3 molec‐2
sec‐1
Propene
p‐xylene
Lumped organic hydroperoxides with 5 or more carbons
(other than those formed following OH addition to aromatic
rings, which is reprsented separately). Mechanism based on
that estimated for 3‐hexyl hydroperoxide.
Organic hydroperoxides formed following OH addition to
aromatic rings, which is reprsented separately because of
their probable role in SOA formation. Mechanism based on
two isomers expected to be formed in the m‐xylene system.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

271

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
Appendix D: Mechanism 5 – SAPRC07TC
Species Name

RCHO
RCO3
RO3H
RNO3
RO2C

RO2X

ROOH

SESQ
SO2
SULF
TBUO
TERP
TOLU
XACE
XACR
XAF1
XAF2
XAF3
XBAC
XBAL
XC
XCCH
XCO
XGLY
XHCH
XHO2

XGLD
XIPR
XMA3
XMAC
XMC3
XMEK
XMEO
XMGL
XMVK
XN
XNO2
XOH
XPD2
XRCH

Description

Lumped C3+ Aldehydes (mechanism based on
propionaldehyde)
Peroxy Propionyl and higher peroxy acyl Radicals
Higher organic peroxy acids (mechanism based on
peroxypropionic acid).
Lumped Organic Nitrates
Peroxy Radical Operator representing NO to NO2 and NO3
to NO2 conversions, and the effects of peroxy radical
reactions on acyl peroxy and other peroxy radicals.
Peroxy Radical Operator representing NO consumption
(used in conjunction with organic nitrate formation), and
the effects of peroxy radical reactions on NO3, acyl peroxy
radicals, and other peroxy radicals.
Lumped organic hydroperoxides with 2‐4 carbons.
Mechanism based on that estimated for n‐propyl
hydroperoxide.
Sesquiterpenes
Sulfur Dioxide
Sulfates (SO3 or H2SO4)
t‐Butoxy Radicals
Terpenes
Toluene
As for xHO2
As for xHO2
As for xHO2
As for xHO2
As for xHO2
As for xHO2
As for xHO2
Lost Carbon or carbon in unreactive products
As for xHO2
As for xHO2
As for xHO2
As for xHO2
Formation of HO2 from alkoxy radicals formed in peroxy
radical reactions with NO and NO3 (100% yields) and RO2
(50% yields)
As for xHO2
As for xHO2
As for xHO2
As for xHO2
As for xHO2
As for xHO2
As for xHO2
As for xHO2
As for xHO2
Lost Nitrogen or nitrogen in unreactive products
As for xHO2
As for xHO2
As for xHO2
As for xHO2

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

272

www.camx.com

March 2016

CAMx User’s Guide Version 6.3
Appendix D: Mechanism 5 – SAPRC07TC
Species Name

XRC3
XRN3
XTBU
Y6PX
YAPX
YRPX
ZRN3

Description

As for xHO2
As for xHO2
As for xHO2
As for ROOH, but for R6PX
As for ROOH, but for RAPX
Formation of ROOH following RO2 + HO2 reactions
Formation of RNO3 in the RO2 + NO, reaction.

COMPREHENSIVE AIR QUALITY MODEL WITH EXTENSIONS

273

www.camx.com



---

Source Exif Data:

File Type                       : PDF
File Type Extension             : pdf
MIME Type                       : application/pdf
PDF Version                     : 1.5
Linearized                      : Yes
Author                          : cemery
Create Date                     : 2016:04:06 14:01:08-07:00
Modify Date                     : 2016:04:06 14:01:08-07:00
XMP Toolkit                     : Adobe XMP Core 5.2-c001 63.139439, 2010/09/27-13:37:26
Format                          : application/pdf
Title                           : Microsoft Word - CAMxUsersGuide_v6.30.docx
Creator                         : cemery
Creator Tool                    : PScript5.dll Version 5.2.2
Producer                        : Acrobat Distiller 10.0.0 (Windows)
Document ID                     : uuid:afaf01aa-a5e0-4033-86f8-78c8998bb395
Instance ID                     : uuid:2f1411e9-73dd-42bb-80b7-71e0eb631901
Page Count                      : 286

EXIF Metadata provided by EXIF.tools