Microsoft Frost_Nelson_Br_Temp_x Frost Pilot Assessment Of Stream Temp

User Manual: Pdf Frost-Pilot-Assessment-of-Stream-Temp

Open the PDF directly: View PDF .
Page Count: 6

Download
Open PDF In Browser	View PDF

4/13/2017

Nelson Branch Data Loggers

Pictrometry Bird’s Eye © Pictrometry International Corp

Pilot Assessment of Stream Temperature
for an Impaired Waterway
Bill Frost, PE, D.WRE, Sr Water Resources Engineer, KCI Technologies, Inc.
Andy Becker, Project Scientist, KCI Technologies, Inc.
National Stormwater and Watershed Conference
Linthicum, MD
April 4, 2017

Temperature Records

Temperature Data from Nelson Branch
Logger
Number*
1
2
4
5
7
8
•
•

%
Exceeding
20C
8%
33%
15%
21%
14%
50%

%
%
Forest Agriculture
22.6%
65.9%
27.2%
48.6%
46.5%
45.9%
28.9%
57.1%
17.6%
73.7%
28.6%
60.6%

%
%
Urban Impervious
11.5%
3.4%
24.2%
6.3%
7.7%
3.1%
14.0%
4.2%
8.8%
1.4%
10.9%
3.3%

Drainage
Area
(acres)
155.2
152.4
59.2
762.6
41.2
1,109.5

Loggers 3 and 6 were air temperature loggers and are not displayed in this table
WQ criteria allow up to 10% exceedance

Logger
Number*
1
2
4
5
7
8

Maximum
Recorded
Water
Temperature
21.3
23.1
22.5
22.6
22.9
24.9

Date of
Maximum
9/9/2016
7/20/2016
6/16/2016
7/20/2016
8/19/2016
7/20/2016

24 to 27 is the upper lethal temperature for Brook Trout

1

4/13/2017

Temperature – Causes?

Modeling Goals

 No apparent correlation
between temperature
exceedance and watershed
land use

 Relate watershed characteristics to
stream temperature for both Nelson
Branch and in the future for other
County watersheds –

• Forest
• Urban
• Impervious

 Other possible causes
• Lack of stream shading
• Low summer instream
flows
• Increased width to depth
ratio of streams
• Warm water from ponds
• Heated run off during rain
events

Types of Models

• Determine and quantify causes of
increased temperature

Land Cover
Type
Forest
Urban
Impervious

Slope
0.026
0.130
0.035

p-value
0.938
0.514
0.504

R2

• Forecast temperature reduction
based on potential improvements

0.002
0.113
0.119

Types of Models
 Deterministic

 Statistical / Stochastic
• Correlation or regression analysis
or modeling of random variables
• Usually unique to the region
where they were developed
• May require a long time series of
measurements in order to
describe a wide range of
conditions

• Physically‐based with an
energy budget approach
• Heat transfer and fluid flow
equations
• Generally capable of
simulating conditions that
may not be present in the
existing watershed
• More complex and require
more input data

2

4/13/2017

Types of Models

Selection Criteria

 Deterministic

Source of Impairment

 Capability to assess of
sources of impairments

• Physically‐based with an
energy budget approach

 Types of management
measures to be modeled

• Heat transfer and fluid flow
equations

Low summer instream flows

 Availability of input data

• Generally capable of
simulating conditions that
may not be present in the
existing watershed

 Resources, complexity,
and expertise required
 Model support

• More complex and require
more input data

 Types of management
measures to be modeled

Meteorology

 Availability of input data

Hydrology

 Resources, complexity,
and expertise required

Channel
Morphology

 Model support

Topography
Shading
Watershed

Increased width to depth ratio of streams

Warm water from ponds

Heated run off during rain events

Selection Criteria
 Capability to assess of
sources of impairments

Lack of stream shading

Possible Remediation
Restore native riparian vegetation;
increase canopy cover and forest height to
cast longer and denser shadows.
Protect riparian area from unnatural
disturbances; remove non‐native
vegetation
Control livestock access to the stream via
fencing and restoring native riparian
vegetation
Restore headwater watershed features
that retain moisture and allow increased
infiltration; e.g. wetlands, wide riparian
buffers.
Implement storm water and agricultural
BMPs to promote infiltration of runoff.
Stabilize stream morphology to reduce
incision and widening.
Implement stream restoration projects to
create deeper, wider baseflow channels
and connect streams with floodplains.
Convert ponds into wetlands
Remove inline pond connection
Where feasible retrofit pond with bottom
release structure to allow for cooler
bottom water to reach the stream
Implement storm water BMPs such as
infiltration, bioretention, swales, and rain
gardens to promote infiltration.
Disconnect runoff by redirecting it away
from impervious areas to turf or forested
land cover.

Deterministic Models

Input data element
Air temperature
Cloud cover
Wind speed
Humidity
Precipitation
Flow volume
Ponds / reservoirs
Water temperature
Reach length
Width / depth
Slope
Gradient / sinuosity
Substrate
Elevation
Stream aspect
Latitude
Elevation
Riparian vegetation
Impervious cover
Forest cover

Nelson
Branch
data
loggers
x

County
GIS
layers

LGF
WQMP

SHA
RWIS
data
x

County
rain
gages

NOAA
climate
data

Hydrodynamics
Model
Sponsor / Time Step
CEQUAL‐RIV1 USACE
Continuous, Sub‐
Daily
HSPF
USGS
Continuous, Sub‐
Daily

x
x
x
x

QUAL2E

USEPA

x
x
x
x
x
x

x
x
x
x

x
x
x
x
x
x
x

x

SNTEMP

USGS

SSTEMP

USGS/
FWS

HEATSOURCE Oregon
DEQ

Description
Hydrodynamic and water quality model for nutrients, sediment, metals,
bacteria, effects of algae and macrophytes in addition to temperature.
Hydrologic and water quality model; simulates watershed processes on
pervious and impervious surfaces. Along with temperature, output
includes water budget, and pollutant loading. Reach and reservoir
nutrient cycle and biological transformations are also modeled.
Sub‐Daily
Receiving water quality model intended for TMDL development.
Hydrologic, temperature, and pollutant mass balance is calculated for
each subreach.
Steady state,
Heat transport model that predicts daily mean and maximum
Daily to monthly temperature based on stream distance and heat flux from radiation,
convection, conduction, shading, and groundwater inflow.
Steady state,
Scaled down version of SNTEMP which handles single stream reaches for
Daily to monthly a single time period per run. Predicts mean and maximum temperatures
based on heat flux processes: convection, conduction, evaporation, air
temperature, solar radiation, and shading.
Continuous, Sub‐ The model simulates dynamic open channel hydraulics, flow routing, heat
Daily
transfer, effective shade and stream temperature. Processes include mass
transfers, groundwater inflows, landscape radiation, adiabatic cooling,
radiation modeling, evaporation, hydrodynamic routing with hyporheic
exchange within the substrate.

Info Source
Deas and
Lowney (2000)
Deas and
Lowney (2000)

Deas and
Lowney (2000)
Deas and
Lowney (2000)
User Manual

User Manual

3

4/13/2017

Deterministic Models
• Capability to assess of
sources of impairments
• Types of management
measures to be modeled
• Availability of input data

Runoff Modeling

Model
Description
CEQUAL‐RIV1 Developed primarily for water
quality modeling. Temperature
HSPF
modeling is an element of
QUAL2E
water quality.
HEATSOURCE

SNTEMP

• Resources, complexity, and
expertise required
SSTEMP

• Model support

Selection
• More data intensive.
• Pollutant loading input
required.
• More complex.
• Data intensive
• Required GIS analysis of
remote sensing data
• Could be considered for a
future modeling
• Stream network
temperature model
Developed solely for modeling
• Could be considered for
stream temperature.
future modeling with a
hydrologic flow model
• Single reach model
• Less complex version of
SNTEMP
• Input data were available

SSTEMP – Sensitivity and Calibration

 One model identified through
literature search:
• Thermal Urban Runoff Model
(TURM)
• Dane County, WI
• Based on Excel spreadsheet
• Site‐level model rather than a
watershed model
 Predicts temperature changes in
runoff from new development that
adds impervious cover

SSTEMP – Instream Results

 Sensitivity

 Instream Improvements

• Which variables would have the
greatest effect on temperature

• Taking ponds offline
Trials were made varying the assumption that upstream
ponds were present. There was no effect on the mean
temperature.

 Calibration

• Stream restoration

• Varying input data so model results
match field measurements
Description

• Results for four reaches

Calibration Change

Mainstem between Loggers 1 and 5 Shade from 65.5 to 77
Mainstem between Loggers 5 and 8 Shade from 50.7 to 45

Changed Total Shade input
Three of four could be
calibrated

Tributary to Logger 4

Shade from 67.6 to 100

Tributary to Logger 2

Shade from 65 to 87

Trials were made varying the width parameters. Mean
temperatures varied by less than one percent, indicating
that this is not a significant factor in this watershed, or
that SSTEMP’s algorithms do not model variations in
stream width or overwidening well.

• Adding riparian buffer / shade
Increasing buffer shading had a positive effect on
temperature.

4

4/13/2017

SSTEMP – Riparian Buffer Results

TURM – Procedure
 Modeling

 Modeling
• Possible sun was increased to
100% to show the worst case
scenario for unshaded streams.
• Percent of shade was increased
with the goal of meeting a
mean temperature of 20oC

 Cost
• King and Hagan (2011)
estimated $33,000 / ac, or
approximately $150 per tree if
planted at 200 trees per acre.

Reach
Mainstem
between
Loggers 1 and 5
Mainstem
between
Loggers 5 and 8
Tributary to
Logger 2

Measured

100%
Change in
Sun Shaded Shading

TURM was run for one subwatershed of Nelson Branch.

•

Headwaters of the stream draining to Logger #2 ‐ the only area with concentrated impervious
cover at St. James Academy

 Procedure
Ac.

Cost
$36,300

19.98 20.26

19.98 77% to 81%

1.1

21.80 22.41

19.97 45% to 80%

8.3 $273,900

21.02 21.26

20.06

87% to
100%

•

1.7

•

Run TURM to find the
temperature from the
urbanized site.

•

Run TR‐55 to find the volume
of flow for the site and the
remainder of the watershed,
then calculate a weighted
average for runoff
temperature.

•

Calculate a weighted accretion
temperature for the stream.

•

Calculate change in stream
temperature using SSTEMP.

$56,100

TURM – Results

Model Summary



 Both models were relatively easy to use and did not have extensive data
requirements.

Site
• 22.0 oC
• 46.4 oC
• 36.2 oC

Rainfall
Runoff from connected impervious area
Runoff from site



Subwatershed
Undisturbed watershed (assumed same as
• 22.0 oC
rainfall)
Runoff from site
• 36.2 oC
o
Subwatershed weighted by flow volume
• 24.8 C



Stream
• 21.0 oC
• 24.8 oC

Without site runoff
With site runoff

Temperature increase:
Approximately 10%

• Use of the two models was feasible for runoff heating but limited by the lack
of a good linkage between the watershed and stream.

 TURM did not provide a module to test improvements from urban BMPs
such as infiltration, impervious disconnection, grass channels, or level
spreaders.
 SSTEMP did not model changes well from stream widening or shallow
water depth.
 Neither model could successfully estimate temperature changes from
heated water in ponds.

5

4/13/2017

Future Work
 For future analyses using SSTEMP:
• Weather data
 Instream and air temperature
 Dew point temperature or relative
humidity
 Cloud cover, at least daily

• Stream data
 Frequent flow measurements at every
data logger
 Average reach width and depth
 Additional temperature readings at
upstream pond discharges

• Detailed Riparian Vegetation Data





Height
Crown
Offset
Density

 Test other models:
• Statistical / Stochastic
 Maryland‐based empirical model
including seasonal and
urbanization effects (Nelson and
Palmer, 2007)

• Deterministic
 SNTEMP stream network with
watershed hydrologic model
(Krause et al, 2004)
 HEATSOURCE with or without
Thermal Infrared Imagery

Pictrometry Bird’s Eye © Pictrometry International Corp

Questions and Comments

6



---

Source Exif Data:

File Type                       : PDF
File Type Extension             : pdf
MIME Type                       : application/pdf
PDF Version                     : 1.5
Linearized                      : Yes
Author                          : Chris
Create Date                     : 2017:04:13 13:30:28-04:00
Modify Date                     : 2017:04:13 13:30:28-04:00
XMP Toolkit                     : Adobe XMP Core 4.2.1-c043 52.372728, 2009/01/18-15:08:04
Format                          : application/pdf
Creator                         : Chris
Title                           : Microsoft PowerPoint - Frost_Nelson_Br_Temp_Model.pptx
Creator Tool                    : PScript5.dll Version 5.2.2
Producer                        : Acrobat Distiller 9.5.5 (Windows)
Document ID                     : uuid:ba8e3c93-3a22-4708-98c7-5e7b8d1bacf5
Instance ID                     : uuid:54c8f69a-708a-44a4-b573-79cbc818a306
Page Count                      : 6

EXIF Metadata provided by EXIF.tools