Sed Sim User Manual 2018 (v. 1.0.0)

User Manual:

Open the PDF directly: View PDF .
Page Count: 112

Download
Open PDF In Browser	View PDF

SedSim
A River Basin Simulation Screening Model for Reservoir
Management of Sediment, Water, and Hydropower
(version 1.0.0)
Documentation and User's Manual

Source: http://www.usbr.gov/pmts/sediment/projects/Matilija/MatilijaDam.html

Please cite this document as: Wild, T.B., Loucks, D.P. and Annandale, G.W. (in review). SedSim:
A River Basin Simulation Screening Model for Reservoir Management of Sediment, Water, and
Hydropower. Journal of Open Research Software.
Download Software at: https://github.com/FeralFlows/SedSim

Table of Contents
1.

Model Overview .................................................................................................................................... 4

2.

Getting Started...................................................................................................................................... 5

2.1. System Specifications ......................................................................................................................... 5
2.2. Quick-Start Guide ............................................................................................................................... 5
3.

Theoretical Development ................................................................................................................... 13
3.1. Motivation........................................................................................................................................ 13
3.2. Sediment Production and Transport................................................................................................ 13
3.3. Sediment Trapping ........................................................................................................................... 18
3.4. Sediment Management ................................................................................................................... 19
3.4.1. Overview ................................................................................................................................... 19
3.4.2. Simulating Flushing ................................................................................................................... 24
3.4.3. Simulating Bypassing................................................................................................................. 31
3.4.4. Simulating Sluicing .................................................................................................................... 33
2.4.5. Simulating Density Current Venting.......................................................................................... 36
3.4.6. Other sediment removal methods............................................................................................ 42
3.5. Dam and Reservoir Design Features ................................................................................................ 43

4.

Overview of Model File Structure ....................................................................................................... 48

5.

Main Model File Description ............................................................................................................... 49
5.1. Specifying Files Names and Locations.............................................................................................. 49

6.

Input File Description .......................................................................................................................... 51
6.1. Overview of Input Workbook........................................................................................................... 51
6.2. Description of Each Worksheet........................................................................................................ 53
6.2.1. System Schematic and Meta File .............................................................................................. 53
6.2.2. Simulation Parameters and Specifications ............................................................................... 54
6.2.3. Network connectivity ................................................................................................................ 58
6.2.4. Sediment Loads ......................................................................................................................... 61
6.2.5. Incremental Flows ..................................................................................................................... 62
6.2.6. Reach Specifications ................................................................................................................. 63
6.2.7. E-V-A-S ...................................................................................................................................... 65
6.2.8. Storage Volume Elevation Target ............................................................................................. 67

Wild et al. (2019)

SedSim User Manual

Page |2

6.2.9. Storage Volume Target ............................................................................................................. 68
6.2.10. Evaporation Data .................................................................................................................... 69
6.2.11. Environmental Flow Data........................................................................................................ 69
6.2.12. Reservoir Specifications .......................................................................................................... 69
6.2.13. Outlet Capacity Data ............................................................................................................... 77
6.2.14. Sediment Management Specifications Worksheets ............................................................... 78
6.2.15. Flushing ................................................................................................................................... 80
6.2.16. Sluicing .................................................................................................................................... 83
6.2.17. Density Current Venting.......................................................................................................... 85
6.2.18. General Sediment Removal .................................................................................................... 87
6.2.19. Bypassing ................................................................................................................................ 88
6.2.20. IncFlowsCalibration1 ............................................................................................................... 88
6.2.21. Calibration1 ............................................................................................................................. 89
6.2.22. FlowsCalibration2 ................................................................................................................... 89
6.2.23. Calibration2 ............................................................................................................................. 89
7.

Runtime File Description ..................................................................................................................... 91

8.

Output File(s) Description ................................................................................................................... 92
8.1. Overview of Model Output Workbook ............................................................................................ 92
8.2. Time Series Output File .................................................................................................................... 92
8.3. Statistics Output File ...................................................................................................................... 100

9.
10.

Assumptions, Limitations and Caveats ............................................................................................. 103
References .................................................................................................................................... 104

Appendix A: Access to Model Software .................................................................................................... 107
Appendix B: Equations for Flow, Sediment and Hydropower Simulation ................................................ 108

Wild et al. (2019)

SedSim User Manual

Page |3

1. Model Overview
This documentation describes the Sediment Simulation Screening Model (SedSim), a
simulation model for the preliminary screening of sediment transport and management in River
Basins. SedSim is an open-source, daily time step river basin simulation model for water and
sediment flows, and hydropower production, in networks of reservoirs and river channels.
SedSim enables water resources systems analysts and planners to explore alternative system
configurations of reservoir sites, designs (i.e., dam outlet structures), and operating policies
(SDO), and their implications for water flows, sediment transport, reservoir sediment trapping,
and hydropower production in any river basin. The model enables simulation of a wide range of
reservoir sediment management techniques, including flushing, sluicing, density current
venting, bypassing, and dredging. The model performs a daily time-step mass-balance
simulation of flow and sediment that is intended to predict in relative terms the spatial and
temporal accumulation and depletion of sediment in river reaches and in reservoirs under
different reservoir operating and sediment management policies. Thus, the model is expected
to be used for estimating sediment transport in river basis including those that have
experienced (or will experience) extensive reservoir development. The model was originally
developed for use in the Mekong River basin, but can be applied in any river basin. The source
code is written in the Visual Basic for Applications (VBA) language, thus permitting users to
interact with the model using Microsoft Excel. The model requires one user-defined input data
file. The SedSim model was developed at Cornell University, in partnership with the Natural
Heritage Institute (NHI), as well as at the University of Maryland (College Park). The model is
freely available at www.github.com/FeralFlows/SedSim.

Wild et al. (2019)

SedSim User Manual

Page |4

2. Getting Started
The following section is a quick start guide for running SedSim. This guide briefly
describes the steps that are necessary in Excel to run the SedSim model. More detailed
information is provided in later sections of this documentation, especially regarding the
preparation of input data to the model. Users may wish to use the SedSim example case
provided on the SedSim Github repository to test the steps below in the quick-start guide.

2.1. System Specifications
For best performance, the SedSim model should be run using MS Excel 2007 or newer
(i.e., Excel 2010). Depending on the size of the model you build, earlier versions of Excel (e.g.,
Excel 2003) have stricter limitations on the maximum RAM that can be used, and may result in
an inability to run the model due to memory usage errors. The model does not require a fast
processor to operate, but a large model (many system elements and/or long time series) can
require extensive memory usage. In general, 2 GB of RAM should be suitable for most
applications of this model.

2.2. Quick-Start Guide
1. Clone SedSim from Github.
Users can either directly download SedSim files from the model’s Github repository
(https://github.com/FeralFlows/SedSim), or can “git clone” SedSim from the command
prompt with the following command:
“git clone https://github.com/FeralFlows/SedSim.git”
If you are unfamiliar working from the command line, we suggest you try downloading Git
for free here: https://git-scm.com/downloads. This download will come with “git bash”,
from which you can attempt the git clone operation mentioned above. Cloning files rather
than downloading them directly will enable you to pull recent SedSim repository updates
directly to the local repository on your computer.

Wild et al. (2019)

SedSim User Manual

Page |5

Figure 2.1. SedSim Github site.

2. Open up the main (macro) workbook (SedSim.xlsm).
Open the main SedSim model file. This file will be referred to as "SedSim_Model.xlsm"
throughout this documentation for convenience, but the file can be given any name by
simply right-clicking on the file icon when the file is closed and selecting "rename".
Upon opening the workbooks, you may be asked if you wish to enable macros. Click the
“Enable Macros” option to allow the sediment model to execute properly.
Figure 2.2 shows the interface associated with the main workbook of the SedSim model. It
is designed to be generic so it can be used without code modification to run any input file.

Wild et al. (2019)

SedSim User Manual

Page |6

Figure 2.2. Interface for the SedSim model. It ("SedSim_Model.xlsm" workbook) has only one worksheet. Clicking
on the right hand box of line (row) 5 results in the arrow shown at the far right of line 5. Clicking on the arrow will
show options available for line 5, any of which can be selected.

3. Enable macros in security settings.
To be certain that the SedSim model will always run on your computer, in the
“SedSim_Model.xlsm” workbook, in Excel 2007 (or Excel 2010), go to File (or MS Office
Button)  Options  Trust CenterTrust Center SettingsMacro Settings Enable All
Macros. When you are finished running the model in Excel, these settings should be
returned to their original status to avoid potential security threats to your computer.
Alternatively, as described in step 1 above, your version of Excel may provide a warning
message when you first open the SedSim model that asks if you wish to enable the currently
open SedSim model Excel file to be run on your computer, among other Macro options. You
can enable the model to be run on your computer this way as well.

Wild et al. (2019)

SedSim User Manual

Page |7

The next two figures below visually depict the steps described above. From the options
menu within Excel’s “File” tab, select “Trust Center”, from which you can enable macros.

Figure 2.3. Accessing trust center settings in Microsoft Excel.

Wild et al. (2019)

SedSim User Manual

Page |8

Figure 2.4. Setting macro security settings within the trust center in Microsoft Excel.

4. Load in the input data and specify assumptions.
Wild et al. (2019)

SedSim User Manual

Page |9

Load your simulation assumptions and data into the main input file (“SedSim_Input.xlsx”)
files. You can obtain an empty version of this file on the SedSim Github repository.
Alternatively, you can copy the input file from the “example” directory on the paper’s
Github repository (SedSim_Input_Example.xlsx) and use it as an example, replacing the
example data with your data. All colored worksheets will require some input, whereas uncolored worksheets will not require user input and are instead populated during the
execution of the macro. Model-related assumptions (e.g., sediment density) can be
modified in the "Simulation Specifications" worksheet of the input data workbook. Please
review the “Input File” section of this user manual for specific details regarding how to
populate each worksheet within the main input file.
The “SedSim_Model.xlsm” workbook is the only file that is required to be open for the
simulation to run properly. The Input file and output file do not need to be opened
beforehand. The input file will be opened and closed automatically when needed by the
main macro, and the output file is automatically created, saved, and closed by the main
macro.
5. Run the model.
This can be performed by clicking the "Run Model" button in the "SedSim_Model.xlsm"
workbook (Figure 2.2). During the execution of the model, the model will automatically
close the input data file. The model was designed to automatically close the input data file
once all data have been imported into internal arrays because keeping the input file open
can exhaust the maximum memory usage limits of Excel for a large reach/reservoir network
or long simulation duration. The model may produce two different types of error messages
during execution: (1) a detailed error message generated by SedSim that the user must
acknowledge, by clicking "OK" on the automatically generated error message box, before
the simulation can proceed; and (2) an excel VBA error message, which is not likely to
contain detailed instructions, and which is likely the result of improper input data
specification or input/output file naming.
Note: If the model will not run and displays an error regarding the Microsoft Excel “Solver”
package, you may need to install “References” within VBA, as described in Step 5.
6. If you experience errors during a model run, install necessary “References” within Excel
Visual Basic.
Do this by opening up the "SedSim_Model.xlsm" and accessing the VBA code by selecting
Alt+F11 on the keyboard. Within the “Project” menu on the left-hand side of the screen,
select (VBAProject (SedSim.xlsm), click on the main model file to reveal its sub-menu, then
double click the “SedSim_Model” module within the sub-menu.

Wild et al. (2019)

SedSim User Manual

P a g e | 10

In the main menu at the top of the screen, select Tools-->References. Find and check two
boxes: (1) “Solver”, which enables sediment calibration; and (2) “Microsoft Scripting
Runtime”, which enables runtime messages to be printed to a text file during model
execution. Click OK to install the solver references. This is shown in the two images below.
After installing these references, save the file within VBA, and close out of VBA and Excel
altogether. Finally, re-open and re-run the model to see if this change permits the model to
run.

Figure 2.5. Installing references in Visual Basic for Applications (VBA).

Wild et al. (2019)

SedSim User Manual

P a g e | 11

Figure 2.6. Specific references that needs to be installed within Visual Basic for Applications (VBA).

7. Evaluate results.
The results of the simulation run are contained in the “SedSim_Output.xlsx” file. Are these
results reasonable given the input data? One approach to gain confidence in the results is
to create input data for relatively simple systems that should lead to obvious results, and
then see if indeed they did.

Wild et al. (2019)

SedSim User Manual

P a g e | 12

3. Theoretical Development
3.1. Motivation
Reservoirs generally trap all of the inflowing river's bedload and some fraction of the
inflowing suspended load. The extent of trapping depends on many factors, including the
residence time; the reservoir's trap efficiency; the amount, texture and size of inflowing
sediment; and the reservoir's operating policy (Morris and Fan, 1998). Flowing water naturally
transports sediment as a means of dissipating energy, so when a reservoir traps sediment and
discharges the clear (or 'hungry') water downstream, that water has an increased capacity to
scour and transport sediment (Kondolf 1997). This can result in a variety of effects, including
bed incision, armoring of the bed, bank failure due to undercutting, lowering of the
groundwater table, and isolation of the river from its floodplain.
The sections below present details regarding the process of simulating sediment
production, transport, trapping, and reservoir management, as well as flow routing in river
channels and reservoir management of water. Appendix B offers additional details regarding
the SedSim approach to water and sediment transport and storage in reaches and reservoirs.
1.

3.2. Sediment Production and Transport
River basin simulation models like SedSim often include representations of natural
physical processes, which describe how water and sediment move through the natural
landscape (e.g., rainfall and runoff); as well as management processes, which describe the
various ways in which these natural processes may be modified by infrastructure (e.g.,
management of water and trapping of sediment in reservoirs). Striking a balance between
computational demands and enduse purpose, SedSim is limited in its abstraction of detailed
natural physical processes. For example, in the water resources domain, natural physical
processes include rainfall, runoff, groundwater infiltration, and channel flow. Among these
processes, PySedSim accounts only for routing of flows through networks of river channels,
relying on external models for the other processes. With respect to sediment, models are often
classified as either loading models, which account for production of sediment suspended in
watershed runoff, or as receiving models, which route sediment through channels and
reservoirs (Kalin et al. 2003). PySedSim is a receiving model, designed to receive input from a
loading model (e.g., SWAT (Neitsch et al. 2009)). An important difference among receiving
models is the degree to which sediment management processes are included. PySedSim
includes numerous such sediment management processes, such as the trapping of sediment in
reservoirs, the distribution of that sediment within reservoirs’ storage geometry, and the
various ways in which sediment can be removed or passed through or around reservoirs.

Wild et al. (2019)

SedSim User Manual

P a g e | 13

SedSim employs relatively simplistic (e.g., empirical) approaches to representing
complex sediment processes, largely to ensure they are simple enough to be applied in datalimited settings. In keeping with its intended use in a screening setting, SedSim simulates a
single median sediment grain size rather than a grain size distribution. As shown in Figure 1 of
Wild et al. (in review), water and sediment can only enter the modeled system at junctions.
Flows specified at junctions represent incremental, rather than cumulative, daily water and
sediment runoff from the local watershed. The local watershed represents the contributing
watershed area between successive incremental flow junctions. SedSim does not simulate
rainfall, runoff and sediment production processes. Instead, these model inputs must be
externally gathered (e.g., from gage station data) or generated (e.g., simulated with a separate
model). For example, Wild and Loucks (2014, 2015a, 2015b) coupled a calibrated Soil and
Water Assessment Tool (Neitsch et al. 2009) with SedSim for use in the Mekong River basin. In
data-limited river basins where sediment data may be sparse, SedSim also offers the ability to
specify parameters for a rating curve function that describe daily sediment load production as a
function of daily hydrologic flow, which can be calibrated using (1) suspended sediment
concentration and flow time series data, or (2) estimates of annual sediment load (e.g., Kondolf
et al. 2014).
Water and sediment entering a junction in a given day immediately enter the next
downstream channel segment or reservoir, and are thereafter routed through that downstream
element along with any sediment and water entering from upstream. Figure 1 in Wild et al. (in
review) depicts this routing process for an example river channel segment i. Channels are
assumed to be unregulated by hydraulic or other structures. With respect to water flow
routing, SedSim seeks to maintain options consistent with those available in the SWAT model,
which is well-suited to provide water and sediment time series inputs to SedSim. SedSim thus
offers several routing options for determining outflow from river channel segments. This
outflow rate is then used to route sediment through the channel segment. Specifically, as
shown in Figure 1 (of Wild et al., in review), each channel segment is assumed to have a
‘carrying capacity’ (Bagnold 1977) to produce suspended sediment in its outflow as a power
function of its already determined water outflow rate. If the concentration of sediment
suspended in the water column exceeds the channel’s carrying capacity, some sediment settles
to the channel bed (i.e., deposition dominates). Otherwise, sediment is scoured from the
channel bed (i.e., resuspension dominates).
The Soil and Water Assessment Tool (SWAT), which was calibrated for the Mekong River
Basin by the Mekong River Commission, may be used to generate local watershed flows (or
incremental flows). Within SedSim, reaches, reservoirs and diversions are connected by
junction nodes. Runoff from the watershed, which is generated by SWAT, enters the SedSim
model at select junction nodes, after which the water instantaneously enters the reach or
reservoir that sits immediately downstream of the junction. The SedSim model conducts
reservoir operations and reach routing procedures, and tracks the accumulation and depletion
of sediment in reservoirs and reaches, independently of the other models (e.g., SWAT and
RESCON). The RESCON model, which is referenced in Figure 3.1 but discussed in more detail
later in the Sediment Management section of this chapter, is a tool that aids in assessing the
feasibility of applying particular reservoir sediment management techniques at particular
Wild et al. (2019)

SedSim User Manual

P a g e | 14

reservoir sites (Palmieri et al., 2003; Kawashima et al., 2003). While this figure references the
specific models that have been used to conduct simulations in the Mekong basin (i.e., SWAT
and RESCON), other models performing similar functions can just as easily be used instead.

Figure 3.1. An example of the suggested SedSim modeling structure, including other modeling tools suggested to be
used in conjunction with SedSim.

A number of studies have indicated a strong correlation between water flow and
suspended sediment concentration (SSC) in both large and small, and gauged and ungauged
rivers (Milliman and Meade, 1983; Walling and Webb, 1983; Milliman and Syvitski, 1992;
Meybeck et al., 2003; Morehead et al., 2003). Factors such as relief and lithology may also play
important roles in sediment production (Vorosmarty et al., 2003). In keeping with this
commonly observed watershed characteristic, the SedSim model assumes that sediment can
only enter the network of reaches and reservoirs at the same exact locations at which water
flows enter. The rating curve, based on the power regression of SSC, C s (kg/m3), on discharge, Q
(m3/s), is given by
C s = kQx

(2.1)

The relationship in Eq. 2.1 is used for two purposes: (1) to generate daily incremental
sediment loads at locations in the modeled system at which incremental flows are generated
(by an external hydrologic model or other means); and (2) to generate sediment loads to be
discharged from river reaches (channels), in keeping with the concept that each reach has a
'carrying capacity' to produce sediment as a function of reach discharge. The parameters 'k' and
'x' in Eq. 2.1 will be referred to as 'c' and 'd', respectively, when discussing the application of
this general equation to incremental sediment load generation (see Eq. 2.2). Conversely, the
parameters 'k' and 'x' in Eq. 2.1 will be referred to as 'a' and 'b', respectively, when discussing
the application of this general equation to sediment discharge from reaches (see Eq. 2.3).
As was discussed previously, most estimates of sediment loads in the lower Mekong
basin predict that about 80 Mt/yr will be generated. Kondolf et al. (2011) partitioned this 80
Wild et al. (2019)

SedSim User Manual

P a g e | 15

Mt/yr of sediment among nine geomorphic regions, which were delineated based on climatic,
geologic, topographic, and tectonic features. Sediment yields (t/km2-yr) were determined by
Kondolf et al. (2011) for each region. For example, the 3S basin lies within two geomorphic
provinces: the Kon Tum Massif and the Tertiary Volcanic Plateau, which have estimated yields
of 280 t/km2-yr and 290 t/km2-yr, respectively (Kondolf et al., 2011). While this annual
sediment yield information is useful, the SedSim model is operated with a daily time step. Thus,
daily sediment load inputs to junction inflow locations in the SedSim model are required. To
accomplish this, sediment is generated on a daily basis with a version of Eq. (2.2) that has been
uniquely calibrated for each incremental input location. The model user must specify a d i value.
The user may wish to set the parameter ‘d’ so that proportionally more sediment is transported
during higher discharge events, as is often observed in practice (Walling, 2009). The model
determines a c i value for each incremental input location such that the mean annual sum of
daily sediment loads generated in the unregulated system equals the product of the watershed
area that contributes to the incremental flows and the annual sediment yield per unit area
(described above) for the input location. In symbolic form, the generated yields will satisfy the
following equality:

1
N

∑ c (Q
T

t =1

ci =

inc
i

i

)

di

(t ) Qiinc (t )∆t = Aiinc Yi inc for all incremental inflow locations i

(2.2)

Aiinc Yi inc
1
N

∑ (Q
T

t =1

inc
i

(t )

)

d i +1

∆t

where T is the simulation duration (in days), N is the average number of simulation years
(=T/365), c i is the parameter being calibrated for location i, d i is a specified parameter for
location i, Qiinc (t ) is the daily incremental flow at location i, Δt is the time step (number of
seconds in simulation time step in one day), Aiinc is the watershed area (km2) that incrementally
contributes to location i, and Yi inc is the average annual sediment yield per square km (Mt/yrkm2) of the incremental watershed.
Each Eq. (2.2) is solved in Excel, assuming the user chooses to perform parameter
calibration within the model. The model also offers two additional options: (1) specifying one
set of two parameter values to be used for all incremental inflow locations, or (2) specifying a
separate set of two parameter values for every incremental inflow location. Selecting one of
these additional options requires that user determine appropriate values externally.
The model currently assumes that there are no limitations to the sediment supply from
the watershed, in that sediment is continually generated as a function of flow without
exhausting sediment supply. However, sediment availability in river reaches can be optionally
Wild et al. (2019)

SedSim User Manual

P a g e | 16

limited. All sediment that exists within the modeled system, including sediment deposits that
existed within the system prior to the start of simulation and the incremental loads that enter
the system during simulation, are subject to several transport processes. These transport
processes are different for reaches and reservoirs.
For reaches, during a one-day time period, any sediment entering a reach element can
either settle (with the possibility of being eroded at a later time), or can be discharged from the
reach as the model attempts to satisfy the sediment discharge from the reach generated based
on an equation that is identical in form to Eq. (2.1). To clarify, previously discussion of Eqs. (1)
and (2) focused on incremental sediment loading. However, the SedSim model permits
sediment to be generated from within the system as well. Thus, if no sediment incrementally
entered the system from watershed runoff, quantities of sediment would be scoured from
reaches to compensate for this input of sediment-deprived water. (Sediment can only be
generated within the system in reaches, not in reservoirs). The amount of sediment discharged
from a given reach is also in the form of Eq. (2.1), where once again the a value is calibrated
given the b, but slightly differently than they were calibrated for the incremental flows. Again,
the user may wish to set the parameter ‘b’ so that proportionally more sediment is transported
during higher discharge events, as is often observed in practice (Walling, 2009). The reach
sediment rating curve coefficients are calibrated for each reach such that the mean annual sum
of daily sediment loads discharged from the reach in the unregulated system is equal to the
sum of the mean annual sediment loads generated incrementally at all upstream incremental
input locations. Note that while a regulated SedSim model consists of both reaches and
reservoirs, the unregulated system consists only of reaches. Thus, the a value is determined for
the locations in the network where reservoirs are sited, treating the unregulated reservoir site
as a reach. In symbolic form, each a value is determined by the model to satisfy the following
equality:

1
N

∑ a (Q
T

t =1

aj =

j

out
j

)

bj

i∈U

∑ (A

1
N

i∈U
T

∑ (Q
t =1

(

inc inc
(t ) Q out
j (t ) ∆t = ∑ Ai Yi

inc inc
i
i

out
j

Y

(t )

)

)

for all reaches j

(2.3)

)

b j +1

∆t

where T is the simulation duration (in days), N is the average number of simulation years
(=T/365), a j and b j are the parameters being calibrated for reach j, U is the group of all
upstream incremental flow locations i that contribute to the outflow at the outlet of reach j,
inc
is the watershed area
Q out
j (t ) is the daily outflow from reach i, Δt is the time step (one day), Ai
(km2) that incrementally contributes to location i, and Yi inc is the average annual sediment yield
(Mt/yr-km2) for the incremental watershed area.
Wild et al. (2019)

SedSim User Manual

P a g e | 17

Each Eq. (2.3) is solved in Excel, assuming the user chooses to perform parameter
calibration within the model. The model also offers two additional options: (1) specifying one
set of two parameter values to be used for all reach sediment rating curves, or (2) specifying a
separate set of two parameter values for every reach sediment rating curve. Selecting one of
these additional options requires that user determine appropriate values externally.
These same parameters a j and b j for each reach j in the unregulated system are then
stored in the model and are used to determine flow-based sediment discharge from each reach
in the regulated system. Thus, Eq. (2.3) assumes that the 3S basin is in relative balance in its
unregulated state, exporting approximately what is eroded on an average annual basis.
However, because the unregulated system coefficients a j and b j are maintained for the reaches
in the regulated system, alterations of reach flow rates by reservoirs and reduction of sediment
availability due to reservoir sediment deposition can both result in significantly altered
sediment discharge characteristics as given by Eq. (2.3).

3.3. Sediment Trapping
The sediment concentration entering a reservoir is diminished due to the trapping or
settling of sediment in the reduced flow behind the dam. Some fraction of the sediment
entering a reservoir is trapped. Sediment that has previously settled in a reservoir can only be
removed by simulating a sediment management practice, such as flushing. The trapped
fraction, TE(t,r), for each reservoir r in each day t is determined using the Brune (1953) method,
which is depicted in Figure 3.2. The Brune (1953) method uses data from reservoirs in the
United States to predict trapping efficiency as a function of the reservoir’s residence time (or
Capacity:Inflow ratio). Residence time for each simulation day is determined in SedSim using
the average total water storage in the reservoir divided by the outflow or release of water from
the reservoir. SedSim will compute trapping efficiency based on either a running monthly or
annual average of residence time, as specified by the user in the input data file.

Wild et al. (2019)

SedSim User Manual

P a g e | 18

Figure 3.2. Representation of the Brune (1953) curve for estimating sediment trap efficiency of reservoirs.

The volume of sediment deposition in SedSim is computed as the ratio of trapped
sediment mass to the average sediment density. Xue et al., 2010 report the density of sediment
in the Vietnam Delta to be about 1.2 g/cm3 or 1200 kg/m3. The model assumes that sediment
volume remains stable in the reservoir, thus ignoring any compaction processes.
Of the remaining sediment, which is assumed to be of equal concentration throughout
the reservoir volume, some is discharged due to the reservoir water release during the time
period, after which a final concentration is computed that accounts for evaporation losses. The
SedSim model carefully accounts for the impact of sedimentation on reservoir storage volume.
As sediment mass accumulates behind the reservoir during a time step, the maximum volumes
of water that can be maintained in the dead and active storage zones are reduced in the next
time step. This may reduce the total residence time and hence the sediment trapping
efficiencies. It can also alter the release of water needed to achieve a specified storage volume
or head.

3.4. Sediment Management
3.4.1. Overview
Wild et al. (2019)

SedSim User Manual

P a g e | 19

The SedSim model simulates several forms of sediment management in reservoirs,
including flushing, bypassing, sluicing, and density current venting. It also can allow for specific
pre-specified amounts of sediment removed by hydrosuction, dredging, and sediment excluder
devices, but does not simulate those processes. The success of any sediment removal method
depends on many factors, including the reservoir channel shape, reservoir water storage
volume, reservoir hydraulic conditions, and sediment mobility (White, 2000; Morris and Fan,
1998; Habib-ur-Rehman et al., 2009). Once a decision is made that a particular sediment
removal method is feasible for a particular reservoir it can be implemented in the SedSim
model but this model is not capable of determining what sediment management techniques
are technically and economically feasible for a particular reservoir. Rather, the SedSim model
will simulate a sediment technique the user indicates should be simulated, without making any
judgment about whether such a management practice appears to be reasonable for each
reservoir. Such feasibility decisions should be made using a combination of expert judgment
and a pre-feasibility sediment screening tool such as the RESCON model (short for REServoir
CONservation) (Palmieri et al., 2003).
The specific approach taken by SedSim to simulate each of these methods is discussed in
below. However, before discussing SedSim’s approach to simulating specific sediment
management techniques, and the data requirements for the user to do so, it will first be of
value to briefly discuss how the methods SedSim simulates fit in among the range of techniques
that are available for sediment management in reservoirs. A variety of options are available for
managing sediment in reservoirs, and they generally fall into three categories: minimizing
sediment inflow (e.g., catchment management), preventing sediment that does enter the
reservoir from depositing (sediment routing), and removing sediment after it has deposited
(sediment removal) (Annandale, 2012d). Other options include designing the reservoir such
that it is large enough to handle significant accumulation of sediment during the desired
operating period, and designing the reservoir so that sediment accumulation occurs in specific
areas that permit future removal (Morris and Fan, 1998). (Note that these techniques are not
generally exclusive, in the sense that multiple techniques can be often be applied at a particular
reservoir, such as routing during times of high sediment inflows and removal during other times
of year). SedSim allows simulation of sediment routing and sediment removal, because these
methods offer the opportunity to preserve a river basin's erosion and sediment transport
characteristics, which are important for ecosystem health and productivity, rather than simply
focusing on preventing sediment deposition in reservoirs as a means of reducing impacts of
sedimentation on hydropower operations. For example, catchment management refers to
practices that reduce sediment flowing into the reservoir of interest, which could include revegetation, tillage practices (e.g., contour farming), and structural approaches (e.g., check dams
located upstream to prevent sediment deposition in a downstream dam). In the Mekong Basin,
for which SedSim was created, it is important not to reduce high sediment loads, but instead to
preserve sediment transport processes as well as preventing sedimentation.
The difference between sediment routing and sediment removal is quite distinct: The
goal of routing is to prevent deposition to the extent possible by hydraulically routing the
Wild et al. (2019)

SedSim User Manual

P a g e | 20

sediment beyond the reservoir, whereas sediment removal focuses on removing previously
deposited sediment. In this sense, sediment routing is advantageous in that regularly
performed routing can produce reservoir sediment outflows that are consistent in timing and
concentration with the natural sediment inflow regime. To prevent deposition, sediment
routing seeks to manage the sediment-laden portion of reservoir inflows differently than the
clear portion, and is generally done in one of two ways: sediment bypassing and sediment passthrough. Sediment bypassing routes the sediment-laden water around the reservoir to prevent
deposition, whereas sediment pass-through routes the water through the reservoir by
maintaining a high sediment transport capacity.
Both sediment bypassing and sediment pass-through are implemented during high flow
events, which in most parts of the world is when the majority of the annual sediment load is
transported. Common examples of bypassing include bypass tunnels (e.g., the Miwa Dam
bypass system in Japan), river modification (e.g., Nagle Reservoir in South Africa), and offchannel reservoir storage (e.g., Fajardo Dam in Puerto Rico) (Annandale, 2012b). The extent to
which hydropower production is affected by these practices depends on the conditions at the
site. Certainly, if significant quantities of water are bypassed around the reservoir during high
flow periods to transport sediment around the reservoir, less water is stored in the reservoir
and benefits from storage are reduced. However, if the reservoir is properly designed from the
beginning, then reduced inflows are anticipated and planned for, as are the associated losses in
power output and water yield. While some forms of sediment bypassing can require expensive
infrastructure such as bypass tunnels, sediment pass-through almost always requires inclusion
of sediment management infrastructure in the dam itself (such as mid- and low-level outlets),
which is much less costly when included in the initial design.
Common examples of sediment pass-through include sluicing (e.g., First Falls Dam in
South Africa) and density current venting (many applications are in China, including Xiaolangdi
Dam), both of which pass sediment directly through the dam via different combinations of
outlets. Sluicing is not as commonly implemented as sediment management techniques such as
flushing, so establishing the suitability of a particular site and dam for sluicing is not as
straightforward. The goal of sluicing is to maintain a sediment balance, such that the annual
sediment inflow equals the annual sediment outflow. Thus, sluicing is more successful when
performed annually. This is accomplished by partially drawing the reservoir down during times
of high sediment inflow to increase the energy slope (and sediment transport capacity). This
drawdown may be done seasonally for long durations, or for individual flood events. The
impact of sluicing on hydropower production depends on several factors, including: whether or
not power production is even possible during sluicing, which depends factors such as
concentration of sediment, quartzite content in sediment (hardness), and duration of sluicing;
the extent of drawdown (reduction in head) required to achieve the desired increase in
transport capacity; and the duration of sluicing.
While sluicing is likely the most applicable sediment pass-through technique for the
Mekong basin that involves some drawdown of the reservoir, other drawdown pass-through
Wild et al. (2019)

SedSim User Manual

P a g e | 21

techniques are used elsewhere. For example, a reservoir may maintain very low storage for
most of the flood season to permit sediment passage, as is practiced at Sanmenxia Dam in
China (Wang et al., 2005). This approach may seem similar to drawdown flushing, but is
different because the sediment is being discharged before it can settle, and is thus released in a
manner that is more consistent with the natural sediment regime. This, just like sluicing, can be
scheduled to occur on a seasonal basis (e.g., every year during the flood season), or can be
implemented during individual flood events, which may be more applicable in smaller
reservoirs that have the capability to monitor in real time the sediment and water flows in the
basin upstream of the reservoir. On the other hand, the goal of density current venting is to
take advantage of high-density plumes of sediment (called density currents) that may form at
times as sediment flows into a reservoir. Depending on the flow, concentration, and
temperature characteristics at a particular site during a particular event, a density current may
form. If and when the current forms, it can be released through the dam's low-level outlets
upon reaching the base of the dam, generally without significant impacts on hydropower
operations given the relatively low quantity of water that is released during this process.
Another option for managing sediment in a reservoir is sediment removal. There are
two categories of such methods. The first is to physically remove sediment from the reservoir
and place it elsewhere (e.g., into the downstream channel). Examples include dredging; use of
an inline sediment collection device; draining the reservoir and performing dry excavation; and
hydrosuction, which siphons sediment from the bottom of the reservoir to the downstream
channel. Some of these methods (e.g., dredging) are typically very expensive, in some cases
approaching the cost of building a new dam. Others (e.g., hydrosuction) are only applicable to
short reservoirs (Palmieri et al., 2003). While SedSim does not explicitly simulate these
techniques, it does allow for the removal of a specified quantity of sediment mass from a
reservoir over a specified period of time without any changes to reservoir operations, which is
an adequate representation of several of the management techniques described above (e.g.,
dredging, but not dry excavation).
The second category of sediment removal is to implement sediment flushing, of which
there are several types. The purpose of sediment flushing is to remobilize and remove sediment
that has been previously deposited in the reservoir (Atkinson, 1996). This can reduce losses in
reservoir water storage capacity; and can increase sediment loads being discharged
downstream. Flushing is conducted by opening low-level (and often mid-level) gates. This
causes an increased flow of water through the reservoir, resuspending deposited sediment and
discharging both through the gates. There are two kinds of flushing: drawdown flushing (free
flow flushing) and pressure flushing (partial drawdown flushing) (Atkinson, 1996; White, 2001;
Palmieri et al., 2003). The SedSim model only simulates drawdown flushing, but the differences
between these two approaches are clarified in the section below that is dedicated to flushing in
SedSim.
If sluicing and flushing are to be performed annually, the significant difference between
the two approaches is in the timing of sediment release. Sediment flows may be more naturally
Wild et al. (2019)

SedSim User Manual

P a g e | 22

preserved with sluicing, whereas a sudden release of sediment over a shorter period of time
may result with flushing. However, sluicing is more likely to remove only the finer fractions of
sediment, whereas flushing can remove sediment sizes up to sand and gravel (depending on
the magnitude of the flushing flow that is employed).
The purpose of sediment flushing is to remobilize and remove sediment that has been
previously deposited in the reservoir (Atkinson, 1996). This can reduce losses in reservoir water
storage capacity; and can increase sediment loads being discharged downstream. Sediment in
river flows impacts the river's geomorphological makeup and ecosystem habitats. Flushing is
conducted by opening low-level flushing gates. This causes an increased flow of water through
the reservoir, resuspending deposited sediment and discharging both through the gates. There
are two types of flushing: drawdown flushing and pressure flushing (Atkinson, 1996; White,
2001; Palmieri et al., 2003). The SedSim model only simulates drawdown flushing. Each will be
discussed separately next.
Drawdown flushing requires reducing water levels in the reservoir enough to permit
free flow conditions through the low-level outlets. For this to happen the low-level outlets
should be located near the original river bed elevation, and should have the capacity to
discharge streamflow during the flushing period without significant ponding behind the dam
(Palmieri et al., 2003). Drawdown typically begins at the beginning of the high flow season after
a period of low inflows and hence low storage volumes. The high flows through the reservoir
are more effective in resuspending sediment than low flow values. Low reservoir water levels
must be maintained during the flushing period. Thus, drawdown flushing is typically performed
at the beginning of the high flow season. The appropriate recurrence interval for flushing
depends on the conditions at the reservoir site. Regularly performed flushing, if conducted at
the right time, can be more environmentally beneficial than less regularly performed flushing.
This is because the amount of sediment released during each flushing event may contain
sediment concentrations (and durations of those concentrations) that more closely resemble
the river's natural high-flow sediment conditions, in comparison to flushing events designed to
discharge sediment that has collected over much longer periods of time. However, more
frequent flushing results in more reductions in hydropower production and hence power
reliability is lower.
Pressure flushing is not included in the SedSim model. It is different from drawdown
flushing in that much higher water levels are maintained in the reservoir during pressure
flushing. While avoiding drawdown of the reservoir to very low storage levels may permit
increased hydropower production in comparison to drawdown flushing, pressure flushing is
only effective at remobilizing and discharging sediment located in the vicinity of the low-level
outlet, as well as relocating sediment from upstream portions of the reservoir to downstream
portions of the reservoir. In general, drawdown flushing is capable of removing larger quantities
of sediment, and from more locations in the reservoir, than pressure flushing.

Wild et al. (2019)

SedSim User Manual

P a g e | 23

3.4.2. Simulating Flushing

The following information must be supplied by the user for every reservoir at which
sluicing will be simulated. Some user inputs are described below as Optional, meaning these
inputs are extra features that are not required to run a simulation. More details on these inputs
are provided in the discussion of “Flushing” worksheet, where most of these inputs are
required to be entered.
1. Target flushing start date. Worksheet: “Flushing”.
2. Flushing duration. Worksheet: “Flushing”.
3. Minimum inflow rate required to initiate drawdown for flushing after the date specified
above (Optional). Worksheet: “Flushing”.
4. Target water surface elevation during flushing. Worksheet: “Outlet Capacity Data”. This is
the water surface elevation target during drawdown and flushing. SedSim will establish this
value by importing the first elevation in the low-level outlet capacity-discharge table, which
should represent the elevation of the low-level outlet). This is generally close to the original
river bed elevation.
5. Maximum water surface elevation (WSE) that will still result in successful flushing.
Worksheet: “Flushing”.
6. Minimum discharge through the low-level outlets that will still result in successful flushing.
Worksheet: “Flushing”.
7. Maximum flushing drawdown rate (Optional). Worksheet: “Flushing”.
8. The representative reservoir bottom width close to the dam. (This information is used to
determine how much sediment is removed during flushing. More details are available later
in this section). Worksheet: “Flushing”.
9. The representative (average) side slope of the reservoir banks. (This information is used to
determine how much sediment is removed during flushing. More details are available later
in this section). Worksheet: “Flushing”.
10. The representative bottom width of the flushing channel. (The model will calculate this as a
function of other inputs described above if the user does not have this information).
Worksheet: “Flushing”. This information is used to determine how much sediment is
removed during flushing.

Wild et al. (2019)

SedSim User Manual

P a g e | 24

11. The representative (average) side slope of the flushing channel banks. (The model will
calculate this as a function of other inputs described above if the user does not have this
information) Worksheet: “Flushing”. This information is used to determine how much
sediment is removed during flushing.
12. Coefficient value, k, for sediment load generation during Flushing, used in equation kQm
(Optional). Worksheet: “Flushing”. (Instead of computing the sediment loads discharged
during flushing via the methods described below in this section, the user can instead specify
parameters to be used in the equation kQm to determine sediment discharge from the
reservoir each day during flushing as a function of reservoir outflow).
13. Exponent value, m, for sediment load generation during Flushing, used in equation kQm
(Optional). Worksheet: “Flushing”. (Instead of computing the sediment loads discharged
during flushing via the methods described below in this section, the user can instead specify
parameters to be used in the equation kQm to determine sediment discharge from the
reservoir each day during flushing as a function of reservoir outflow).
14. A discharge capacity vs. elevation table for the low-level outlet that will be used for flushing.
Worksheet: “Outlet Capacity Data”.
The SedSim model flushing procedure consists of three components: Drawdown,
Flushing, and Refill. The user must supply inputs related to these three processes, all of which
are described below and depicted in Figure 3.3, which demonstrates an example of SedSim
simulation results for reservoir water storage during flushing.

Wild et al. (2019)

SedSim User Manual

P a g e | 25

Figure 3.3. Depiction of the flushing process in SedSim. This figure plots simulation results for an example reservoir
being flushed. In the figure, Qin(t) represents the reservoir inflow in time period t, Qflush represents the minimum
flow required to achieve successful flushing, Qmin represents the minimum inflow required before reservoir
drawdown is initiated, WSEres(t) represents the water surface elevation in the reservoir in period t, and WSEmax
represents the maximum reservoir water surface elevation for flushing to be successful (typically this is equal to or
within a few meters the elevation of the low-level outlet(s)).

On or after the date on which the user specifies drawdown is to be initiated, the SedSim
model initiates the drawdown process once the reservoir inflow exceeds the minimum inflow
target set by the model user. (In other words, the user can establish a minimum reservoir
inflow required to initiate drawdown that is different from the minimum inflow/outflow
required to achieve successful flushing. The reservoir's low-level gates are opened to attempt
to draw down the reservoir to the lowest possible storage so flushing can occur. (The model
assumes these gates are already installed). The low-level gates are only opened when flushing
is to be attempted, and they are closed as soon as flushing is complete. Other outlets can be
used to drain the reservoir during the drawdown period, including the hydropower outlets, as
long as the reservoir's water surface elevation during the time period of interest is large enough
that the outlet has discharge capacity. Beginning with the first day of reservoir drawdown, the
reservoir's pre-established operating policy is temporarily overridden to conduct flushing. In
other words, the model does not require the user to modify the pre-existing operating policy
(e.g., elevation targets, storage targets, or other policy data) during the time period over which
flushing is to occur. Rather, during the time frame when flushing is to occur, flushing is assumed
to be the primary goal of operation, and the model operates the reservoir to attempt to satisfy
flushing criteria, which are discussed below. The duration of drawdown is not specified by the
user, but instead depends on the water storage at the start of drawdown, the maximum
drawdown rate, and the capacity of the reservoir's outlets to release water as the reservoir is
drained.
During the drawdown process, the Trapping Efficiency (TE) of the reservoir is assumed
to be zero. Similar to normal reservoir operations, the only sediment that can be released from
the reservoir during drawdown is sediment contained in suspension in the reservoir water
volume. That is, no sediment is removed from the sediment mass that has previously settled to
the bottom of the reservoir. Water is released from the reservoir only through outlets that have
capacity given the reservoir's water surface elevation.
The goal is to keep the reservoir’s storage volume as low as possible, so as not to exceed the
maximum water surface elevation required to achieve successful flushing (as defined by the
user).
Flushing is assumed to begin on the first date on which the following criteria are
satisfied at the reservoir:
1. The water surface elevations (mamsl) at the beginning and end of the time period do not
exceed the maximum flushing water surface elevation specified by the user.
Wild et al. (2019)

SedSim User Manual

P a g e | 26

2. The flushing discharge (discharge through the low-level outlets) exceeds the minimum
flushing discharge requirement (m3/s).
Once these criteria have been satisfied for the number of days over which the user
specifies flushing should occur, flushing is complete and refill begins in the next time period. If
either constraints are not satisfied on a particular day, then a day is added to the number of
specified flushing days. If flushing requirements fail to be satisfied, no sediment is removed
from the deposited sediment mass, although already suspended sediment can be discharged
from the reservoir via the low-level outlets, or can remain in suspension in the water stored in
the reservoir. The model will continue to attempt to satisfy the flushing requirements until the
specified number of flushing days has occurred. Due to the significant uncertainty in estimating
the discharge requirements to achieve successful flushing, the model assumes that if the flow
constraints are satisfied to within 20% of the provided values, flushing is successful.
During the flushing period, the TE of the reservoir is assumed to be zero. The volume of
sediment that is removed from the reservoir as a result of flushing is removed from the settled
sediment mass and is equally distributed in the discharge downstream over the user-specified
flushing horizon. No more sediment mass than is available can be removed from the reservoir
as a result of flushing. Sediment is assumed to be removed from segments of the elevationvolume-area curve in the same manner in which sediment was assumed to deposit in the
reservoir. For example, if sediment is deposited linearly throughout the elevation range, then
flushing will result in removal of sediment from all elevations in the same manner. The quantity
of sediment removed during each flushing event is determined via a process that is described at
the end of this section on flushing. Flushed sediment accumulates downstream of the flushing
channel and is subject to being picked up and further transported downstream depending on
the unsatisfied sediment carrying capacities of the reach flows.
During the flushing period, only those outlets that maintain a positive release capacity
at the reservoir's water surface elevation each day can be used to release water and sediment.
Generally, only the low-level outlets will have discharge capacity at such low elevations, thus
preventing any hydropower production during flushing. The water surface elevation cannot
drop below the minimum elevation at which the low-level outlet has capacity to release water.
Thus, if any water remains in storage below the low-level outlet, which is not likely to be much
water given that the low-level outlets are best positioned near the original river bed elevation,
the low-level outlets would not have capacity to release this water. Any storage or elevation
target set to a level below the low-level outlet will result in a surface elevation close to that of
the low-level outlet.
Note that the goal of maintaining low storage during flushing does not mean that zero
water volume is maintained in the flushing channel; rather, this means that water volume
inflow is similar to water volume outflow during the time period. The goal of flushing is to
permit free flow through the reservoir. Thus, at the beginning and end of every time period
during which flushing occurs, in a real reservoir some water will always exist in storage in the
Wild et al. (2019)

SedSim User Manual

P a g e | 27

flushing channel, because there is a constant flow of water into and out of the channel
throughout the day. However, one result of conducting a daily mass balance in the reservoir
without routing is that all water flowing into the reservoir each day is assumed to be
immediately available for release, with storage recorded only at the beginning and end of each
day. During the flushing period, this effectively results in all of the inflow being stored
immediately behind the dam, and the low-level outlet thus has the capacity (and the goal) of
releasing the stored water right away. This means that during flushing, while a storage close to
zero m3 is recorded at the beginning and end of each day, in a real reservoir there is storage of
water maintained within the flushing channel throughout the flushing period.
The refill period begins on the day after drawdown is completed (i.e., the day after the
drawdown maximum elevation and minimum discharge goals have been satisfied for the
specified number of days). The reservoir's pre-established operating policy is re-established
during the refill period. For example, if the reservoir has a pre-established water storage or
water surface elevation target for the day after flushing is completed, the reservoir will not
release any water until this target is met. Hydropower production is possible during refill, but
only once the water surface elevation is high enough to permit a turbine outlet discharge
capacity greater than zero.
Next, note that density current venting, flushing and sluicing are assumed to be exclusive
activities, in that they cannot be conducted at the same time. Multiple sediment management
techniques can be simulated in the same reservoir at different times. However, flushing,
sluicing and density current venting cannot be simulated concurrently. Any management
technique being simulated will be allowed to finish before a new technique is begun. For
example, if sluicing is being simulated at a particular reservoir and flushing is meanwhile
scheduled to occur, the start of flushing will be delayed until sluicing is completed. If the user
schedules two or more sediment management techniques to start on the same date, priority is
given first to flushing, then to sluicing, and finally to density current venting.
The SedSim approach to determining the quantity of sediment removed during a flushing event
is as follows.
1. During each time step, determine the deposited sediment volume, V d (t)
2. During each time step, determine the depth of the deposited sediment layer, d(t).
To accomplish Step 2, SedSim first determines the average Area, A, over which the sediment is
deposited during the time step. This value is taken to be constant for the duration of
simulation. This is estimated using the average surface area of the reservoir, or

𝐴𝐴 =

𝑉𝑉𝑇𝑇𝑇𝑇

𝐸𝐸𝐸𝐸𝑎𝑎 −𝐸𝐸𝐸𝐸𝑏𝑏

Wild et al. (2019)

(2.4)
SedSim User Manual

P a g e | 28

where A is the average water surface area in the reservoir, V TK is the reservoir’s total storage
capacity, El a is the elevation at the top of the active storage zone, and El b is the elevation at the
bottom of the reservoir (likely the original river bed elevation). These two values will be taken
from the user-specified elevation-volume table.
The area, A, is then used to determine the depth of the deposited sediment layer, d(t), as
follows:
𝑑𝑑(𝑡𝑡) =

𝑉𝑉𝑑𝑑 (𝑡𝑡)

(2.5)

𝐴𝐴

(Note: This approach is a simplification. In reality, in each time step the sediment is deposited
over the reservoir water surface area, which changes in each time step).
3. Determine the fraction of the sediment layer deposited in time period t that sits within the
incised channel to be formed by flushing, which represents the quantity of sediment that can
be removed via flushing.
Every flushing event results in removal of some fraction of the volume of sediment that has
settled since the last flushing event. If the reservoir reaches its sustainable long-term storage
capacity, K f , which is determined in SedSim, the model assumes that all of the sediment that
has settled since the last event can be removed. The Long Term Capacity Ratio (LTCR) (see
Atkinson (1996) for more details) represents the ratio of the long term storage capacity that can
be sustainably maintained (in perpetuity) with frequent and successful flushing, K f , to the initial
storage capacity, K o , as given by the following:
𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿 =

𝐾𝐾𝑓𝑓

(2.6)

𝐾𝐾𝑜𝑜

Figure 3.4 represents a simplified version of a reservoir’s cross-sectional geometry that enables
a quick calculation of the LTCR of any reservoir that will be frequently flushed. In this figure, the
area within the inner trapezoid (denoted by the letter “B”) represents the cross-sectional area
that can be maintained in perpetuity by frequent and effective flushing. (Note that this
sustainable area is assumed to extend the length of the reservoir, thus forming a sustainable
storage volume.) The area of the outer trapezoid (denoted by the letter “A”) represents the
total representative cross-sectional area of the reservoir. While the flushing bottom elevation
(elevation of low-level outlet) is higher than the reservoir bed elevation at the dam (original
river bed elevation) in this figure, the user can locate the low-level outlet at the bottom
elevation of the dam, in which case the bottom of the flushing channel coincides with the
bottom of the dam.

Wild et al. (2019)

SedSim User Manual

P a g e | 29

Figure 3.4. Simplified cross-sectional geometry of a reservoir (outer trapezoid) and the sustainable channel (inner
trapezoid) that can be formed by frequent and effective flushing.

The ratio of the two areas “B” and “A” in Figure 3.4 defines the LTCR, or
𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿 =

𝐵𝐵

(2.7)

𝐴𝐴

Note that SedSim accounts for many possibilities regarding the geometries of areas “A” and
“B”. For example, SedSim accounts for the circumstance in which the flushing channel side
slope is low enough that the flushing channel will eventually meet the side of the simplified
reservoir geometry before the flushing channel reaches the top of the reservoir.
SedSim tracks the evolution of the flushing channel as sediment layers deposit within the
reservoir in each time step. When the flushing channel has finally reached its sustainable
storage capacity (given by the product of the area “B” in Figure 3.4 and the reservoir length),
flushing is capable of removing all of the sediment that has deposited since the previous
flushing event. Also, if the flushing channel has the potential to be larger than the reservoir’s
geometry from the start (i.e., “B” is greater than “A” in Figure 3.4), then complete removal of
settled sediment is possible throughout simulation. Generally, it takes some time for enough
sediment accumulation to occur before a reservoir reaches the LTCR. Thus, if a reservoir that is
regularly flushed has not yet reached its LTCR, the fraction of settled sediment that is removed
during a particular flushing event is assumed to be equal to the ratio of the current flushing
channel top width in the simplified reservoir geometry, W f (t), to the width of accumulated
sediment deposits in the simplified reservoir geometry, W s (t). The assumption is that sediment
deposits in an equally-distributed manner within this simplified reservoir geometry. Thus, a
fraction of any deposited sediment layer will fall within the confines of the channel formed by
flushing and will thus be removed when flushing occurs, whereas the rest of the sediment will
be located outside of the flushing channel boundary and will thus never be removed. In
equation form, the fraction of the mass, f m (t) in a sediment layer deposited in a reservoir in
time period t that can be removed in the next flushing event is given by the following
relationship
Wild et al. (2019)

SedSim User Manual

P a g e | 30

𝑓𝑓𝑚𝑚 (𝑡𝑡) =

𝑊𝑊𝑓𝑓 (𝑡𝑡)

(2.8)

𝑊𝑊𝑠𝑠 (𝑡𝑡)

Given that the layer of sediment deposited outside the confines of the flushing channel is
growing as the simulation proceeds and more sediment accumulates, and given that the
flushing channel geometry and reservoir geometry have different bottom widths and side
slopes, the fraction in Eq. (2.8) can change in every time step. The fraction may increase or
decrease depending on the relative shapes of the cross-sectional geometries.
The widths in Eq. (2.8) above is depicted in Figure 3.5 below.

Figure 3.5. Simplified cross-sectional geometry of a reservoir (outer trapezoid) and the sustainable channel (inner
trapezoid) that can be formed by frequent and effective flushing. The brown area represents an example of
sediment that has previously deposited in the reservoir up to time period t. Only the fraction of any newly
deposited sediment layer in time period t+1 that deposits within the boundaries of the theoretical flushing channel
will be removed in the next flushing event.

3.4.3. Simulating Bypassing

Bypassing divides the flow into two parts. The bottom portion flows into the reservoir
and the top portion gets bypassed to a point downstream of the reservoir outlet. Bypassing can
be implemented any time but is usually implemented when the flow is high, and hence carrying
more sediment than a lower flow would. All sediment and flow directed into the bypass are
discharged into the downstream channel without any routing considered.

Wild et al. (2019)

SedSim User Manual

P a g e | 31

To implement bypassing in the SedSim model the user must specify the minimum inflow
rate (m3/s) when bypassing would start, and the maximum bypass discharge capacity.
Bypassing is assumed to start when the total reservoir inflow (before bypassing is considered)
equals or exceeds that minimum specified inflow rate, and stops when the inflow rate drops
below the specified minimum inflow rate. Only inflow in excess of the minimum inflow rate is
bypassed, with the remainder entering the reservoir. However, when the reservoir inflow in
excess of the minimum bypass flow rate also exceeds the maximum bypass discharge capacity,
the flow in excess of the discharge capacity also enters the reservoir.
The default model assumption is that sediment is partitioned between the bypass and
reservoir in proportion to the fractions of total inflow that are distributed into the bypass and
reservoir. If desired, the user can specify what fraction of the sediment that would otherwise
have entered the reservoir (based on the proportion of total inflow that enters the reservoir)
should instead be distributed into the bypass. This option was implemented to reflect that
concentration increases with depth of flow, and thus the bypass may remove more of the
inflowing sediment than just the proportion of flow diverted into the bypass.
To further explain this reservoir inflow fraction assumption, suppose for a particular
reservoir that the minimum bypass flow is 50 m3/s, the bypass capacity is 80 m3/s, and the total
inflow is 120 m3/s. Since the inflow exceeds the minimum flow requirement, a bypass will
occur. All flow in excess of the minimum threshold will be bypassed, or 120 m3/s – 50 m3/s = 70
m3/s. The remaining 50 m3/s will enter the reservoir. At least 70/120 = 0.583 (58%) of total
sediment inflow (kg) will be bypassed. However, by specifying a bypass fraction in the user
interface, the user can establish what fraction of the remaining 42% of the sediment load will
be diverted into the bypass instead of entering the reservoir. If the user enters no fraction (0%),
this corresponds to the default assumption, which is that all of the remaining 42% of sediment
flows into the reservoir. If the user chose, for example, 50% instead, in this example the bypass
would receive 58% + 0.5*42% = 79% of the total sediment load, while the reservoir would
receive the remaining 21%.
The following information must be supplied by the user for every reservoir at which bypassing
will be simulated. More details on these inputs are provided in the discussion of the sediment
management specifications worksheet (“Bypassing”), where these inputs are required to be
entered.
1. Flow rate above which bypass is activated during wet season (m3/s). Worksheet:
“Bypassing”. Minimum reservoir inflow rate at which the sediment bypass is opened during
the monsoon season and sediment and flow begins to be discharged around the reservoir.
2. Bypass discharge capacity (m3/s). Worksheet: “Bypassing”.
3. Fraction of sediment load in reservoir inflow (SedSim will establish a default value if the user
does not specify a value). Worksheet: “Bypassing”. Allows the user to describe how
Wild et al. (2019)

SedSim User Manual

P a g e | 32

sediment is partitioned between the bypass, which diverts sediment around the reservoir,
and the remaining sediment that enters the reservoir.
4. Minimum required bypass flow (m3/s). Worksheet: “Bypassing”. This is the minimum water
flow that is required to be diverted into the bypass channel(s) each day (if inflow is
sufficient).
5. Minimum required bypass flow (fraction). Worksheet: “Bypassing”. Similar to the minimum
bypass required flow, except this value represents the fraction of the total daily site water
inflow required to be diverted into the bypass channel at all times.
3.4.4. Simulating Sluicing

When the simulation date reaches a user-specified beginning date of sluicing, the
reservoir will be drawn down to the associated user-specified water surface elevation (mamsl)
target, permitted that all optional inflow rate and concentration conditions (described below)
are met. Note that density current venting, flushing and sluicing are assumed to be exclusive
activities, in that they cannot be conducted at the same time. Any management technique
being simulated will be allowed to finish before a new technique is begun. For example, if
sluicing is being simulated at a particular reservoir and flushing is meanwhile scheduled to
occur, the start of flushing will be delayed until sluicing is completed. If the user schedules two
or more sediment management techniques to start on the same date, priority is given first to
flushing, then to sluicing, and finally to density current venting.
While the reservoir is drawn down to the target level, the drawdown rate (m/day) will
be restricted to the user-specified maximum rate. On the beginning date of sluicing, and for the
duration of sluicing, sediment trapping is defined by the Churchill curve (see below for more
discussion). When the simulation date reaches the day after the user-specified sluicing end
date, normal sediment trapping will resume (Brune's curve trapping), unless another form of
trapping is activated immediately after the completion of sluicing. During sluicing, even if the
reservoir cannot be drawn down to the target elevation on a given day, sluicing is still assumed
to successfully occur. Thus, if drawdown takes a significant amount of time compared to the
total sluicing duration, a significant percentage of the sluicing period could consist of
drawdown.
If the user specifies that no hydropower will be produced during sluicing, then only the
mid- and low-level gates can be used to drain the reservoir (not the hydropower outlets). If the
user specifies that hydropower production is possible during sluicing, then water will be
released through the hydropower outlets throughout the sluicing process, assuming the
hydropower outlets have capacity to release flow.

Wild et al. (2019)

SedSim User Manual

P a g e | 33

Once the reservoir is drained down to the target sluicing elevation, the mid-level outlets
and low-level outlets are used to achieve sluicing. The low-level outlets are only used during
sluicing if (1) Only low-level outlets exist (no mid-level outlets are installed); and/or (2) the
inflow is too high for the mid-level outlets to release to maintain the target elevation. If the
low-level outlets are used to release water during sluicing, SedSim assumes that no sediment is
released as a result of scour in the vicinity of the low-level outlet.
The Churchill (1948) Curve, which appears in Figure 3.6, is used to determine trap
efficiency during the sluicing period (on or in between the beginning and ending dates of
sluicing). A new trap efficiency value is computed every day depending on the residence time of
water in the reservoir. The Churchill Curve is used instead of the Brune Curve because it
produces improved sediment passage approximation for reservoirs that have been drawn
down, and are therefore hydrologically smaller than during normal operations.

Figure 3.6. Representation of the Churchill (1948) Curve for estimating sediment release efficiency (100-trap
efficiency). The yellow uncertainty band demonstrates the ability to sample capture uncertainty in sediment
passage rate in PySedSim by varying model parameters.

The Sedimentation Index (SI) is used to predict the percentage of sediment predicted to
pass through the reservoir (or 1 - TE), and is given by the following:

Wild et al. (2019)

SedSim User Manual

P a g e | 34

𝑆𝑆𝑆𝑆(𝑡𝑡) =

𝑅𝑅(𝑡𝑡)
𝑣𝑣(𝑡𝑡)

=

𝑆𝑆(𝑡𝑡)
𝑄𝑄𝑜𝑜 (𝑡𝑡)
𝑄𝑄𝑜𝑜 (𝑡𝑡)
𝐴𝐴(𝑡𝑡)

=

𝑆𝑆(𝑡𝑡)
𝑄𝑄𝑜𝑜 (𝑡𝑡)
𝑄𝑄𝑜𝑜 (𝑡𝑡)
𝑆𝑆(𝑡𝑡)
𝐿𝐿

=

𝑆𝑆(𝑡𝑡)
𝑄𝑄𝑜𝑜 (𝑡𝑡)
𝑄𝑄𝑜𝑜 (𝑡𝑡)
𝑆𝑆(𝑡𝑡)
𝐿𝐿

=

�

𝑆𝑆(𝑡𝑡) 2
�
𝑄𝑄𝑜𝑜 (𝑡𝑡)

𝐿𝐿

=

𝑆𝑆(𝑡𝑡)
𝑄𝑄𝑜𝑜 (𝑡𝑡)
𝑄𝑄𝑜𝑜 (𝑡𝑡)
𝐴𝐴(𝑡𝑡)

(2.9)

where R(t) is the residence time (s) of water in the reservoir during the time period t (one day);
S(t) is the water storage (m3) in the reservoir at the beginning of the time period; Q o (t) is the
reservoir release rate (m3/s) during the time period; A(t) is the cross-sectional area (m2)
through which the reservoir inflow is discharged during the time period; L is the reservoir
length (m) at full supply level.
The SI(t) is then used to determine the percentage of sediment passing, P(t), through
the sluiced reservoir by applying the following approximation to Churchill's original regression
curve:
P(t) = 800*(SI(t)/3.28)-0.2-12
TE = (100 - P(t))/100

(2.10)

The following information must be supplied by the user for every reservoir at which
sluicing will be simulated. Some user inputs are described below as Optional, meaning these
inputs are extra features that are not required to run a simulation. More details on these inputs
are provided in the discussion of the sediment management specifications worksheet
(“Sluicing”), where most of these inputs are required to be entered.
1. Beginning date of sluicing. Worksheet: “Sluicing”.
2. Sluicing starting criterion: minimum reservoir inflow rate (m3/s) (Optional). Worksheet:
“Sluicing”.
3. Sluicing duration. Worksheet: “Sluicing”.
4. Target drawdown water surface elevation (mamsl) or storage (m3). Worksheet: “Sluicing”.
5. Maximum sluicing drawdown rate (m/d) (Optional). Worksheet: “Sluicing”.
6. Maximum sluicing refill rate (m/d) (Optional). Worksheet: “Sluicing”.
7. Reservoir Length (m) at Full Supply Level (FSL). Worksheet: “Reservoir Specifications”.
8. Does hydropower production occur during sluicing? (Yes or No.) Worksheet: “Sluicing”.
9. Mid-level outlet elevation vs. discharge capacity table. Worksheet: “Outlet Capacity Data”.
This table will be used to limit the capacity to release water from the mid-level outlets during
Wild et al. (2019)

SedSim User Manual

P a g e | 35

sluicing based on reservoir water surface elevation. Note that this input is not optional, as the
model makes the assumption that mid-level outlets are required to conduct sluicing.
2.4.5. Simulating Density Current Venting
The following are some important assumptions SedSim makes in simulating density
current venting. First, there is no user-specified start date for venting. Instead, it can occur on
any day, at any time of year. It begins whenever the combination of inflow rate and sediment
concentration result in a theoretical venting efficiency that exceeds the use-specified minimum
venting efficiency (defined below). Likewise, there is no user-specified end date for venting.
Instead, venting ends whenever the combination of inflow rate and sediment concentration
result in a theoretical venting efficiency that is less than the user-specified minimum venting
efficiency. (Venting efficiency, which is defined by the user as a function of inflow conditions,
describes the percentage of sediment concentration flowing into a reservoir that can be
released by the low-level outlets at the dam during density current venting).
Next, if density current venting is initiated in a time period (if the venting efficiency in
time period t exceed the user-specified minimum venting efficiency), the reservoir maintains its
originally specified rule curve, but four user-specified constraints are imposed on the reservoir
during this time that may alter the reservoir’s targets from their pre-specified course. These
constraints are described here. The latter three require user input, as described at the end of
this section.
1. The target outflow rate (m3/s) for the low-level outlets Q vent (t), is set equal to the
reservoir inflow, Q in (t) (i.e., Q vent (t) = Q in (t)). To limit the capacity of the low-level outlets
to release the inflow during density current venting, the user should establish an
elevation vs. discharge capacity table for the low-level outlet used for venting.
2. The user can impose a minimum daily power production requirement during venting.
Since the inflow rate will be released through the low-level outlets during venting, it is
important to specify some minimum power production requirement if one exists,
because the reservoir may otherwise not produce any power during venting, depending
on the current water surface elevation (or storage) target on the original guide curve.
Note that any water released from the hydropower outlets during venting will result in
some drawdown, given that the inflow is likely being released during this time.
3. Additional water may be released from the reservoir through both the hydropower and
mid-level outlets, assuming capacity exists, if the user specifies a maximum downstream
sediment concentration (mg/l) during venting. The hydropower outlets will not be used
to release additional water if their release is diverted away from the reservoir’s
downstream channel.

Wild et al. (2019)

SedSim User Manual

P a g e | 36

4. Density current venting (and all associated density current releases) will be stopped if
the reservoir’s water surface elevation drops below a user specified minimum water
surface elevation.
Additionally, venting efficiency (%) is assumed to be a function of inflow rate (m3/s) and
concentration (mg/l). This means that in one simulation time step (each day) a density current
may have a particular set of concentration and potential venting efficiency, whereas on the
very next day these two parameters may change due to a change in the inflow conditions. This
would be fine if a density current entered and exited the reservoir within each day. However, a
density current (1) needs time to proceed from the inflow point to the low-level outlets, and (2)
will dissipate if the sediment supply driving the current stops (or drops below some threshold).
SedSim ignores these issues for the sake of simplicity.
Next, note that density current venting, flushing and sluicing are assumed to be exclusive
activities, in that they cannot be conducted at the same time. Multiple sediment management
techniques can be simulated in the same reservoir at different times. However, flushing,
sluicing and density current venting cannot be simulated concurrently. Any management
technique being simulated will be allowed to finish before a new technique is begun. For
example, if sluicing is being simulated at a particular reservoir and flushing is meanwhile
scheduled to occur, the start of flushing will be delayed until sluicing is completed. If the user
schedules two or more sediment management techniques to start on the same date, priority is
given first to flushing, then to sluicing, and finally to density current venting.
A few additional simplifying assumptions are important to mention. For example,
density current venting and sluicing cannot be conducted at the same time. When sluicing is
underway, density current venting cannot be activated. Also, no muddy lakes exist at the
bottom of the reservoir, from which sediment can be released. Next, the bathymetry of the
reservoir's submerged channel does not change over time. Infilling of channel running along the
thalweg could lead to decreased effectiveness of current conveyance over time. Next, SedSim
ignores the multi-dimensional spatial variability in concentration and velocity of currents.
Finally, SedSim ignores the within-day temporal variability of the inflow and concentration that
create the currents.
The following information must be supplied by the user for every reservoir in which
density current venting will be simulated. Some user inputs are described below as Optional,
meaning these inputs are extra features that are not required to run a simulation. More details
on these inputs are provided in the discussion of the sediment management specifications
worksheet (“Density Current Venting”), where most of these inputs are required to be entered.
1. Minimum venting efficiency (%).Worksheet: “Density Current Venting”. This represents the
lowest acceptable percentage of sediment removal that must occur for density current venting
to be an attractive option (e.g., 35%). See below for additional comments regarding
Wild et al. (2019)

SedSim User Manual

P a g e | 37

determination of venting efficiency in SedSim, against which this minimum venting efficiency is
compared to determine whether or not to conduct venting in each time step.
2. Minimum reservoir water surface elevation (mamsl) during density current venting
(Optional). Worksheet: “Density Current Venting”.
3. Reservoir length (km). Worksheet: “Density Current Venting”. Used to determine a default
minimum venting efficiency value for the user if the user does not specify one.
4. Mid-level outlet elevation vs. discharge capacity table (Optional). Worksheet: “Outlet
Capacity Data”. This input is used to limit the capacity to release water from the mid-level
outlets during density current venting. Mid-level outlets can be used to release clear water
downstream to reduce the concentration of density current releases.
5. Low-level outlet elevation vs. discharge capacity table. Worksheet: “Outlet Capacity Data”.
This table will be used to limit the capacity to release water from the low-level outlets during
venting.
6. Maximum concentration (mg/l) of sediment released from reservoir during density current
venting (Optional). Worksheet: “Density Current Venting”.
7. Whether to continue venting if maximum specified release concentration is exceeded, or to
stop venting. Worksheet: “Density Current Venting”.
8. Minimum power requirement during density current venting (MW) (Optional). Worksheet:
“Density Current Venting”.
9. Reservoir bottom width (m). This input is used to determine venting efficiency during each
density current venting event. Worksheet: “Density Current Venting”.
10. Reservoir bed slope (m/m). This input is used to determine venting efficiency during each
density current venting event. Worksheet: “Density Current Venting”.
Some additional comments are necessary regarding the determination of venting efficiency in
SedSim. As described below, the SedSim internally employs a methodology proposed by Morris
and Fan (1998) that determines the efficiency with which a current can be vented given a
reservoir inflow rate (m3/s) and inflow concentration (mg/l) on each simulation day. In each
time step, if the inflow rate and concentration combine to produce a venting efficiency lower
than the user-specified minimum venting efficiency, then density current venting will not occur
on that simulation day, and the sediment will settle. Aside from assumptions about flow rate
and concentration of inflow, input data/assumptions required to develop this venting efficiency
table include temperature (for density calculations), grain size distribution, river slope, and
Wild et al. (2019)

SedSim User Manual

P a g e | 38

velocity vs. grain size curve. These internal assumptions cannot be modified by the user. SedSim
determines the venting efficiency iteratively as follows
1. Determine the velocity of the turbidity current in the ith iteration.
𝑣𝑣𝑖𝑖 = �

8 𝜌𝜌𝑤𝑤𝑤𝑤 (𝑡𝑡)−𝜌𝜌𝑤𝑤

𝑓𝑓

𝜌𝜌𝑤𝑤

𝑔𝑔

𝑄𝑄𝑖𝑖𝑖𝑖 (𝑡𝑡)
𝐵𝐵

1/3

(2.11)

𝑆𝑆�

where v is the velocity of the density current (m/s); f is a coefficient that represents the total
interfacial frictional effects between the density current and the both the river bed and
overlying clear water layer; g is the gravitational constant (9.81 m/s2); ρ ws (t) is the density of
water at 20° C with a suspended solids concentration equal to the concentration of the
reservoir’s inflow during time period t, C in (t); ρ w is the density of pure water at 20 °C (zero
suspended solids concentration); B is the representative bottom width of the reservoir; S is the
representative bed slope of the reservoir; and Q in (t) is the reservoir inflow rate.
Note that the density of water, ρ ws , with suspended sediment concentration C (ppm), was
assumed to be given by the following relationship:
ρ ws = 0.9982 + 0.0006C

(2.12)

Eq. (2.12) was determined using data from Washburn (1928), as appears in Table 3.1 below.
Table 3.1. Density of water and sediment mixtures as a function of temperature (°C) and suspended solids concentration (g/L).
Original Source: Washburn 1928. Table taken from Morris and Fan (1997).

2. Using the velocity, v i , calculated in Step 1 of iteration i, determine the maximum grain size
that can be transported by v i using Figure 3.7 below. This figure represents the relationship
between turbidity current velocity and the grain size that can be maintained in suspension (Fan
1996).

Wild et al. (2019)

SedSim User Manual

P a g e | 39

Figure 3.7. Relationship between density current velocity (m/s) and the sediment grain size (mm) that can be
maintained in suspension. Figure borrowed from Fan (1986).

The fitted curve appearing in Figure 3.7 representing velocity, v, as a function of 90th percentile
particle size, d 90 , can be approximated by the following relationship:
d 90 = -0.0074(v2) + 0.0369(v) + 0.0007

(2.13)

3. Remove all grain sizes from suspension that are too large (> d 90 ) to be transported by the
velocity of the current, v, in iteration i.
To complete this step, the following particle size distribution is assumed (Figure 3.8).

Wild et al. (2019)

SedSim User Manual

P a g e | 40

100
90
80

Percentage finer

70
60
50
40
30
20
10
0
0.00

Clay
0.01

Silt

Gravel

Sand
0.10

1.00

10.00

Particle size, mm
Figure 3.8. Particle size distribution used to compute percent of sediment in suspension that does not settle in the
reservoir (is fine enough to be transported by the velocity of the density current).

The following relationship can be used to approximate the data points in Figure 3.8.
% Finer = 15.226ln(d 90 ) + 95.839

(2.14)

4. Apply the % finer value determined in iteration step 3 to the concentration of the inflow,
and repeat steps 1 - 4 until convergence of the % finer value. The % finer value represents the
venting efficiency (%). It is the portion of the inflowing suspended sediment load that can
remain in suspension, which is the percentage that can be vented and will not settle. As this
%finer value affects water density and thus the velocity of the current, which affects the
particle size that can be transported, this iterative process is required.
Once convergence of the % finer value is achieved, the venting efficiency is given by the final %
finer value.
If the user chooses to allow density current venting at a particular reservoir, and if the venting
efficiency in this table corresponding to the current inflow and concentration exceeds the userspecified minimum venting efficiency (see below), then a percentage of the inflowing
concentration equal to the corresponding venting efficiency will be released from the low-level

Wild et al. (2019)

SedSim User Manual

P a g e | 41

gates of the reservoir. The remaining percentage (100-venting efficiency) will settle in the
reservoir.
Consider the following example of how the venting efficiency computed in SedSim is used:
Average daily flow into reservoir from upstream reach, Qin: 100 m3/s
Average daily concentration into reservoir from upstream reach, Cin: 20,000 mg/l
Venting efficiency = f(Qin, Cin) = 30%
Venting flow, Qout = Qin = 100 m3/s
Mass released from venting in one day, assuming user defines 30% efficiency as acceptable:
0.3*100 (m3/s)*20,000 (mg/L)*1000 (L/m3)*86400 (s/day)*10-9 (kg/mg) = 51,840 kg.
In the example above, the sediment concentration released in the vented flow, Q vent , is equal to
the product of the sediment concentration in the flow entering the reservoir (from the
upstream channel) and the venting efficiency. The venting efficiency is equal to the percentage
of sediment remaining in suspension by the time the current reaches the dam
3.4.6. Other sediment removal methods

To account for the removal of sediment from a reservoir in SedSim, the following
information must be supplied in the “General Sediment Removal” Worksheet by the user for
every reservoir in which a sediment removal practice will be simulated:
1. Calendar date on which to begin sediment removal. Worksheet: “General Sediment
Removal”.
2. Removal duration (days). Worksheet: “General Sediment Removal”.
3. Amount of sediment to be removed (tons). Worksheet: “General Sediment Removal”.
4. The user-defined name of the system element into which the removed sediment will be
distributed. Worksheet: “General Sediment Removal”.
5. The fraction of sediment that is removed from the active storage zone (the remainder of
sediment is assumed to be removed from the dead storage zone). Worksheet: “General
Sediment Removal”.
For these techniques, the SedSim model approach is very simple. On the date on which
the user specifies sediment removal is to be initiated, the model attempts to remove the userspecified amount of sediment volume from the reservoir, while first making sure that no more
sediment than is initially available can be removed from the reservoir. The volume removed
may be distributed over a user-specified number of days, and is transferred to the settled
Wild et al. (2019)

SedSim User Manual

P a g e | 42

sediment reserve at a location in the modeled system that the user must specify. If no
destination location for the sediment is specified, the sediment is assumed to permanently
leave the modeled system. The user can optionally specify the fraction of sediment volume
removed that contributes to recovery of active storage, while the remainder of storage is
assumed to be recovered in the dead storage zone. Hydropower production and reservoir
storages/releases are not altered during the removal process.

3.5. Dam and Reservoir Design Features
To estimate hydropower production and release capacity of reservoir outlets, the model
requires that Elevation-Volume-Area information be provided for each reservoir, so the model
can determine the elevation (mamsl) corresponding to the storage volume (m3) at the
beginning and end of every simulated day. The original Elevation-Volume-Area relationship
provided by the user is modified over time as either (1) sedimentation reduces the water
volume and surface area at each water surface elevation, or (2) sediment management
practices increases the water volume available at each water surface elevation.
For each reservoir, the SedSim model requires the user to select a reservoir type from
four different options, shown in Figure 3.9. A reservoir can either have, or not have, the
capability to produce hydropower. Within the two hydropower categories, a reservoir can
either only have the capability to release water downstream, or can have the added capability
to divert water to another location in addition to downstream. At diversion reservoirs, all
diverted water and sediment is immediately transferred to the specified location without
routing (i.e., there is no routing time lag in the transfer of diverted water and sediment from
one location to another).

Figure 3.9. Diagram of the four different reservoir types a user can simulate using the SedSim model.

These reservoir type distinctions are important for the user to establish so the model
can determine appropriate release outlets for each reservoir type, as well as the priority of
releases from those outlets. The outlet types established for each reservoir type are listed in
Wild et al. (2019)

SedSim User Manual

P a g e | 43

Figure 3.9., but each is described in more detail below. First, it will be important to discuss
some important general points about specifying outlets in SedSim.
First, it is suggested that for each reservoir to be simulated, the user specify the same
type of outlets listed in Figure 3.9 for the reservoir type being simulated, though this is not a
requirement. For example, if a reservoir with no hydropower or diversion capabilities is to be
simulated, a controlled and overflow outlet should be specified by the user. However, if the
user only includes the overflow outlet (i.e., omits the controlled outlet), only the overflow
outlet will be capable of releasing flow. If the user specified a hydropower outlet for this
reservoir type, it would not be used, because this reservoir type is not capable of producing
power.
When the user selects a reservoir type (in the "Reservoir Specifications" worksheet), the
model must be supplied with information for each of its outlets. It is very important that the
user have a maximum of only one outlet of each type that is appropriate for the reservoir
selection (again, see Figure 3.9). For example, even if the reservoir in reality has 5 spillway
gates, the individual capacities of these gates should be combined into one larger gate when
reservoir data are provided by the user. Additionally, the maximum capacity of each outlet to
release flow is dependent only on the water surface elevation in the reservoir. The user must
establish the maximum discharge capacity of each outlet for elevations at which the capacity is
different. Finally, all outlets are assumed to be controllable. For example, even if the water
surface elevation in a reservoir is at the level of the spillway outlet, the spillway is not assumed
to release any water unless the storage or elevation target for the reservoir dictates a release is
needed. So, while possible it is not likely water will be stored above the spillway elevation since
the target elevation or volume would not normally be set greater than this limit.
Note that the discussion below references specific sediment management techniques, detailed
descriptions of which are presented later in this chapter.
Reservoir outlets include:
1. Controlled Outlet. This outlet type, which does not result in any hydropower generation, is
used to ensure that water is released into the downstream channel. For example, diversion
reservoirs can divert a significant volume of water away from the downstream channel, but
this outlet is located at a low point in the reservoir to increase the likelihood that some
water can be released downstream.
2. Hydropower Outlet. This outlet type directly supplies the turbines with flow to generate
hydropower. Water released through this outlet type is discharged into the downstream
channel after passing through the turbines. Importantly, the model does not limit discharge
through the hydropower outlet when the provided capacity (MW) of the powerhouse is
exceeded. That is, the user-supplied capacity vs. elevation curve for the hydropower outlet
applies to the ability to supply water to the turbines, rather than a power production
Wild et al. (2019)

SedSim User Manual

P a g e | 44

capacity-based discharge limit. The model does, however, prevent power production in
excess of the user-provided maximum power production capacity of the plant. Any water
being released through this outlet that does not generate hydropower is also classified as
spilled flow in the model output.
3. Diverted Outlet. This outlet type, which does not result in any hydropower generation,
diverts water away from reservoir's downstream channel. The location to which water is
diverted can be within the basin being modeled, or outside of the modeled system.
4. Hydropower/Diversion Outlet. This outlet type directly supplies the turbines with flow to
generate hydropower. This outlet type is only different from the hydropower outlet type
described above in that the water is diverted away from the downstream channel after
producing power.
5. Spillway (overflow) Outlet. This outlet type, which does not result in any hydropower
generation, is a spillway. Thus, this outlet is assumed to be located at the top of the active
storage zone, which is the zone of the reservoir that is operated for hydropower
production. These outlets are used to drain any storage space above the active storage zone
(i.e., the flood storage zone). Assuming the reservoir’s storage or elevation targets are
below the spillway outlet, any water stored above the spillway outlet is spilled into the
downstream channel, assuming the outlet has the capacity to release the water.
6. Low level Outlet. This outlet type, which does not result in any hydropower generation, is
used to release water and sediment into the downstream channel for sediment
management purposes. This outlet type can be used to release water during flushing,
density current venting and sluicing. For flushing and sluicing, the outlet should be properly
sized to enable drawdown (either complete or partial) of the reservoir, as well as properly
sized to release reservoir inflows during the flushing and/or sluicing period. For effective
flushing, the invert elevation of this outlet should be close to the original river bed elevation
at the reservoir site. (Important: The first elevation entry in the elevation-capacity table for
this outlet will be assumed to be the original river bed elevation for flushing calculations).
This outlet is the only operable outlet during flushing once the water levels drop below the
operating levels of other outlets, due to the low water surface elevation maintained during
flushing. For density current venting, the low-level outlet should be sized to discharge the
reservoir inflow at normal operating water surface elevation during inflow events that
produce concentrations significant enough to warrant density current venting.
7. Mid Level Outlet. This outlet type, which does not result in any hydropower generation, can
be used to release water during sluicing (a form of sediment management that includes
drawing the reservoir level down to the mid-level outlets and releasing the reservoir inflow,
usually during high inflow season). The mid-level outlets are opened when all criteria are
satisfied to initiate drawdown for sluicing, and are generally kept open throughout sluicing.
The goal of sluicing is to release the reservoir inflow, so in the event that hydropower
Wild et al. (2019)

SedSim User Manual

P a g e | 45

outlets exist and have the capacity to release the inflow during sluicing, the mid-level
outlets would not be used.
All outlets are assumed to be functional regardless of the amount of sediment
contained in the reservoir. That is, even if sedimentation significantly reduces a reservoir’s
water storage capacity, all of the outlets are assumed to remain functional for water release
throughout simulation.
In addition to selecting a reservoir type, the user must define an operating policy for
every reservoir (i.e., how much water to release every day in m3/s). The model allows
operating policies based on (1) storage volume targets or (2) storage elevation targets. Both
options require that the user pre-establish a time series of targets. (Future versions of the
model may allow release decisions in response to the reservoir's storage or elevation state,
rather than pre-established targets).
Regardless of which target option is selected, the model implements a similar approach
in operating each reservoir, which is to determine how much water must be released, if any, to
meet the target, given the initial reservoir storage, potential evaporation, constraints on the
release capacity of the reservoir's user-defined outlets (discussed above), and minimum
environmental flow constraints. Once the model has determined how much water must be
released during the simulation period to meet the specified target, discharge is distributed
among the outlets using a set of priorities that depend on the reservoir type. For reservoirs that
can only release water downstream (regardless of whether or not the reservoir produces
hydropower), the primary outlet (controlled or hydropower) discharges as much of the target
flow release as possible, and the overflow outlet only receives the remainder of flow that could
not be distributed to other outlets due to release capacity constraints at those outlets. If
drawdown flushing is being conducted, any flow that could not be released by other outlets is
released by the low-level outlet. For diversion reservoirs, which can divert water away from the
downstream channel, the approach is only slightly different. In this case, water is first allocated
to the controlled outlet (the primary outlet responsible for releasing water downstream) to
meet any user-established minimum environmental flow constraints, after which the
hydropower/diversion outlet, overflow outlet and low-level outlet are allocated flow (assuming
they have capacity), in that order.
2. Storage Targets. If this option is selected, a water storage target (m3) must be established
for the end of each simulation day. This input must be provided in the form of a time series.
If sediment accumulates in a reservoir and a "Storage Target" policy is selected, the model
will only store the water required to meet the storage target, without regard to the impact
of sediment on the elevation of the water. For example, if one specifies an operating policy
that assigns an identical target value for the reservoir for the duration of simulation, the
model will attempt to meet this target, but the water surface elevation to which the
constant storage target corresponds will continue to rise as simulation proceeds. (The
Wild et al. (2019)

SedSim User Manual

P a g e | 46

second reservoir operations option, presented below, offers an approach that accounts for
this sediment accumulation issue).
Elevation Targets. If this option is selected, a water surface elevation target (m3) must be
established for the end of each simulation day. This input must be provided in the form of a
time series. If sediment accumulation in the reservoir is negligible in comparison to the storage
capacity, then this policy option will result in the same policy one would establish just using
storage targets, because an elevation corresponding to every water storage value can be
determined from the Elevation-Volume data. However, if sediment accumulation in the
reservoir is significant, this option will allow specified elevations to be maintained in the
reservoir over time, which may require that less water be maintained in storage as the
simulation proceeds due to sediment accumulation in the reservoir's storage space.

Wild et al. (2019)

SedSim User Manual

P a g e | 47

4. Overview of Model File Structure
The SedSim model consists of three different types of Excel workbooks (as summarized in Figure

4.1 below):

1. A data input file (e.g., “SedSim_Input.xlsx”),
2. The main model file (e.g., “SedSim_Model.xlsm”), and
3. Output file(s) (e.g., “SedSim_Output.xlsx”).
Users can name these files as they wish. For example, the input file for the single reservoir
simulation used in this manual could be “Sambor_Input.xlsx” or “INPUT_SAMBOR.xlsx”.

Figure 4.1. Schematic of data flow in the SedSim model.

From this point forward, excel files (e.g., “SedSim_Input.xlsx”) will be referred to as
workbooks, whereas the tabs within the workbooks will be referred to as worksheets. While the
three files in the list above are given names (e.g., "SedSim_Input.xlsx"), the user will control the
naming of all workbooks in their application of the SedSim model. Referring throughout this
manual to the files by the names in the list above (e.g., "SedSim_Input.xlsx") is only done for
convenience.
For the model to run properly, only the main model file (“SedSim_Model.xlsm”) must be
open. The location of the files does not affect the ability of the model to execute properly as
long as the file locations are all properly specified in the “SedSim_Model.xlsm” file. However,
the time required to load the data, run the model and save the results may be faster if the files
are located on the computer’s hard drive, rather than on an external hard drive or flash drive.
A description of the three primary file types (main file, input file and output files) is provided in
Chapters 5, 6 and 7, respectively. The discussion will focus on the purpose of each workbook
and the actions a user must take to properly prepare each worksheet within each workbook.

Wild et al. (2019)

SedSim User Manual

P a g e | 48

5. Main Model File Description
The main model file ("SedSim_Model.xlsm") has two primary purposes:
(1) It contains the Excel VBA code needed for simulation and permits its execution by
clicking on the “Run Model’ button, and
(2) It contains an interface into which users are required to specify the names and directory
locations of input and output files, as well as which output files the user wants the
model to create. See Figure 2.2.
While this file is referred to here as "SedSim_Model.xlsm", the user can specify any name for
this file (e.g., "ModelFile.xlsm"), as long as the ".xlsm" extension is maintained. To run the
model, after providing all required inputs click on the "Run SedSim Model" button located
within the worksheet titled "Run Model". (Note: the name of the main worksheet within this
workbook must remain "Run Model", or the model will not execute). During the execution of
the simulation run, the model will attempt to automatically close the “SedSim_Input.xlsx” file
because the data will no longer be needed.
If your model input and output files are particularly large, you may wish to increase the “Auto
save” time increment under File  Excel Options. This will avoid long auto-save delays. You can
customize the auto-save time to a particular workbook so that standard auto-save settings are
still maintained for other workbooks not related to the sediment model.

5.1. Specifying Files Names and Locations
a) Input File name/location. Specify the location and file name of the input file on the
computer. For example, the following would be a valid file name and location:
C:\Users\YourName\My Documents\SedSim_Input.xlsx. The model will use this information
to locate, open and import input data from the specified file. The file can be located
anywhere on the computer, but the file name must have a .xlsx or .xls extension. (If this file
is to be located in the same directory as the main model workbook, only a file name must
be specified, without any file routing information. If no file routing or name are specified,
the model will assume the input file is located in the same directory as the main model
workbook, and is named "Input.xlsx").
b) Time series output file name/location. Specify the desired location and file name of the
time series output file that will be automatically created and saved by the model, assuming
the user chooses to have this file created (see the "File creation specifications" options
below). For example, the following would be a valid file name and location:
C:\Users\YourName\My Documents\Time_Series_Output.xlsx. The model will use this
information to create and save the specified file, and export data to this file. The file can be
Wild et al. (2019)

SedSim User Manual

P a g e | 49

located anywhere on the computer, but the file name must have a .xlsx or .xls extension
and must not have the same name as pre-existing files in the specified location or the preexisting files will be replaced. (If this file is to be located in the same directory as the main
model workbook, only a file name must be specified, without any file routing information. If
no file routing or name are specified, the model will assume the time series output file
should be located in the same directory as the main model workbook, and will be named "
Time_Series_Output.xlsx".).
c) Statistical output file name/location. Specify the desired location and file name of the
statistical summary output file that will be automatically created and saved by the model,
assuming the user chooses to have this file created (see the "File creation specifications"
options below). For example, the following would be a valid file name and location:
C:\Users\YourName\My Documents\Statistics_Output.xlsx. The model will use this
information to create and save the specified file, and export data to this file. The file can be
located anywhere on the computer, but the file name must have a .xlsx or .xls extension
and must not have the same name as pre-existing files in the specified location or the preexisting files will be replaced. (If this file is to be located in the same directory as the main
model workbook, only a file name must be specified, without any file routing information. If
no file routing or name are specified, the model will assume the statistics output file should
be located in the same directory as the main model workbook, and will be named
"Statistics_Output.xlsx".).
d) File creation specifications. Select one of the following options from the drop-down menu:
i.
ii.
iii.

Create only Time Series Output File
Create only Statistical Output File
Create Time Series File and Statistical Output File

These four options allow the user to specify as many, or as few, output data files as wanted.
These options are available because the output files, depending on their size, can take
significant time and storage space to be created, populated with data, and saved.

Wild et al. (2019)

SedSim User Manual

P a g e | 50

6. Input File Description
6.1. Overview of Input Workbook
To illustrate how SedSim works a simple single reservoir problem will be simulated. The
simplified example will be based on the proposed Sambor reservoir in the Mekong River.
The input workbook file, “INPUT_SAMBOR.xlsx,” to the SedSim model consists of a
variety of separate worksheets, each responsible for storing a different type of information.
This file should be updated, saved and closed before the simulation is conducted. The
simulation will run if the file is open at the start of simulation, but this may make the simulation
proceed more slowly (or fail to execute) if your RAM is sufficiently low given the duration of
simulation and number of system elements being simulated.
Some of the worksheets in this workbook are required to contain time series. For such
cases, additional discussion about formatting of the input data is provided below.
In time series worksheets, each column represents a unique location in the modeled
system, whereas each row represents a date. Time series should begin in the second column on
the second row, as the first column should just contain dates (in DD/MM/YYYY format).
The name of the location for which each time series applies must be listed in the first
row of the associated column. The name of the location must contain the same exact name as
is listed in the "Network Connectivity" worksheet for that reservoir. For example, if a reservoir
in the system is defined in the network connectivity matrix as "LowerSeSan3", then the string
typed into the third row for a particular column must contain the letters "LowerSeSan3" in that
exact order. There are no case restrictions in this regard; that is, "LoWeRsEsAn3" would also
suffice, as would "LoWeRsEsAn3-POOL", because the correct name is still contained within the
string, even though the word POOL is added. However, "LoWeR sE sAn 3-POOL" would not
suffice as a time series column heading, because spaces are placed in locations in which they
did not appear in the name of the reservoir in the "Network connectivity" worksheet.
For each variable for which you are providing time series input, there is no limit to the
number of contiguous columns or rows for which data can be provided. The model will search
for the time series data for each element by name (located at the top of each time series) for
the dates contained within the simulation horizon, and will thus skip data that do not pertain to
elements and dates being modeled in the current simulation. This is especially useful for cases
in which simulations corresponding to different levels of development (e.g., different numbers
of reservoirs) or different time horizons in the basin are of interest. This is because one large
data set can be stored in the input file corresponding to the maximum extent of development,
and simulations with fewer reservoirs will not then require new, smaller input files to be
created. This being said, changes to time series files can still be required when a new system
Wild et al. (2019)

SedSim User Manual

P a g e | 51

configuration or time horizon is simulated. For example, do not store incremental flow data in
the "Incremental Flow" worksheet for junctions that exist in the modeled system but at which
incremental flows do not occur in the current simulation.
The date on which each time series value occurs must appear in the first column on the
same row on which the associated time series value occurs. If no data exist for a particular
element, then no data are required to be entered for that element. For example, if one
worksheet corresponds to reservoir evaporation, then if the system has 41 reservoirs but
evaporation time series for only 40 of them are available, then the worksheet only needs 40
time series columns. If you omit the time series input for a particular system element (or if you
incorrectly spell the name of the element in the column heading), the model will assume all
zero values for the omitted element. However, the model will automatically generate a warning
message (before simulation proceeds) that indicates you have omitted data, and for what
element. (No such error messages will be generated for any incremental flow data supplied by
the user in the "Incremental Flow" worksheet). The time series columns can be placed in any
order in the worksheet, as long as the reservoir's name is correctly spelled in the third row of
the time series column, as explained above.
No row gaps or column gaps in time series data are permitted. That is, the SedSim
model assumes that data are provided for every day of simulation, and that time series are
stacked as contiguous columns. For example, there should not be a sequence of two numbers
in the time series that skips a date, nor should there be any blank rows or columns.
Note that some of the conditions discussed above for time series worksheets also hold
for worksheets that do not require time series input. For example, no gaps between columns or
rows should exist in any of the input data worksheets. Additionally, data for more than one
element can be provided, as the model will locate and import only data pertaining to elements
that will be included in the simulation.
The following worksheets are required at a minimum to conduct a simulation:
A.
B.
C.
D.
E.
F.
G.
H.
I.
J.

Simulation Specifications
Network Connectivity
Sediment Loads
E-V-A-S
Storage Volume Target OR Storage Volume Elevation Target (depending on preferences
established in "Reservoir Specifications" worksheet)
Incremental Flows
Evaporation Data
Environmental Flow Data
Reach Specifications
Reservoir Specifications

Wild et al. (2019)

SedSim User Manual

P a g e | 52

K. Flushing (only if Flushing will be simulated as specified in "Reservoir Specifications"
worksheet)
L. Sluicing (only if Sluicing will be simulated as specified in "Reservoir Specifications"
worksheet)
M. Density Current Venting (only if Density Current Venting will be simulated as specified
in "Reservoir Specifications" worksheet)
N. Bypassing (only if Bypassing will be simulated as specified in "Reservoir Specifications"
worksheet)
O. General Sediment Removal (only if General Sediment Removal will be simulated as
specified in "Reservoir Specifications" worksheet)
P. Outlet Capacity Data

6.2. Description of Each Worksheet
6.2.1. System Schematic and Meta File

In this worksheet, users are encouraged to place a map (or schematic) of the system
being modeled. Any figures or data placed in this worksheet will not be used in the execution of
the sediment model. Meta data pertaining to the model inputs and assumptions can also be
included in this worksheet. An example of this worksheet is shown below.

Figure 6.1. Illustration of a Map and Meta data on the “System Schematic” worksheet page of Input file
“INPUT_SAMBOR.xlsx”

Wild et al. (2019)

SedSim User Manual

P a g e | 53

This worksheet is accessed by clicking on the purple “System Schematic” worksheet tab
at the bottom of the worksheet. Nothing in this worksheet is used during the simulation.
6.2.2. Simulation Parameters and Specifications

This worksheet is shown below. It is accessed by clicking on the "Simulation Specifications"
worksheet tab at the bottom of the input data workbook.

Figure 6.2. The “Simulation Specifications” worksheet of the input file for the SedSim model.

In this worksheet, users are required to enter a variety of assumptions for the model to
use during the simulation run. Any cell highlighted in green requires input from the user,
whereas any cell highlighted in red only requires user input if the model feature pertaining to
the cell will be used. Do not leave blank any cells that are highlighted green. Also, do not
change the location of any data descriptions within the workbook, as the model assumes that
all parameters will be located in particular, pre-established cells.
6.2.2.1. System Properties Category

The following information must be specified by the user in this section:
a) Regulated or unregulated simulation. Select either "Regulated" or "Unregulated" from the
drop-down menu. The regulated system is defined as one in which reservoirs and diversions
are present, with the number and connectivity of reservoirs and diversions, and the system
flows that have been affected by these structures, are specified in the input data. The
unregulated simulation simply assumes that no reservoirs or diversions exist, and therefore
Wild et al. (2019)

SedSim User Manual

P a g e | 54

all locations at which reservoirs are expected to exist (or currently exist) become reaches.
This option is not necessarily the same as simulating the "current" state of an already
regulated basin.
b) Simulation start and end dates. Specify the simulation start date and end dates in
MM/DD/YYYY format. In the model input files, several time series inputs for each system
location (and several variables) may be required, so users should be careful to specify start
and end dates of simulation for which a corresponding value exists for every variable of
input time series data.
6.2.2.2. Sediment-related Assumptions Category

a) Incremental sediment loads: calibration preferences for determining coefficient c (in
sediment rating curve equation cQd) for each incremental inflow location i. Select one of
the following options from the drop-down menu:
i.
ii.
iii.
iv.

Calibrate a coefficient for each incremental inflow location.
Use coefficients calibrated in most recent simulation.
Specify one coefficient for all incremental flow locations.
Specify a separate coefficient and exponent for each incremental inflow location.

A few comments about these options are now important to make, but note that detailed
discussion of the calibration process is available in Chapter 2 under the SedSim Model
Development section.
Note the units of sediment concentration resulting from this equation are kg/m3.
Option 1 results in a series of calibrations to determine an appropriate c value for every
location in the system at which an incremental flow (and therefore sediment load occurs). This
requires that the model calls Microsoft Excel's Solver to calibrate a parameter for each location.
The model will store and save the calibrated parameters in the output files.
Option 2 allows the user to save time and CPU usage by running a simulation using
incremental sediment load parameters that were determined in a previous calibration run. If
you have selected Option 1 in a previous simulation run, the model will store and save the
calibrated parameters in the input data file used to supply data for that simulation run.
Selecting Option 2 will mean the model imports those most recently calibrated incremental
sediment load parameters stored in the current input data file.
Option 3 allows the user to specify one set of two parameters (the same coefficient and
exponent to be used at all incremental inflow locations for sediment load generation). This is a
useful option for studies with very limited data availability. For example, there may not be
Wild et al. (2019)

SedSim User Manual

P a g e | 55

enough evidence that having different parameters at different locations in the system is
reasonable. The two values should be specified in the "Simulation_Specifications" worksheet in
the two cells highlighted in red, as described below. The two cells are highlighted in red, rather
than green, because these specifications are only required if Option 3 is selected.
Option 4 allows the user to specify a different set of two parameters (coefficient and
exponent) to be used for sediment load generation at each incremental inflow location. As the
model will not perform a calibration to determine these values, the values must be supplied by
the user in the "Annual_Sed_Loads" worksheet. The "Annual_Sed_Loads" worksheet stores
mean annual sediment loads for each incremental inflow location. Thus, additional columns are
available for also storing coefficients and exponents to be used for generating incremental
sediment loads.
Note that, depending on the watershed in which this model is applied, you may wish to
establish the parameters ‘c’ and ‘d’ so that proportionally more sediment is transported during
higher discharge events, as is often observed in practice (Walling, 2009).
b) Sediment discharge from reaches (channels): calibration preferences for determining
coefficient a (in sediment rating curve equation aQb). Select one of the following options
from the drop-down menu:
i.
ii.
iii.
iv.
v.

Calibrate a coefficient for each reach.
Use coefficients calibrated in most recent simulation.
Specify one coefficient for all reaches.
Specify a separate coefficient and exponent for each reach.
Sediment mass out (kg) = Sediment mass in (kg).

A few comments about these options are now important to make, but note that
detailed discussion of the calibration process is available in Chapter 2 under the SedSim Model
Development section.
Note the units of sediment concentration resulting from this equation are kg/m3.
Option 1 results in a series of calibration to determine an appropriate a value for the
sediment carrying capacity function for every reach (channel) in the system. This requires that
the model calls Microsoft Excel's Solver to calibrate a parameter for each reach, which can take
a significant amount of time depending on the number of locations and simulation duration.
The model will store and save the calibrated parameters in the input data file used to supply
data for the simulation run
Option 2 allows the user to save time and CPU usage by running a simulation using
reach sediment carrying capacity parameters that were determined in a previous calibration
run. If you have selected Option 1 in a previous simulation run, the model will store and save
Wild et al. (2019)

SedSim User Manual

P a g e | 56

the calibrated parameters in the input data file used to supply data for that simulation run.
Selecting Option 2 will mean the model imports those most recently calibrated reach carrying
capacity parameters stored in the current input data file.
Option 3 allows the user to specify one set of two parameters (coefficient and
exponent) to be used in the sediment carrying capacity function for all reaches in the system.
These two values should be specified in the "Simulation_Specifications" worksheet in the two
cells highlighted in red, as shown in Figure 6.2. The two cells are highlighted in red because
these specifications are only required if Option 3 is selected.
Option 4 allows the user to specify a different set of two parameters (coefficient and
exponent) to be used in the sediment carrying capacity function for each reach. As the model
will not perform a calibration to determine these values, the values must be supplied by the
user in the "Reach_Data" worksheet.
Option 5 allows the user to assume that each reach maintains a sediment transport
capacity that results in a daily discharge of sediment from each reach that is equal to the
sediment inflow to the reach, regardless of the water inflow and outflow rates. This is a steady
state assumption, in that the sediment leaving the system every day should equal the sum of
the sediment incrementally generated within the system every day. This is not a physically
realistic assumption, but can be useful for evaluating the impact of assumptions regarding
sediment transport capacity in studies in which very little is known about sediment production
in channels.
Note that, depending on the watershed in which this model is applied, you may wish to
establish the parameters ‘a’ and ‘b’ so that proportionally more sediment is transported during
higher discharge events, as is often observed in practice (Walling, 2009).
c) Sediment density. Specify one value for sediment density (kg/m3). This value is only used to
determine the volume (m3) of sediment settles in each reservoir in each day, given a mass
(kg) of settled sediment.
d) The 'a' value for all reach carrying capacities (aQb), and 'c' value for all incremental
sediment loads (cQd), if desired. A value must be specified here only if option 3 is selected
in either the Incremental Sediment Load category ("Specify one coefficient for all
incremental flow locations") or the reach carrying capacity category ("Specify one
coefficient for all reaches").
Note the units of sediment concentration resulting from this equation are kg/m3.
e) The 'b' value for all reach carrying capacities (aQb), and 'd' value for all incremental
sediment loads (cQd), if desired. The value in this cell is jointly imported by the incremental
sediment load generation function and by the reach carrying capacity function to be used as
Wild et al. (2019)

SedSim User Manual

P a g e | 57

the exponent in each case. The only reason not to specify a value in this cell is if either (1) all
of the calibrated parameter values from a previous simulation are to be used, or (2)
appropriate options have been selected in order for the user to specify exponent values for
each location in separate worksheets (the "Annual_Sed_Loads" worksheet for the
incremental sediment load function, and the "Reach_Specifications" worksheet for the
reach carrying capacity function).
Note the units of sediment concentration resulting from this equation are kg/m3.
f) Channel Routing Method. Select one of the following options from the drop-down menu:
i.
ii.

Storage-Outflow Routing
Null Routing (Flow in = Flow out)

Storage-outflow routing determines daily reach outflow rates (m3/s) as a function of
reach storage (m3). This option is described in Appendix B. Null routing results in reach inflow
rate equal to reach outflow rate (a steady state assumption).
6.2.3. Network connectivity

The “Network connectivity” worksheet contains the network connectivity matrix, which
describes how system elements (reaches, reservoirs, junctions, and diversions) are connected
together.

Wild et al. (2019)

SedSim User Manual

P a g e | 58

Figure 6.3. Portion of "Network connectivity" worksheet of input file containing network configuration of the
system being simulated.

The "Network connectivity" worksheet is used to assign unique names, ID numbers, and
element types to every system element, and to describe to which junctions the upstream and
downstream ends of each reach, reservoir and diversion are connected. Column A contains the
element name and column B the element number. Column C is the element name used in the
simulation, and columns D and E are the inflow and outflow nodes associated with a particular
reach or reservoir element. Nodes are also called junctions.
The network connectivity matrix must be 5 columns wide. Every element that exists in
the modeled system, which includes reaches, reservoirs, junctions and diversions, must be
represented in at least one row of the network connectivity matrix. The process is different
depending on what type of element is of concern. Each will be addressed separately below.
The following describes what users should type into each specified cell. The string
enclosed in quotes should be entered into the specified cell without including the quotation
marks. Follow very closely the syntax suggested below. For example, when establishing a reach
element, the single string "ReachElement" must be typed in the first column, not the two
strings "Reach Element". Note that reaches must have a single inflow node and a single outflow
node to be modeled properly in the SedSim model. Conversely, reservoirs can have multiple
inflow nodes. For reaches and reservoirs, users should follow the steps described below to (1)
establish the existence of the element, and (2) define its inflow and outflow nodes (inflow
nodes must appear first, and outflow nodes must appear second). Names and ID numbers must
be unique for every system element. No element should ever be defined with an ID of "0".
Reaches:
Row 1 [required; used to establish existence of element, its ID, and its name)]
Column 1: "ReachElement"
Column 2: "Unique Reach ID #" (e.g., 105)
Column 3: "Unique Reach Name" (e.g., 12008)
Column 4: No input required
Column 5: No input required
Row 2 [required; used to establish the inflow node for the reach and the ID of that inflow
node]
Column 1: "Inflow Node"
Column 2: No input required
Column 3: No input required
Column 4: "Unique ID of the Inflow Junction; must be the same ID that is established
when the node/Junction is defined"
Column 5: No input required
Wild et al. (2019)

SedSim User Manual

P a g e | 59

Row 3 [required; used to establish the outflow node for the reach and the ID of that outflow
node]
Column 1: "Outflow Node"
Column 2: No input required
Column 3: No input required
Column 4: No input required
Column 5: "Unique ID of the Outflow Junction; must be the same ID that is established
when the node/Junction is defined"
Reservoirs:
Row 1 [required; used to establish existence of Reservoir, its ID, and its name)]
Column 1: "ReservoirElement"
Column 2: "Unique Reservoir ID #" (e.g., 105)
Column 3: "Unique Reservoir Name" (e.g., V009-Buon Tua Srah)
Column 4: No input required
Column 5: No input required
Row 2 [required; used to establish the first inflow node for the Reservoir and the ID of that
inflow node]
Column 1: "Inflow Node"
Column 2: No input required
Column 3: No input required
Column 4: "Unique ID of the first Inflow Junction; must be the same ID that is
established when the node/Junction is defined"
Column 5: No input required
Row 3 [optional; used to establish the second inflow node for the Reservoir and the ID of that
inflow node]
Column 1: "Inflow Node"
Column 2: No input required
Column 3: No input required
Column 4: "Unique ID of the second Inflow Junction, established when defining the
Junction"
Column 5: No input required
Row 4 [required; used to establish the outflow node for the Reservoir and the ID of that inflow
node]
Column 1: "Outflow Node"
Column 2: No input required
Column 3: No input required
Column 4: No input required
Column 5: "Unique ID of the Outflow Node, established when defining the Junction"
Wild et al. (2019)

SedSim User Manual

P a g e | 60

Junctions:
Row 1 [required; used to establish the existence of a Junction node]
Column 1: "JunctionElement"
Column 2: "Unique Junction ID #" (e.g., 2675)
Column 3: "Unique Junction Name" (e.g., Junction 10)
Column 4: No input required
Column 5: No input required
Diversions:
Row 1 [required; used to establish the existence of the Diversion, and the reservoir at which it
originates]
Column 1: "DivertedOutletElement"
Column 2: "Unique Diversion ID #" (e.g., 1385)
Column 3: No input required
Column 4: No input required
Column 5: “Name of Reservoir at which diversion originates, same as listed when
initially establishing reservoir.”
Row 2 [required; used to establish the ID of the junction node to which the diversion flows]
Column 1: “Outflow Node”
Column 2: No input required
Column 3: No input required
Column 4: No input required
Column 5: “Unique ID# of Junction to which diversion flows”
6.2.4. Sediment Loads

The “Sediment Loads” worksheet contains estimates of the cumulative average annual
sediment loads (in kg/yr) that are discharged past every location in the system at which
incremental flows enter. See Figure 6.4 below.

Wild et al. (2019)

SedSim User Manual

P a g e | 61

Figure 6.4. “Sediment Loads” worksheet of the input file.

This worksheet should contain four columns. In the first column, each row should
contain sediment load data relevant to each system element (reach or reservoir) into which
incremental sediment loads (and incremental flows) will enter the system at an upstream
inflow junction of the element. Data entry for the first element should begin on the second
row. The reservoir or reach name in each row of the first column must be identical to the name
as defined in the "Network connectivity" worksheet in the input file. Data for more locations
than just those being modeled can be stored in this worksheet, as the model will only import
information relevant to elements being simulated. Each column represents a different input
data requirement. Column data entries should begin on the second column.
a) Mean total annual cumulative sediment load (kg/year). In column 2, users should enter
the mean annual cumulative sediment yield (kg/km2-yr) that is expected to flow into each
incremental inflow location in a completely unregulated system. This should be the mean
annual sediment load that is discharged past each reservoir site, assuming no reservoirs
exist in the system to trap sediment, and assuming that the system is in relative balance.
Assuming the system is in balance requires that there are no long-term sediment sinks. All
sediment generated upstream of an incremental inflow location is expected to be
discharged past the incremental inflow location, on average, when the system is in balance).
The model will only import this value for calibration purposes (calibrating c and d in Eq.
(2.2)). Thus, this column is only required if the user selects the option to calibrate a
coefficient for each incremental inflow location in the "Simulation Specifications" worksheet
of the input workbook.
b) The c value for incremental sediment load generation (Eq. 2.2). In column 3, specify a
coefficient value to be used by the model directly in Eq. (2.2). This column is only required if
the user chooses not to conduct a calibration of incremental sediment load coefficients
using the model, and instead prefers to supply the model directly with these coefficient
values. The preference to supply these values instead of conducting a calibration must be
established in the "Simulation Specifications" worksheet of the input workbook.
c) The d value for incremental sediment load generation (Eq. 2.2). In column 4, specify an
exponent value to be used by the model directly in Eq. (2.2). This column is only required if
the user chooses not to conduct a calibration of incremental sediment load coefficients
using the model, and instead prefers to supply the model directly with these exponent
values. The preference to supply these values instead of conducting a calibration must be
established in the "Simulation Specifications" worksheet of the input workbook.
6.2.5. Incremental Flows

The “Incremental Flows” worksheet, shown in Figure 6.5, defines where the incremental
flows take place and when.
Wild et al. (2019)

SedSim User Manual

P a g e | 62

Figure 6.5. Portion of the “Incremental Flows” worksheet of the input file. In this example these are the flows that
enter the upstream reach of the reservoir, Junctrion 4, and get routed through the reach and enter the reservoir.

The sheet is only required to contain average daily incremental flow rates (m3/s) that
enter each junction in the modeled system. Incremental flows should only include those flows
that locally enter the modeled system, rather than flows that have already entered the system
upstream. Each column should represent a time series of incremental flow rates that enter a
particular junction, whereas each row represents the incremental flow rates for all junctions for
a particular date. As with all other time series sheets, time series data should begin on the
second column and second row, dates should begin on the first column and second row, and
the reservoir name should be placed on the first row of every time series column (beginning
with the second column). The junction name in each column must be identical to the junction's
name as defined in the "Network connectivity" worksheet in the input file. Time series data for
more than just those junctions for which elevation targets will apply can be stored in the
worksheet, as can data for more dates than just those contained within the simulation horizon.
The model will locate and import only the data necessary to conduct the simulation. However,
be careful not to store incremental flow data in this worksheet for junctions that exist in the
modeled system but at which incremental flows do not occur in the current simulation.
6.2.6. Reach Specifications

“Reach Specifications” worksheet stores all data relevant to sediment transport and
flow routing in river reaches (channels). Each row corresponds to a different reach in the
Wild et al. (2019)

SedSim User Manual

P a g e | 63

modeled system for which data exist. The reach name stored in each row must be identical to
the reach name as defined in the "Network connectivity" worksheet in the input file. Each
column in this worksheet corresponds to a different category of information for which the user
should supply data for all reaches for which the data category is relevant. (Each of these
categories is introduced in detail below). If all data columns do not apply to a particular reach,
no row for the reach is needed. Do not rearrange the locations of columns, as the model
searches in specific columns for specific information. Importantly, the model converts
reservoirs to reaches when the user runs an unregulated simulation. Thus, each reservoir that
will exist in the regulated system simulation should be listed in a separate row in this worksheet
(in addition to all reaches that will exist in the regulated system). The data you specify in each
column on the row corresponding to each reservoir name will only be applied to the
unregulated simulation.

Figure 6.6. “Reach Specifications” data worksheet.

a) Flow Routing Flow routing coefficient δ. This data column is required only if the StorageOutflow routing method is selected in the "Simulation Specifications" worksheet within the
input file. The specified coefficient value will be used in the pair of reach routing Eq. (1) and
Eq. (2) for the corresponding reach.
b) Flow Routing Flow routing exponent γ. This data column is required only if the StorageOutflow routing method is selected in the "Simulation Specifications" worksheet within the
input file. The specified exponent value will be used in the pair of reach routing Eq. (1) and
Eq. (2) for the corresponding reach.
c) Ponding storage volume (m3). This data column is required only if the Storage-Outflow
routing method is selected in the "Simulation Specifications" worksheet within the input
file. The specified value will be used in the pair of reach routing Eq. (1) and Eq. (2) for the
corresponding reach.
d) Initial reach storage (m3) at beginning of day on user-specified simulation start date. This
information is required because the model predicts end-of-period storage volume values for
every reach. To predict the end-of-period storage volume for the simulation start date, the
user must supply the beginning of period storage volume for the simulation start date. This
Wild et al. (2019)

SedSim User Manual

P a g e | 64

value is identical to the end-of-period storage in the date before the specified simulation
start date.
e) Sediment routing coefficient 'a'. This information is required only if the user chooses the
"Specify a separate coefficient and exponent for each reach" calibration option in the reach
sediment discharge category (within the "Simulation Specifications" worksheet in the input
file). The specified coefficient value will be used in the sediment routing Eq. (2.3) for the
corresponding reach.
f) Sediment routing coefficient 'b'. This information is required only if the user chooses the
"Specify a separate coefficient and exponent for each reach" calibration option in the reach
sediment discharge category (within the "Simulation Specifications" worksheet in the input
workbook). The specified coefficient value will be used in the sediment routing Eq. (2.3) for
the corresponding reach.
g) Initial sediment mass (kg) available in the reach at beginning of simulation start date. This
represents the amount of sediment (kg) available in storage (bed sediment) in each reach at
the beginning of the day on the simulation start date. (Sediment in suspension at the
beginning of simulation in all reaches is assumed to be zero. Absence of sediment in
suspension in reaches during the first simulation time period will briefly affect the quantity
of sediment that is scoured from (or that settles onto) the river channel bed. For this
reason, you may wish to run the simulation for a few extra days, or simply ignore the first
few days of results.). An accurate value may be difficult to determine in practice, in which
case the user could assume some large value of initial sediment availability, simply to
prevent complete exhaustion of sediment supply in reaches. Relative Changes in sediment
mass stored in the reach from the initial value assumed would then become more
important than absolute changes. This initial sediment mass value is required because the
model predicts end-of-period sediment mass values for every reach. To predict the end-ofperiod sediment mass on the simulation start date, the user must supply the beginning of
period mass for the simulation start date. This value is identical to the end-of-period mass in
the date before the specified simulation start date.
6.2.7. E-V-A-S

The “E-V-A-S” worksheet contains the Elevation (meters above mean sea level: mamsl)Volume (m3)-Area (ha)-Sediment (cumulative fraction) data for each reservoir in the modeled
system.
Starting with the first column on the left, the user must specify exactly four columns of
information for each reservoir, in the exact order in which they appear below:
a) Column 1: Elevation data
Wild et al. (2019)

SedSim User Manual

P a g e | 65

b) Column 2: Water volume data
c) Column 3: Water surface area data
d) Column 4: Cumulative fraction (in the range from 0-1) of settled sediment to be stored
below each corresponding elevation in Column 1 in each time step of simulation.
No blank columns should be placed between the sets of 4 columns (e.g., if your system
includes 41 reservoirs, then you should populate 41 sets of 4 columns, or 41*4=164 columns of
E-V-A-S data, with no blank columns). The first three rows are reserved for column headers and
reservoir identification. Thus, data should be first entered on the fourth row. In the first row of
the worksheet, above the first of the four column entered for each reservoir, users should enter
the name of the reservoir to which the data correspond, exactly as the name appears in the
"Network connectivity" worksheet.
The elevation, volume and area data are generally measured or estimated using GIS.
These data are used for four purposes. First, the data is used in conjunction with each
reservoir's empty and full supply level elevations to estimate each reservoir's active and dead
storage capacity. Second, the data are used to determine the water surface elevation
corresponding to the reservoir's storage in each time period, and therefore the capacities of
discharge outlets and the head available for hydropower production. Third, the Area (ha) data
are used to determine the reservoir's average surface area during each time period
corresponding to each water level elevation, which is used to determine evaporation during
each time period. Fourth, the sediment information is used to continually adjust the originally
specified elevation-volume data (from the “E-V-A-S” worksheet) as the simulation proceeds to
account for sediment accumulation in the water storage space. That is, less space is available to
store water as sediment accumulates, and more space is available to store water as sediment is
removed from the reservoir via sediment management techniques.
The cumulative fraction of sediment stored below different elevations in the reservoir
depends on several factors, including the reservoir operating policy, reservoir shape, and
predominant grain size of settling sediment. (More information about how to specify a
cumulative curve for a particular reservoir can be found in Morris and Fan, 1997 and Strand and
Pemberton, 1987).
If the user does not specify any sediment storage information, the model will assume
that sediment is continually deposited equally in the reservoir at all elevations. For example, if a
reservoir’s total water volume (dead and active) is stored over 20 m of depth, when sediment is
deposited SedSim will reduce the available water storage capacity at each of the depths (i.e.,
reduce the cumulative water storage volume value for each elevation) in proportion to the
fraction each depth represents out of the total 20 m differential.
When a sediment removal practice is simulated for a particular reservoir (e.g., flushing
or general sediment removal), this sediment is assumed to be removed equally from all
elevations in the reservoir. For flushing, this assumption is particularly appropriate because the
Wild et al. (2019)

SedSim User Manual

P a g e | 66

flushing channel extends from the upstream end of the reservoir (at the highest elevation) to
the base of the dam (the lowest elevation). Thus, when flushing occurs, some sediment is likely
removed from every elevation in the reservoir’s profile. When this sediment is removed, more
volume at each elevation becomes available for water storage (i.e., there is a partial recovery of
the elevation-volume curve towards its original profile).

Figure 6.7. Portion of "E-V-A-S" worksheet of input data file.

6.2.8. Storage Volume Elevation Target

“Storage Volume Elevation Target” worksheet is only required if the elevation target
option is selected for any reservoir in the reservoir operations goal column in the "Reservoir
Specifications" worksheet in the input data file. Otherwise, this worksheet will not be used by
the model, and thus can be ignored (deleted, left blank, or kept in its current state). The sheet
is only required to contain time series of the end-of-day water elevation (mamsl) for reservoirs
for which the elevation target option is selected in the reservoir operations goal column in the
"Reservoir Specifications" worksheet in the input data file. The time series of targets must
contain a target value for the end of the day the simulation starts, but not for the end of the
day on the date before the simulation begins. Each column should represent a time series of
elevation target values for a particular reservoir, whereas each row represents the elevation
target values for all reservoirs for a particular date. As with all other time series sheets, time
series data should begin on the second column and second row, dates should begin on the first
column and second row, and the reservoir name should be placed on the first row of every time
series column (beginning with the second column). The reservoir name in each column must be
identical to the reservoir's name as defined in the "Network connectivity" worksheet in the
input file. Data for more than just those reservoirs for which elevation targets will apply can be
stored in the worksheet, as can data for more dates than just those contained within the
simulation horizon. The model will locate and import only the data necessary to conduct the
simulation.
Wild et al. (2019)

SedSim User Manual

P a g e | 67

6.2.9. Storage Volume Target

This worksheet is only required if the storage target option is selected for any reservoir
in the reservoir operations goal column in the "Reservoir Specifications" worksheet in the input
data file. Otherwise, this worksheet will not be used by the model, and thus can be ignored
(deleted, left blank, or kept in its current state). The sheet is required to contain time series of
the end-of-day water storage targets (m3) for reservoirs for which the storage target options
are selected in the reservoir operations goal column in the "Reservoir Specifications" worksheet
in the input data file. The time series of targets must contain a value for the target for the end
of the day the simulation starts, but not for the end of the day on the date before the
simulation start date. Each column should represent a time series of storage target values for a
particular reservoir, whereas each row represents the storage targets for all reservoirs for a
particular date. As with all other time series sheets, time series data should begin on the second
column and second row, dates should begin on the first column and second row, and the
reservoir name should be placed on the first row of every time series column (beginning with
the second column). The reservoir name in each column must be identical to the reservoir's
name as defined in the "Network connectivity" worksheet in the input file. Data for more than
just those reservoirs for which storage targets will apply can be stored in the worksheet, as can
data for more dates than just those contained within the simulation horizon. The model will
locate and import only the data necessary to conduct the simulation.

Figure 6.8. Portion of the “Storage Volume Elevation Target worksheet of the input file.

Wild et al. (2019)

SedSim User Manual

P a g e | 68

6.2.10. Evaporation Data

This worksheet is only required to be created if the user wants to account for
evaporation during simulation of reservoir operations. “Evaporation Data” worksheet stores
average monthly evaporation data (mm) for each reservoir. Each row corresponds to a different
reservoir in the modeled system for which data exist. The reservoir name stored in each row
must be identical to the reservoir's name as defined in the "Network connectivity" worksheet in
the input file. Each column in this worksheet corresponds to a different month for which the
average monthly evaporation (mm) is required. If no evaporation data is available for a
particular reservoir, a row for the reservoir is not required. Do not rearrange the locations of
columns, as the model searches each column in order (1 through the 12), assuming they are in
chronological order.

Figure 6.9. “Evaporation data” worksheet of input file.

6.2.11. Environmental Flow Data

This worksheet is only required to be created if the user wants to require minimum
environmental flow releases (m3/s) at any reservoirs during simulation. The “Environmental
Flow Data” worksheet stores the minimum average daily flow (m3/s) that must be released into
the downstream channel at each reservoir site. This requirement is assumed to take
precedence over rule curve-based requirements. For example, if the reservoir's pre-established
operating policy dictates that no water should be released in a particular time period, the
model will override this goal to make a release downstream to attempt to satisfy the minimum
environmental flow requirement. Each row corresponds to a different reservoir in the modeled
system for which data exist. The reservoir name stored in each row must be identical to the
reservoir's name as defined in the "Network connectivity" worksheet in the input file. Each
column in this worksheet corresponds to a different month for which the flow data are to be
specified. If no data are available for a particular reservoir, a row for the reservoir is not
required. Do not rearrange the locations of columns, as the model searches each column in
order (1 through the 12), assuming they are in chronological order.
6.2.12. Reservoir Specifications

Wild et al. (2019)

SedSim User Manual

P a g e | 69

“Reservoir Specifications” worksheet stores a significant quantity of data (aside from
time series) required to simulate sediment and water flows in reservoirs. Each row corresponds
to a different reservoir in the modeled system for which data exist. The reservoir name stored
in each row must be identical to the reservoir name as defined in the "Network connectivity"
worksheet in the input file. Each column in this worksheet corresponds to a different category
of information for which the user should supply data for all reservoirs for which the data
category is relevant. (Each of these categories is introduced in detail below). If a particular data
column does not apply to a particular reservoir, no input is required in that column. Do not
rearrange the locations of columns from the order in which they appear below, as the model
searches in set columns for specific information.

Wild et al. (2019)

SedSim User Manual

P a g e | 70

Figure 6.10. Portions of "Reservoir Specifications" worksheet of input data file.

a) Full supply level elevation (mamsl). This information represents the upper elevation
(mamsl) threshold of the active storage zone, and is used by the model to estimate each
reservoir's dead and active storage capacity. If you know the dead and active storage
capacity, skip this column and enter the capacity data in the two appropriate capacity
columns (discussed below). The only reason the model attempts to determine dead and
active storage capacity is to report in the model output separate estimates of the loss in
storage capacity within each zone that results from sediment deposition. This elevation
information is only required when the "Compute active and dead storage capacity using
elevation-volume-area data and dead/active elevations" option is selected in the column
(within the "Reservoir Specifications" worksheet) that represents the method for
determining active and dead storage capacity (see below for a description of this column). If
the user selects this option, indicating the low and full supply level elevations will be used to
determine active and dead storage capacity, the model will import the specified upper and
lower elevations and use them to interpolate over the Elevation-Volume-Area data (within
the "E-V-A-S" worksheet) to determine the dead and active storage capacities. Note that
the active storage zone is defined here as the zone of the reservoir within which water
elevations are fluctuated for purposes of hydropower production, whereas the dead
storage is defined as all storage that remains at an elevation below active storage zone (i.e.,
below the low supply level elevation).
b) Low supply level elevation (mamsl). This information represents the lower elevation
(mamsl) threshold of the active storage zone, and is used by the model to estimate each
reservoir's dead and active storage capacity. If you know the dead and active storage
capacity, skip this column and enter the capacity data in the two appropriate capacity
columns (discussed below). The only reason the model attempts to determine dead and
active storage capacity is to report in the model output separate estimates of the loss in
storage capacity within each zone that results from sediment deposition. This elevation
information is only required when the "Compute active and dead storage capacity using
elevation-volume-area data and dead/active elevations" option is selected in the column
(within the "Reservoir Specifications" worksheet) that represents the method for
determining active and dead storage capacity (see below for a description of this column). If
Wild et al. (2019)

SedSim User Manual

P a g e | 71

the user selects this option, indicating the low and full supply level elevations will be used to
determine active and dead storage capacity, the model will import the specified upper and
lower elevations and use them to interpolate over the Elevation-Volume-Area data (within
the "E-V-A-S" worksheet) to determine the dead and active storage capacities. Note that
the active storage zone is defined here as the zone of the reservoir within which water
elevations are fluctuated for purposes of hydropower production, whereas the dead
storage is defined as all storage that remains at an elevation below active storage zone (i.e.,
below the low supply level elevation).
c) Brune Curve Type (L=Low trapping; M=Median trapping; H=High trapping; number
(fraction) = constant trapping). Sediment trapping in reservoirs is given by Brune's curve
(1953). Brune (1953) originally specified three curves, each fitted to different portions of his
available data set. Type one of the following options into the cell for the reservoir of
interest (type only the value appearing in quotation marks).
i.
ii.
iii.
iv.
v.

“L”. This represents the Brune (1953) original low trapping curve (for colloidal/dispersed
and fine-grained particles).
“M”. This represents the Brune (1953) original median curve.
“H”. This represents the Brune (1953) upper (or high) trapping curve (for highly
flocculated and coarse sediments).
“C”. This represents use of the Churchill (1948) Method, as described in Chapter 2.
Enter a numeric fraction from 0-1 to specify a constant trapping efficiency instead of
using the Brune curve (e.g. enter “0.4” for a constant trapping efficiency of 40%
throughout simulation at the reservoir).

Note: If the user specifies no value in this cell for a particular reservoir, the default is to assume
zero trapping efficiency (0%).
d) Time scale over which Trap Efficiency (TE) is computed (A=Annual; M=Monthly). This value
is only required if the user specifies “L”, “M”, or “H” in the Brune Curve Type Column of the
“Reservoir Specifications” worksheet (i.e., if the user elects to use Brune curve trapping
instead of specifying a constant trapping efficiency). Type one of the following options into
the cell for the reservoir of interest (type only the value appearing in quotation marks).
i.
ii.

“A”. A represents Annual. This indicates the model should apply the Brune (1953) curve
using Annual residence time (computed using data from the previous 365 days).
“M”. M represents Monthly. This indicates the model should apply the Brune (1953)
curve using Monthly residence time (computed using data from the previous 30 days).

To provide more detail on these two options, trapping in reservoirs in SedSim is determined
using Brune's curve (1953). Trap Efficiency depends on a reservoir's residence time. The first
option computes the residence time based on the average reservoir storage (m3) and
average outflow volume (m3) over the previous 30 days. The second option computes the
Wild et al. (2019)

SedSim User Manual

P a g e | 72

residence time based on the average reservoir storage (m3) and average outflow volume
(m3) over the previous 365 days.
Note: If the user specifies “L”, “M”, or “H” in the Brune Curve Type Column of the
“Reservoir Specifications” worksheet but does not specify a value in the column described
here, the default will be “A”, as described above. If the user elects to apply a constant trap
efficiency in the in the Brune Curve Type Column of the “Reservoir Specifications”
worksheet, no information from this worksheet will be used.
e) Initial sediment in reservoir (kg) at beginning of simulation start date (time t=0). This
represents the amount of sediment (kg) available in storage (at the bottom of the reservoir
rather than in suspension) in each reservoir at the beginning of the day on the simulation
start date. (Sediment in suspension at the beginning of simulation in all reservoirs is
assumed to be zero. Absence of sediment in suspension in reservoirs during the first
simulation time period will briefly affect the quantity of sediment that settles in the
reservoir, and that is discharged from the reservoir, in the first few days of simulation. For
this reason, you may wish to run the simulation for a few extra days, or simply ignore the
first few days of results.) This column represents sediment that is available to be released
from the reservoir, and that exists in the active and/or dead storage zones. This option is
designed to allow a simulation of a reservoir that has already experienced sedimentation.
Hence, for a new reservoir, the value in this initial sediment availability column should be 0,
as no sediment has accumulated in either zone by the start of simulation. Clearly, a new
reservoir will have some sediment availability where the original river bed existed, but this
model assumes none of this sediment is available to be released from the reservoir, and is
therefore ignored. As with other initial (time zero) values, this is required because the
model attempts to predict the end-of-period sediment mass on the simulation start date, so
the user must supply the beginning of period mass for the simulation start date. This value
is identical to the end-of-period mass in the date before the specified simulation start date.
f) Describe Reservoir's Hydropower and Diversion Capabilities. Select one of the following
options from the drop-down menu:
i.) Power generation only
ii.) Diversion only
iii.) Power generation & diversion
If no drop-down menu is available, type one of the listed options into the cell exactly as it
appears above. Leave the cell blank if the reservoir has neither power production nor
diversion capabilities. As detailed in Chapter 2, the model simulates four different types of
reservoirs. A reservoir can have (or not have) hydropower production capabilities, and can
have (or not have) the capability to divert water away from the downstream channel and
instead to another site within the modeled system. In this column, the user specifies which
of the reservoir types described in Chapter 2 is applicable to the reservoir of interest. This
Wild et al. (2019)

SedSim User Manual

P a g e | 73

column is designed for users to specify only whether the reservoir of interest has a
hydropower and/or diversion capability.
g) Reservoir operations goal. Select one of the following options from the drop-down menu:
i.) Meet specified daily water elevation (mamsl) targets.
ii.) Meet specified daily water storage targets (m3).
If no drop-down menu is available, type one of the listed options into the cell exactly as it
appears above. This column allows the user to specify the operational goal of each
reservoir. If sediment accumulation in the reservoir is negligible in comparison to the
storage capacity, then this option will result in essentially the same policy one would
establish using storage targets, because an elevation corresponding to every water storage
value can be determined from the user-supplied Elevation-Volume data. However, if
sediment accumulation in the reservoir is significant, the water elevation targets option
allows specified elevations to be maintained in the reservoir over time, which may require
that less water be maintained in storage as the simulation proceeds due to sediment
accumulation in the reservoir's storage space.
h) Initial reservoir storage volume (m3) at beginning of simulation start date (time t=0). This
information is required because the model predicts end-of-period storage volume values for
every reservoir. To predict the end-of-period storage volume for the simulation start date,
the user must supply the beginning of period storage volume for the simulation start date.
This value is identical to the end-of-period storage volume on the date before the specified
simulation start date.
If the user specifies no value, the model default assumption is that the reservoir’s initial
water storage value is equal to the dead storage capacity (i.e., the water surface elevation is
at the bottom of the active storage elevation and top of the dead storage elevation).
i) Maximum of turbines' centerline elevation and reservoir's tailwater elevation (mamsl). In
this column, specify the maximum of the constant tailwater elevation (the elevation of the
water in the channel immediately downstream of the reservoir) and the turbines' centerline
elevation. The larger of these two values will serve as the lower elevation used to compute
hydropower head (and therefore hydropower production) during each time period, as given
in Chapter 2. The value specified in this cell is assumed to remain constant throughout
simulation. The hydropower head during the time period is assumed to be the average
difference between the reservoir's water elevation and the constant tailwater elevation at
the beginning and end of each time period. These data are only required if the user has
selected that the reservoir has hydropower capabilities.
j) Hydropower plant capacity (MW). In this column, specify the hydropower plant capacity
(MW) for each reservoir. This value should represent the sum of the rated (nameplate)
Wild et al. (2019)

SedSim User Manual

P a g e | 74

capacities of all installed generators. No combination of turbine flow, net head, and
efficiency will be permitted to produce an amount of hydropower exceeding this plant
capacity value. This information is only required if the user has selected that the reservoir
has hydropower capabilities.
k) Hydropower plant efficiency (fraction). In this column, specify the hydropower plant
efficiency. This value is assumed to remain constant throughout simulation (not assumed to
be a function of head and discharge). This information is only required if the user has
selected that the reservoir has hydropower capabilities. The value in this column will be
used to compute hydropower production during each time period.
l) Reservoir Length (m). This is the reservoir length at the normal reservoir operating level.
This input is used to simulate sluicing and density current venting. For sluicing, it is used to
compute the Sedimentation Index (SI), which is used in SedSim to implement the Churchill
Method (1948) to compute reservoir trap efficiency during sluicing. For density current
venting, this input is optional, as it is used to determine a default minimum venting
efficiency value for the user if the user does not specify one.
m) Perform sediment Flushing? Select one of the following options from the drop-down menu:
i.) Yes
ii.) No
If no drop-down menu is available, type one of the listed options into the cell exactly as it
appears above. In this column, specify whether or not sediment flushing should be
attempted in the reservoir of interest. (Note that more detailed background information
about flushing, and the approach to flushing taken by this model, is contained in Chapter 2).
If you select "Yes", you must supply additional flushing information in (1) the "Flushing"
worksheet, and (2) the "Outlet Capacity Data" worksheet. See descriptions of these two
worksheets for more details. Also, sediment flushing can only be performed if the user
chooses to perform the hydrologic simulation using the SedSim model.
n) Remove a specified sediment mass from reservoir without altering reservoir operations?
Select one of the following options from the drop-down menu:
i.) Yes
ii.) No
In this column, specify whether or not general sediment mass removal should be simulated
at this reservoir, via a technique that does not require reservoir operational changes (e.g.,
dredging). If no drop-down menu is available, enter either “Yes” or “No” into the cell for
each reservoir. Any entry other than “Yes” (e.g., a blank cell) will result in no mass removal.
(Note that more detailed background information about this technique, which results in net
Wild et al. (2019)

SedSim User Manual

P a g e | 75

sediment removal from a reservoir without altering the reservoir's operations, is provided
in Chapter 2). If you select "Yes", you must supply additional information in the "General
Sediment Removal" worksheet.
o) Perform Sediment Bypass? Select one of the following options from the drop-down menu:
i.) Yes
ii.) No
In this column, specify whether or not sediment bypassing should be simulated at this
reservoir. If no drop-down menu is available, enter either “Yes” or “No” into the cell for
each reservoir. Any entry other than “Yes” (e.g., a blank cell) will result in no sediment
bypassing. (Note that more detailed background information about sediment bypassing is
provided in Chapter 2). If you select "Yes", you must supply additional information in the
"Bypassing" worksheet.
p) Perform Density Current Venting? Select one of the following options from the drop-down
menu:
i.) Yes
ii.) No
In this column, specify whether or not density current venting should be simulated at this
reservoir. If no drop-down menu is available, enter either “Yes” or “No” into the cell for
each reservoir. Any entry other than “Yes” (e.g., a blank cell) will result in no density current
venting. (Note that more detailed background information about density current venting is
provided in Chapter 2). If you select "Yes", you must supply additional information in (1) the
"Density Current Venting" worksheet, and (2) the "Outlet Capacity Data" worksheet.
q) Perform Sluicing? Select one of the following options from the drop-down menu:
i.) Yes
ii.) No
In this column, specify whether or not sluicing should be simulated at this reservoir. If no
drop-down menu is available, enter either “Yes” or “No” into the cell for each reservoir. Any
entry other than “Yes” (e.g., a blank cell) will result in no sluicing. (Note that more detailed
background information about density current venting is provided in Chapter 2). If you
select "Yes", you must supply additional information in (1) the "Sluicing" worksheet, and (2)
the "Outlet Capacity Data" worksheet.

Wild et al. (2019)

SedSim User Manual

P a g e | 76

6.2.13. Outlet Capacity Data

“Outlet Capacity Data” worksheet provides elevation vs. discharge capacity tables for
every outlet that will be operated on each reservoir. This worksheet is only required if the user
chooses to perform a hydrologic simulation using the SedSim Model. For every outlet at every
reservoir, the user must supply two columns of data (elevation and discharge, from left to
right). The number of outlets for which data must be provided for a particular reservoir will
depend on the number of outlets the reservoir has, which is strictly dictated by the reservoir
type (as selected in the "Reservoir Specifications" worksheet). More detailed discussion of
outlets is provided in Chapter 2. Specifically, Figure 3.9 describes the number and types of
outlets required for each reservoir type. For example, if a model is being built for ten reservoirs,
and every reservoir has the same basic capability (hydropower production), then only two
outlet types are required per reservoir (Hydropower Outlet and Overflow Outlet). For these ten
reservoirs, a total of 10*2 = 20 pairs (40 columns) of elevation vs. max. discharge data must be
supplied by the user. If flushing were to be performed in each of the ten reservoirs, an
additional 10 outlets would be required for a total of 80 columns. The most outlets required for
a reservoir occurs in the case of a diversion dam. For example, if the ten reservoirs were instead
hydropower/diversion reservoirs with flushing capabilities, a total of 10*4 = 40 outlets would
be required.

Figure 6.11. “Outlet Capacity Data” worksheet.

Every time outlets for a particular reservoir are defined, the name of the reservoir for
which the outlet capacity vs. elevation data are provided must be listed in the first row of the
first of all the elevation-discharge columns defined for a particular reservoir. (The reservoir
name at the top of each column must be identical to the reservoir's name as defined in the
"Network connectivity" worksheet in the input file). For example, if a particular reservoir has
four outlets, then a total of 8 consecutive columns should be defined for the reservoir, with the
name of the reservoir appearing only once at the very top of the first of the 8 columns.

Wild et al. (2019)

SedSim User Manual

P a g e | 77

Every time a new outlet for a reservoir is defined, the user should select the outlet type
from a drop-down menu in the second row in the column in which the elevation data are
provided (the left-hand column in each pair of two elevation-discharge columns). If no dropdown menu is available, type one of the listed options into the cell exactly as it appears below.
For example, if a particular reservoir has four outlets, then an outlet type should be selected
four times, each in the second row of the elevation data column. The drop-down menu offers
the following six options:
i.
ii.
iii.
iv.
v.
vi.

Controlled Outlet
Hydropower Outlet
Diversion Outlet
Hydropower/Diversion Outlet
Spillway Outlet
Low level Outlet

Data in each column should not begin until the fifth row. This is only to allow space for
the user to list data type headings (e.g., "Flow") and data units ("cms"), if desired (this is not a
requirement).
Data for more than just those reservoirs that will be modeled in the current simulation can be
stored in the worksheet. The model will use only the data for those reservoirs that will be
modeled. This saves time in the input file preparation process in instances in which one wishes
to run simulations that explore different of reservoir development.
6.2.14. Sediment Management Specifications Worksheets

The input file can contain up to five different sediment management worksheets:
a)
b)
c)
d)
e)

Flushing
Sluicing
Density Current Venting
General Sediment Removal
Bypassing

The “Flushing”, “Sluicing”, and “General Sediment Removal” worksheets have similar
formatting requirements. Likewise, the “Density Current Venting” and “Bypassing” worksheets
have similar formatting requirements. Thus, common formatting requirements are presented
here before each of the worksheets is presented, to avoid repeating this information in the
sections for each individual worksheet.
Note that multiple sediment management techniques can be simulated in the same
reservoir at different times. However, flushing, sluicing and density current venting cannot be
Wild et al. (2019)

SedSim User Manual

P a g e | 78

simulated concurrently. Any management technique being simulated will be allowed to finish
before a new technique is begun. For example, if sluicing is being simulated at a particular
reservoir and flushing is meanwhile scheduled to occur, the start of flushing will be delayed
until sluicing is completed.
6.2.14.1. Comments on “Flushing”, “Sluicing”, and “General Sediment Removal” worksheets

Each worksheet is only required if the user will implement the corresponding sediment
management technique in any reservoir (e.g., if you specify "Yes" in the Flushing column of the
“Reservoir Specifications” worksheet, then you need a “Flushing” worksheet to store additional
information). For every in which a sediment management technique is to be applied to a
reservoir, a new block of 12 (for “Flushing”), 8 (for “Sluicing”), and 5 (for “General Sediment
Removal”) columns of information must be supplied by the user. For example, if a model is
being built for ten reservoirs, and both Flushing and General Sediment Removal are to be
conducted at some time during simulation at every one of the ten reservoirs, then a total of
10*12 + 10*5 = 170 columns of information must be supplied by the user.
Within each worksheet, the name of the reservoir for which the technique and data
apply must be listed in the first row and in the first of the set of columns (e.g., in the “Sluicing”
worksheet, the reservoir name should appear in (row 1, column 1) for the data set of the first
reservoir in which sluicing will occur, in (row 1, column 9) for the data set for the second
reservoir in which sluicing will occur, etc. The reservoir name should only be listed once in each
worksheet for each reservoir in which the corresponding sediment management technique will
be simulated.
The reservoir name at the top of each column must be identical to the reservoir's name
as defined in the "Network connectivity" worksheet in the input file. Data for more than just
those reservoirs for which you have selected to implement a sediment management technique
can be stored in the worksheet, as the actual implementation of the technique is controlled by
the preferences selected in the "Reservoir Specifications" worksheet. The model will use only
the data for those reservoirs at which sediment management techniques will actually be
implemented.
Each row in these three worksheets, starting with the third row, represents a different
event date. The events should appear in chronological order. For example, if sluicing will be
simulated twice at one reservoir during the simulation horizon, from 9/1/1980 to 9/15/1980, as
well as from 8/3/1985-8/8/1985, data corresponding to each of these separate events should
appear in separate, consecutive rows. If some of the required user input data will remain the
same every time the technique is simulated (e.g., the flushing duration, flushing minimum
discharge, etc.), then the user is only required to enter this information in the first row
(corresponding to the first event). SedSim will import information from the first row when the
user does not specify information for future events. If particular specifications are different for
Wild et al. (2019)

SedSim User Manual

P a g e | 79

different events (e.g., the target sluicing water surface elevation), then the user can also just
specify new information in each row. For example, if ten sluicing events are scheduled in ten
different rows in column 1, then the user can specify ten different sets of input data in the
input data columns.
6.2.14.2. Comments on “Density Current Venting” and “Bypassing” worksheets

In these two worksheets, each row corresponds to a different reservoir in the modeled
system for which data exist. The reservoir name stored in the first column of each row must be
identical to the reservoir name as defined in the "Network connectivity" worksheet in the input
file. Each column in this worksheet corresponds to a different category of information for which
the user should supply data for all reservoirs for which the data category is relevant. (Each of
these categories is introduced in detail below). If a particular data column does not apply to a
particular reservoir, no input is required in that column. The Data entry should begin on the
second row and second column. Do not rearrange the order of columns from how they appear
in each section below, as the model searches in set columns for specific information.
6.2.15. Flushing

The following information must be supplied by the user for every reservoir at which
flushing will be simulated. Some user inputs are described below as Optional, meaning these
inputs are extra features that are not required to run a simulation. The columns should be
specified in exactly the order (left to right) in which they are listed below.

Figure 6.12. “Flushing” worksheet of input file.

Wild et al. (2019)

SedSim User Manual

P a g e | 80

a) Target Date: Beginning of Drawdown. In this column, enter the calendar date(s) defining
when reservoir drawdown for flushing can begin (if the inflow rate criterion defined below
is satisfied) and the number of days of flushing. Every time flushing (and therefore
drawdown) are to occur, a new date should be listed in a new row in the column for this
reservoir.
b) Flushing Duration. This is the number of days during which flushing criteria (flow and
elevation requirements discussed below) must be satisfied, and does not include any days
that only serve to draw down (empty) the reservoir.
c) The maximum flushing water surface elevation (WSE). This is the maximum elevation of
water in the reservoir that will still result in successful flushing. (For example, if the original
river bed elevation, and/or low-level outlet invert elevation, is 56 masl, then 56 masl will be
the target drawdown elevation. However, flushing can still be successful if the reservoir is
not able to fully draw down to 56 masl. If, for example, successful flushing will still occur if
the reservoir is drawn down to 58 masl (2 m above the target) because free flow conditions
are relatively well maintained, then the user would input 58 masl into this column for the
flushing date(s) of interest.
d) Minimum inflow rate at which drawdown is initiated. Optional. The minimum inflow rate
that will permit the drawdown process to be initiated on or after the target date specified in
the date column. The user must specify the date on which drawdown should first be
considered. The model waits until this specified date to consider drawdown, but does not
actually initiate drawdown until the reservoir inflow exceeds the value specified in this cell.
This will prevent a drawdown that begins too early in the dry season, before flows at the
beginning of the wet season begin to increase. If the user does not enter a value in this
column, the model will assume that no such threshold exists for the initiation of drawdown.
e) Flushing channel bottom width (m). This is the width of the bottom of the channel that will
form during flushing. Assuming the channel will form a trapezoidal cross-section over time,
this width represents the (smaller) bottom width of the trapezoid. This value is used to
determine the dimensions of the flushing channel, which is used to determine the quantity
of sediment removed during flushing as the flushing channel grows over time. The channel
width can be approximated fairly well as a function of flushing discharge, though it also
likely depends on channel slope and sediment properties. In absence of field measurements
or other data, Atkinson (1996) suggests estimating this property using the equation given
below.
If the user does not specify any channel width, the model will use this relationship below to
determine channel width as a function of the user-specified minimum flushing flow value
(the reservoir inflow (m3/s) below which flushing is assumed not to be successful on a
flushing day).
Wild et al. (2019)

SedSim User Manual

P a g e | 81

W f = 12.8Q f 0.5
where W f is the width (m) of the channel formed during flushing and Q f is the flushing
discharge (m3/s).
This value cannot be different for each flushing date, so the value is imported from the first
row of flushing data.
f) Flushing channel average side slope (m/m, 1 horizontal to SS f vertical). This is the slope of
the side walls of the incised channel formed during flushing. This value is used to determine
the dimensions of the flushing channel, which is used to determine the quantity of
sediment removed during flushing as the flushing channel grows over time. Flushing
channel side slopes can vary widely, and depend upon the degree of sediment
consolidation, sediment properties, the depth of the deposits through which the incised
channel is cut, and the extent of water level fluctuation during flushing. In the absence of
field measurements or other data, Atkinson (1996) suggests using the formulation proposed
by Migniot (1981), as given below. If the user does not specify any side slope, the model will
use this relationship to determine side slope as a function of the user-specified sediment
density.
31.5

1

Side slope = 5 𝜌𝜌4.7
𝑑𝑑 �10�

where ρ d is the average dry density (t/m3) of the sediment through which the flushing
channel will be cut.
This value cannot be different for each flushing date, so the value is imported from the first
row of flushing data.
g) Maximum flushing drawdown rate (m/day). Optional. This is the maximum rate at which
the water level of the reservoir can be drawn down per day during the drawdown phase of
sediment flushing. Rapid drawdown of a reservoir can lead to bank failure, landslides, or
similar events, in which large quantities of soil fall into the reservoir storage space. In the
absence of better information, the user may wish to restrict the drawdown rate to within
the range of 1-3 m/day.
h) Minimum flushing discharge. In this column, enter the minimum discharge through the
low-level outlets that will still result in successful flushing. No removal of sediment from the
reservoir's settled sediment storage will occur if this minimum flow rate criterion is not
satisfied, which will extend the number of total days the reservoir spends trying to meet the
flushing duration goal.

Wild et al. (2019)

SedSim User Manual

P a g e | 82

i) Reservoir bottom width (m). This is the representative bottom width of the reservoir. This
value is used in SedSim to determine the relative volume of the reservoir, which, when
compared to the volume of the channel formed during flushing, aids in determining the
quantity of sediment released during a flushing event. It is suggested to use the widest
section of the reservoir bottom close to the base of the dam, as a wider value will produce a
more conservative flushing result (a wider reservoir results in lower quantities of sediment
removed during flushing events).
This value cannot be different for each flushing date, so the value is imported from the first
row of flushing data.
j) Reservoir average side slope (m/m, 1 horizontal to SS res vertical). This is the representative
bank side slope of the reservoir. This value is used in SedSim to determine the relative
volume of the reservoir, which, when compared to the volume of the channel formed
during flushing, aids in determining the quantity of sediment released during a flushing
event.
This value cannot be different for each flushing date, so the value is imported from the first
row of flushing data.
k) Coefficient value, k, for sediment load generation during Flushing (kQm). Optional. Instead
of computing the sediment loads discharged during flushing via the methods described in
the Flushing section of Chapter 2, the user can instead specify parameters to be used in the
equation kQm to determine sediment discharge from the reservoir each day during flushing
as a function of reservoir outflow). Sediment discharge during a flushing day (kg) is
determined by kQm*Q(dt), where Q (m3/s) is the reservoir inflow and dt is the daily time
step in seconds. Entering information in this column will dictate the use of this method for
sediment flushing discharge, whereas leaving this column blank will dictate the use of the
long term capacity ratio for the flushing simulation.
l) Exponent value, m, for sediment load generation during Flushing (kQm). Optional. Instead
of computing the sediment loads discharged during flushing via the methods described in
the Flushing section of Chapter 2, the user can instead specify parameters to be used in the
equation kQm to determine sediment discharge from the reservoir each day during flushing
as a function of reservoir outflow). Sediment discharge during a flushing day (kg) is
determined by kQmQ*(dt), where Q (m3/s) is the reservoir inflow and dt is the daily time
step in seconds. Entering information in this column will dictate the use of this method for
sediment flushing discharge, whereas leaving this column blank will dictate the use of the
long term capacity ratio for the flushing simulation.
6.2.16. Sluicing
Wild et al. (2019)

SedSim User Manual

P a g e | 83

The following information must be supplied by the user for every reservoir at which sluicing will
be simulated. Some user inputs are described below as Optional, meaning these inputs are
extra features that are not required to run a simulation. The columns should be specified in
exactly the order (left to right) in which they are listed below.
Some comments on default sluicing assumptions. Before describing the user inputs
specifically, some comments are important regarding default assumptions. Except for the
sluicing start and stop dates each year, the user is only required to specify assumptions for the
categories above in the row corresponding to the first sluicing event. As long as preferences
(e.g., target drawdown elevation) are established for the first sluicing event, the model will
continue to use these values if the user neglects to specify preferences for future sluicing dates.
For categories including the sluicing inflow-based starting criteria, drawdown rate, and refill
rate, the model assumes the constraints do not exist if values of zero are specified for the first
sluicing event (or if cells are left blank). For the sluicing drawdown elevation, if no value is
specified the default is Elevation 0 masl. For power production during sluicing, if no value is
stored in the first row, then the model assumes a default that no power is produced during
sluicing.
a. Beginning date of sluicing. This is the first date on which drawdown for sluicing will begin.
Application of the Churchill curve to determine sediment trapping begins on this date. If
sluicing is to occur annually, the user must specify this date for every year in which sluicing
will occur. Sluicing is typically performed for an extended period of time when water and
sediment inflows are high (e.g., the monsoon season). To prevent the majority of inflowing
sediment from depositing in the reservoir, sluicing should be performed for as long as
possible during the season in which sediment production is highest.
Note: specifying no sluicing start date or duration will result in sluicing that can occur on any
given day (with no set start date or duration). In this case sluicing will be triggered by the
minimum reservoir inflow rate as specified below.
b. Sluicing starting criterion: minimum reservoir inflow rate (m3/s). Optional. This input is
optional. If the reservoir inflow is lower than this value on the sluicing beginning date, the
start of sluicing will be delayed by one day, and the condition will be checked again on the
next day. The purpose of this option is to allow the user to avoid initiation of drawdown for
sluicing during conditions that are not typical of high sediment inflows. For example, if the
monsoon season begins in October on average, the user might specify 10/1 as the sluicing
date every year. However, in a given year the monsoon might actually begin in November.
This optional input is thus an extra check that could hold sluicing off until the high inflows
actually begin in a given year. If sluicing is delayed, the original sluicing duration will be
maintained (as defined by the number of days between the user-specified beginning and
ending dates).
c. Sluicing Duration. Sluicing will last for this many days.
Wild et al. (2019)

SedSim User Manual

P a g e | 84

Note: specifying no sluicing start date or duration will result in sluicing that can occur on any
given day (with no set start date or duration). In this case sluicing will be triggered by the
minimum reservoir inflow rate as specified below.
d. Sluicing stopping criterion: minimum reservoir inflow (m3/s). Optional. This input is
optional. If the reservoir inflow is higher than this value on the sluicing end date, the end of
sluicing will be delayed by one day, and the condition will be checked again on the next day.
The purpose of this option is to allow the user to avoid ending sluicing if reservoir inflows
are high for longer than expected, in which case extending the duration of sluicing could be
beneficial from a sediment management perspective.
e. Target sluicing drawdown water surface elevation (mamsl) or storage (m3). This is the
reservoir water surface elevation or storage to which the reservoir will be drawn down to
begin the sluicing process. The extent to which the reservoir is drawn down affects the
energy slope and ultimately the percentage of sediment that will pass through the reservoir
during sluicing. In the absence of better information, the target elevation could be set to an
elevation at or slightly above the mid-level gates.
f. Maximum sluicing drawdown rate (m/d). Optional. This is the maximum rate at which the
water level of the reservoir can be drawn down per day during the drawdown phase of
sediment sluicing. Rapid drawdown (and/or refill) of a reservoir can lead to bank failure,
landslides, or similar events, in which large quantities of soil fall into the reservoir storage
space. In the absence of better information, the user may wish to restrict the drawdown
rate to within the range of 1-3 m/day.
g. Maximum sluicing refill rate (m/d). Optional. This is the maximum rate at which the water
level of the reservoir can be refilled per day upon the completion of sluicing. In the absence
of better information, the user may wish to restrict the refill rate to 4 m/day or less.
h. Does hydropower production occur during sluicing? Valid responses are either "Yes" or
"No". If sluiced sediment is of high concentration for a long duration, and/or contains
significant quartz content, hydropower-related infrastructure (e.g., turbines) can be
damaged due to abrasion if power is produced during sluicing operations.
6.2.17. Density Current Venting

The following information must be supplied by the user for every reservoir at which density
current (or turbidity current) venting will be simulated. Some user inputs are described below
as Optional, meaning these inputs are extra features that are not required to run a simulation.
The columns should be specified in exactly the order (left to right) in which they are listed
below.
Wild et al. (2019)

SedSim User Manual

P a g e | 85

a. Minimum venting efficiency (%). This represents the lowest acceptable percentage of
sediment removal that must occur for density current venting to be an attractive option
(e.g., 35%). In determining this value, the user should weigh the relative importance of
releasing sediment compared to the water that will be wasted during venting without
power production.
If the user does not specify a minimum venting efficiency, Figure 14.13 from Morris and Fan
(1998) is used to establish a default value based on the reservoir length (km), using the
following equation:
Min. Efficiency = 0.5384 - 0.08*ln(Reservoir Length)
b. Minimum reservoir water surface elevation (mamsl) during density current venting.
Optional. Density current venting will not be allowed to reduce the water surface elevation
below this specified elevation. The user should leave this cell blank if no minimum level
exists, in which case venting will proceed until completed. Note that if the reservoir’s water
surface elevation approaches this minimum elevation, there exists the possibility that flow
releases through the low-level outlets for venting may be significantly reduced (< Q in (t)) or
even entirely eliminated, because any releases in this circumstance are first allocated to
satisfying minimum hydropower production requirements.
c. Maximum concentration (mg/l) of sediment released from reservoir during density
current venting. Optional. This input is optional. If the sediment mass released during
density current venting divided by water volume released from the hydropower and low
level outlets during venting cannot achieve this target 'concentration' (total mass/total
volume), extra water will be released from the hydropower outlets (up to their capacity),
and if this still does not achieve the target concentration, more water will be released from
the mid-level outlets to attempt to achieve the target level. SedSim will not reduce the
release of the density current flow from the low-level outlets in order to satisfy this target
concentration, as such a practice could theoretically reduce the effectiveness of the density
current release.
d. Continue venting if max concentration exceeded? If the user specifies a maximum
concentration (mg/l) of sediment released from the reservoir during density current
venting, the reservoir will attempt to meet this target by releasing additional clear water
through available outlets, if necessary. The option described here allows the user two
options: (1) to continue density current venting despite the inability of the reservoir to
satisfactorily dilute the sediment released during the venting process (enable this option by
entering “Yes”), and (2) to end density current venting due to the inability of the reservoir
to satisfactorily dilute the sediment released during the venting process (enable this option
by entering “No”).
Wild et al. (2019)

SedSim User Manual

P a g e | 86

e. Reservoir bottom width (m). This input is used in Eq. (2.11) as part of the SedSim procedure
to determine venting efficiency during density current venting events.
f. Reservoir bed slope (m/m). This input is used in Eq. (2.11) as part of the SedSim procedure
to determine venting efficiency during density current venting events.
g. Minimum daily power requirement during density current venting (MW). Optional. During
density current venting, water will be released through available hydropower outlets to
satisfy this minimum power production requirement, subject to hydropower outlet
capacity-elevation constraints. While the SedSim approach to density current venting is to
attempt to release the reservoir’s inflow (minus evaporation) through the low-level outlets,
thus maintaining the water surface elevation, any additional releases to satisfy this
minimum power requirement will result in a more significant reduction in the water surface
elevation of the reservoir. If the user also specifies a minimum water surface elevation
during density current venting and the reservoir reaches this minimum elevation, priority is
first given to satisfying the minimum hydropower production requirements, after which
remaining releases are allocated to the low-level outlets for venting flow release.
6.2.18. General Sediment Removal

The following information must be supplied by the user for every reservoir at which the user
wishes to remove a specified sediment mass during a specified period of time without explicitly
simulating the process. Some user inputs are described below as Optional, meaning these
inputs are extra features that are not required to run a simulation. The columns should be
specified in exactly the order (left to right) in which they are listed below.
a) Removal start date. This is the calendar date on which to begin removing sediment mass.
b) Removal duration (days). Specify number of days over which the specified sediment mass
should be removed during each event. The sediment mass to be removed will be equally
distributed among this number of days.
c) Sediment mass removal (tons). Specify the sediment mass to be removed during each
event. This mass will be equally distributed over the removal duration.
d) Destination element name. Specify the user-defined (in the "Network connectivity"
worksheet) name of the element into which the removed sediment mass will be deposited.
If the user leaves this cell blank, the model's default assumption is to discharge this
sediment out of the modeled system (the modeled system permanently loses this mass).
e) Fraction of sediment removed from active storage zone. Specify the fraction of sediment
that is removed from the active storage zone when the sediment is removed from the
Wild et al. (2019)

SedSim User Manual

P a g e | 87

reservoir during each event. The remaining fraction of sediment is assumed to be removed
from the dead storage zone. If no value is entered, the model maintains the following
default percentages: 50% of removed sediment is removed from the active storage zone,
while the remaining 50% is removed from the dead storage zone.
6.2.19. Bypassing

The following three columns of information must be supplied to perform a sediment bypass
around a reservoir. The columns should be specified in exactly the order (left to right) in which
they are listed below.
1. Flow rate above which bypass is activated during wet season (m3/s). Specify the minimum
reservoir inflow rate at which the sediment bypass is opened during the wet season and
sediment and flow begins to be discharged around the reservoir. If the inflow rate is lower
than this value, the sediment and water will enter the reservoir without being bypassed
(except for any minimum required bypass flow rate as specified below).
2. Bypass discharge capacity (m3/s). Specify the flow capacity of the sediment bypass. If the
reservoir inflow rate exceeds this value, any inflow in excess of the bypass discharge
capacity will enter the reservoir. However, only a fraction of the sediment concentration in
this flow will enter the reservoir, whereas the remaining fraction of sediment will be
distributed into the bypass. This fraction must be specified by the user, as described below.
3. Fraction of sediment load in reservoir inflow. The model's default assumption is that
sediment is partitioned between the bypass and reservoir in proportion to the fractions of
total inflow that are distributed into the bypass and reservoir. The user should thus enter
nothing in this column if this is the desired assumption. Alternatively, the user can specify
what fraction of the sediment that would otherwise have entered the reservoir (based on
the proportion of total inflow that enters the reservoir) should instead be distributed into
the bypass. This option was implemented to reflect that concentration increases with depth
of flow, and thus the bypass may remove more of the inflowing sediment than just the
proportion of flow diverted into the bypass.
4. Minimum required bypass flow (m3/s). This is the minimum flow that is required to be
maintained in the bypass channel(s) at all times (when possible).
5. Minimum required bypass flow (fraction). Similar to the minimum bypass required flow,
except this value represents the fraction of the total daily site water inflow required to be
diverted into the bypass channel at all times.
6.2.20. IncFlowsCalibration1
Wild et al. (2019)

SedSim User Manual

P a g e | 88

The SedSim model will automatically create this worksheet if it is needed. This
worksheet is only required if (1) the user chooses to perform internal calibration of incremental
sediment load coefficients (i.e., selects the "Calibrate a coefficient for each incremental inflow
location" option in the preferences in the main .xlsm model file); or (2) the user chooses to use
coefficients that were previously calibrated within this worksheet (i.e., selects the "Use
coefficients calibrated in most recent simulation." option in the preferences in the main .xlsm
model file). Once the worksheet is created, do not modify the contents of this worksheet. The
worksheet contains time series of the daily average incremental flow rates (m3/s) that enter
every reach or reservoir location in the modeled system.
6.2.21. Calibration1

The SedSim model will automatically create this worksheet if it is needed. This
worksheet is only required if (1) the user chooses to perform internal calibration of incremental
sediment load coefficients (i.e., selects the "Calibrate a coefficient for each incremental inflow
location" option in the preferences in the main .xlsm model file); or (2) the user chooses to use
coefficients that were previously calibrated within this worksheet (i.e., selects the "Use
coefficients calibrated in most recent simulation." option in the preferences in the main .xlsm
model file). Once the worksheet is created, do not modify the contents of this worksheet. The
worksheet contains time series of the daily average incremental sediment loads that enter
every reach or reservoir location in the modeled system (in kg/day). The incremental flows,
Q(t), in the "IncFlowsCalibration 1" worksheet are used as input to the incremental load
generation function, cQ(t)d*Q(t)*Δt, to generate daily incremental sediment loads at the
location represented by each column. This worksheet is used as a calibration worksheet,
because the value of c in the sediment generation function is contained in a cell at the bottom
of the time series that is manipulated by Excel's LP solver until the average annual sediment
load matches the desired value. The desired value should be provided in the "Sediment Loads"
worksheet.
6.2.22. FlowsCalibration2

The SedSim model will automatically create this worksheet if it is needed. This
worksheet is only required if the user chooses either of the following two options in the
preferences in the main “SedSim.xlsm” file: (1) Calibrate a coefficient for each reach; or (2) Use
coefficients calibrated in most recent simulation. Once the worksheet is created, do not modify
the contents of this worksheet. The worksheet contains the time series of the daily average
flow rates out of each reach (m3/s) in the unregulated system. Thus, these flows simply
represent the sum of all incremental flows that enter points upstream of the location of
interest.
6.2.23. Calibration2

Wild et al. (2019)

SedSim User Manual

P a g e | 89

The SedSim model will automatically create this worksheet if it is needed. This
worksheet is only required if the user chooses either of the following two options in the
preferences in the main “SedSim.xlsm” file: (1) Calibrate a coefficient for each reach; or (2) Use
coefficients calibrated in most recent simulation. The worksheet contains time series of the
daily average sediment loads that flow out of every reach or reservoir location in the modeled
system (in kg/day). The outflows, Q(t), in the "FlowsCalibration2" worksheet are used as input
to the sediment load generation function, aQ(t)b*Q(t)*Δt, to generate daily sediment loads
flowing out of each location represented by each column. This worksheet is used as a
calibration worksheet, because the value of a in the sediment generation function is contained
in a cell at the bottom of the time series that is manipulated by Excel's LP solver until the
average annual sediment load matches the desired value. The desired value is determined
internally in the model by summing the values of the incremental sediment loads that are
generated at all points upstream of the outflow point of interest.

Wild et al. (2019)

SedSim User Manual

P a g e | 90

7. Runtime File Description
Users have the option of producing runtime output during the execution of a simulation. This
will print information to a text file as a simulation runs. Each row represents a new piece of information.
This information includes the current time step, which system elements (e.g., reservoirs) are being
simulated, which model sub-routines are being entered, and a confirmation of a completed simulation.
This file will not contain error messages. SedSim-generated error messages are instead printed to the
screen for users to view directly.
The default is for no runtime file to be created. This is because printing to the runtime file takes
time and the file can become large. If you want a runtime file to be produced, specify a desired
name/location within the SedSim.xlsm file (cell B6) as shown in Figure 2.2. You can either specify the
name of the file, in which case it will be saved in the same directory in which you are running
SedSim.xmlsm, or you can specify a full file path to any directory on your computer. Your runtime file
must be a text file, so always end your file name or path with a “.txt”.
An example of runtime file output is shown below:

Figure 7.1. An example of runtime output.

Wild et al. (2019)

SedSim User Manual

P a g e | 91

8. Output File(s) Description
8.1. Overview of Model Output Workbook
Output file(s) are automatically created during the simulation and saved (using the userspecified file location and name in the “SedSim.xlsm” file). As discussed in Chapter 5 (the
“SedSim.xlsm” file preferences), users have the option of creating up to two output files, or
none at all. Currently, the two output files users can create are (1) a time series output for a
variety of variables for all locations in the system to which the variables are applicable, and (2) a
statistical summary of the time series output for a variety of variable values at all locations
where they apply.
In the time series output file, each worksheet within the workbook corresponds to a
different variable. Each column of time series data corresponds to a different location, which is
listed at the top of each time series column. The number of variables (and system elements)
included in the two output files depends on whether the system being simulated is Regulated
or Unregulated. Fewer variables and locations are applicable in an unregulated simulation, as
no reservoirs are present. In the statistical output file, in general, each worksheet contains
output that corresponds to a unique combination of model variables and set of statistics taken
over a particular time period. One worksheet will contain the mean, standard deviation,
median, maximum and minimum of the values of a variable for each month at all system
locations, whereas another worksheet will contain the same statistical manipulations but on an
annual time scale. The automatically loaded worksheet labels, column and row labels, and
worksheet/variable descriptions (in the first row of every worksheet) should clarify the
organization of data in both files.
Worksheets in both files are color coded such that if the variable primarily relates to
water the tab color is blue, whereas if the variable primarily relates to sediment the tab color is
brown.
The model does not automatically generate figures (charts, graphs, tables, etc.) to
graphically summarize the data in Excel. However, the user can easily access Excel’s plotting
capabilities to prepare any desired plots.
Each of the output files will now be discussed separately. The time series output file
section provides a separate discussion of every worksheet included in the file, whereas the
statistical output file section provides a more general discussion of the worksheets included in
the file.

8.2. Time Series Output File

Wild et al. (2019)

SedSim User Manual

P a g e | 92

The following is a description of the information conveyed by each variable for which
there is a worksheet in the time series output file. Worksheet names appear below in bolded
text, although the actual worksheet names do not include the variable units listed below. The
first row of each output file worksheet contains a brief description of the information contained
in the worksheet, and those descriptions are given below to provide very brief summaries of
the information included in the time series output file. However, much more detailed
descriptions of the model output contained in each worksheet are provided immediately
afterward.
•
•
•
•
•
•
•
•
•

•
•

“Water Storage” Worksheet Description: Represents the total volume of water (m^3)
stored in a reservoir or reach at the end of each time period.
“Water Surface Elevation” Worksheet Description: Represents the elevation (mamsl)
associated with the water storage in each reservoir.
“Active Storage Volume” Worksheet Description: Represents the volume of water (m^3)
held within the active storage zone at the end of each time period.
“Dead Storage Volume” Worksheet Description: Represents the volume of water (m^3)
held within the dead storage zone at the end of each time period.
“Storage Volume Target Deviation” Worksheet Description: Represents the difference
(m^3), or error, between the storage target for the end of each time period and the
simulated reservoir storage at the end of each time period.
“Stor. Vol. Target Deviation (%)” Worksheet Description: Represents the % difference
(m^3), or error, between the reservoir storage target for the end of each time period
and the simulated reservoir storage at the end of each time period.
“Elevation Target Deviation” Worksheet Description: Represents the difference (mamsl),
or error, between the reservoir elevation (mamsl) target for the end of each time period
and the simulated reservoir elevation at the end of each time period.
“Elevation Target Deviation (%)” Worksheet Description: Represents the % difference, or
error, between the elevation (mamsl) target for the end of each time period and the
simulated reservoir elevation at the end of each time period.
“Active Storage Volume Capacity” Worksheet Description: Represents maximum
capacity (m^3) of a system element to store water within its active storage zone during
each time period. This value will not remain constant over time in a reservoir if
sediment volume accumulates in the reservoir.
“Act. Stor. Capacity Reduction” Worksheet Description: Represents the percentage (%)
reduction in size of the initial capacity of the active storage zone in each reservoir during
each time period.
“Dead Storage Volume Capacity” Worksheet Description: Represents the maximum
capacity (m^3) of a reservoir to store water in the dead storage zone during each time
period. The value of this variable for a particular reservoir is only different from the
initial value if sediment accumulates in the dead storage zone of the reservoir.

Wild et al. (2019)

SedSim User Manual

P a g e | 93

•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•

•

“Dead Stor. Capacity Reduction” Worksheet Description: Represents the percentage (%)
reduction in size of the initial capacity of the dead storage zone in each reservoir during
each time period.
“Flow_inflow” Worksheet Description: Represents water discharge (m^3/s) into a reach
or reservoir during each time period.
“Flow_outflow” Worksheet Description: Represents the water discharge (m^3/s) from
each element (reaches and reservoirs), not including evaporation.
“Storage_evaporation” Worksheet Description: Represents water evaporation rate
(m^3/s) at each reservoir site during each time period.
“Downstream Flow” Worksheet Description: Represents water discharge (m^3/s)
released from a reservoir during each time period that enters the reach immediately
downstream of the reservoir.
“Turbine Flow” Worksheet Description: Represents water discharge (m^3/s) released
from a reservoir during each time period through the hydropower outlet (turbines).
“Spilled Flow” Worksheet Description: Represents water discharge (m^3/s) released
from a reservoir during each time period that does not generate any power.
“Overflow Worksheet” Description: Represents water discharge (m^3/s) released from a
reservoir during each time period through the spillway outlet.
“Diversion Flow” Worksheet Description: Represents water discharge (m^3/s) released
from a reservoir during each time period through the diversion outlet.
“Controlled Flow” Worksheet Description: Represents water discharge (m^3/s) released
from a reservoir during each time period through the controlled outlet.
“Low level flow” Worksheet Description: Represents water discharge (m^3/s) released
from a reservoir during each time period through the low-level outlet.
“Power Production (MW)” Worksheet Description: Represents the power (MW)
generated at a hydropower dam during each time period.
“Energy Production (MWH)” Worksheet Description: Represents the energy (MWH)
generated at a hydropower dam during each time period.
“Suspended Sediment Mass Inflow” Worksheet Description: Represents the mass of
suspended sediment (kg) that enters a system element during one time period.
“Suspended Sediment Mass Outflow” Worksheet Description: Represents the mass of
suspended sediment (kg) that exits a system element during one time period.
“Trap Efficiency” Worksheet Description: Represents the trap efficiency (as a fraction)
for each reservoir in the system during each time period.
“Residence Time” Worksheet Description: Represents the residence time (years) of
water in each reservoir at the end of each time period.
“Settled Sediment Mass” Worksheet Description: Represents the sediment mass (kg)
held in bottom storage in the element of interest at the end of each time period. This
value can increase or decrease depending on whether scour or deposition is the
dominant process.
“Suspended Sediment Mass” Worksheet Description: Represents the mass (kg) of
sediment in suspension in a reach or reservoir during a time period.

Wild et al. (2019)

SedSim User Manual

P a g e | 94

•
•
•
•
•

“Total Sediment Mass” Worksheet Description: This variable represents the sum (kg) of
the bottom sediment mass and suspended sediment mass, which are defined in their
respective worksheets.
“Total Sediment Surplus Deficit” Worksheet Description: Represents the total sediment
(kg) that exists in a reach or reservoir at the end of each time period that is in excess (or
deficit) of the amount of sediment that existed in the element at the start of simulation.
“Flow_Junction” Worksheet Description: Represents the water discharge (m^3/s) at
each user-defined junction.
“Flow_bypass” Worksheet Description: Represents the water discharge (m^3/s) in the
reservoir bypass channel.
“Bypass Suspended Sediment Mass” Worksheet Description: Represents the suspended
sediment mass (kg) discharged into the reservoir bypass channel

The following is a more detailed discussion of the model output contained within each of the
time series output file worksheets. The worksheet names are listed in bold type.
1. Water Storage (m3). This variable represents the volume of water (m3) stored at the end of
every time period in a reservoir or reach.
2. Water Surface Elevation (mamsl). This variable represents the elevation (mamsl) associated
with the water storage (m3) in each reservoir in each time period. This is determined via
interpolation over the user-provided Elevation-Volume table. The reported elevation does
account for impact of accumulation of sediment volume on the water surface elevation.
3. Active Storage Volume (m3). This variable represents the volume (m3) of water held within
the active storage zone at the end of each time period. This value can vary over time.
However, active storage is limited to the active storage capacity, which is given by the
volume of water held between the low and full supply level elevations (the elevations
between which the reservoir is expected to be operated for hydropower generation). The
active storage capacity can also change over time, as will be discussed below.
4. Dead Storage Volume (m3). This variable represents the volume (m3) of water held within
the dead storage zone in each reservoir at the end of each time period. This value can vary
over time as the water storage in the reservoir varies (assuming the water level drops below
the lower limit of the active storage zone). However, dead storage is limited to the dead
storage capacity, which is given by the volume of water held below the low supply level
(below the bottom of the active storage).
5. Storage Volume - Target Deviation (m3). This variable represents the difference (m3), or
deviation, between the user-established reservoir storage volume target (m3) for the end of
each time period and the simulated reservoir storage at the end of each time period. This
variable is only applicable to a particular reservoir if the user is simulating hydrology

Wild et al. (2019)

SedSim User Manual

P a g e | 95

internally in the model and selects the option to operate the reservoir based on storage
volume targets (rather than storage elevation targets).
6. Storage Volume - Target Deviation (%). This variable represents the error between the
user-established reservoir storage target (m3) for the end of each time period and the
simulated reservoir storage at the end of each time period, represented as a % of the
storage target. This variable is only applicable to a particular reservoir if the user is
simulating hydrology internally in the model and selects the option to operate the reservoir
based on storage targets (rather than storage elevation targets).
7. Elevation Target Deviation (m3). This variable represents the difference (m3), between the
user-established reservoir elevation target (mamsl) for the end of each time period and the
simulated reservoir storage elevation at the end of each time period. This variable is only
applicable to a particular reservoir if the user is simulating hydrology internally in the model
and selects the option to operate the reservoir based on storage elevation targets (rather
than storage volume targets).
8. Elevation Target Deviation (%). This variable represents the difference between the userestablished reservoir storage elevation target (mamsl) for the end of each time period and
the simulated reservoir storage elevation at the end of each time period, represented as a
% of the storage elevation target. This variable is only applicable to a particular reservoir if
the user is simulating hydrology internally in the model and selects the option to operate
the reservoir based on storage elevation targets (rather than storage volume targets).
9. Active Storage Volume Capacity (m3). This variable represents maximum capacity (m3) of a
reservoir to store water within its active storage zone. This value is defined at the
simulation start date by the user in the "Reservoir Specifications" worksheet. This value will
not remain constant if sediment volume accumulates in the reservoir.
10. Active Storage Capacity Reduction (%). This variable represents the percentage reduction
in size of the initial capacity of the active storage volume zone in each reservoir. The active
capacity will only be reduced in size from the initial value if sedimentation in the active
storage volume zone occurs.
11. Dead Storage Volume Capacity (m3). This variable represents maximum capacity (m3) of a
reservoir to store water within its dead storage volume zone. This value is defined for the
simulation start date by the user in the "Reservoir Specifications" worksheet. This value will
not remain constant if sediment volume accumulates in the reservoir.
12. Dead Storage Capacity Reduction (%). This variable represents the percentage reduction in
size of the initial capacity of the dead storage zone in each reservoir. The dead capacity will
only be reduced in size from the initial value if sedimentation in dead storage zone occurs.
Wild et al. (2019)

SedSim User Manual

P a g e | 96

13. Flow_inflow (m3/s). This variable represents the water flow rate (m3/s) into a reach or
reservoir during each time period. The value reported for each time period is assumed to
remain constant for the duration of that time period.
14. Flow_outflow (m3/s). This variable represents the water flow rate (m3/s) out of a reach or
reservoir during each time period. The value reported for each time period is assumed to
remain constant for the duration of that time period. Note that for reservoirs, this value
represents the sum of the discharge from all of the reservoir's outlets during the time
period.
15. Flow_junction (m3/s). This variable represents the water flow rate at each user-defined
junction during each time period. The value reported for each time period is assumed to
remain constant for the duration of that time period.
16. Suspended sed. mass junction (kg). This variable represents the suspended sediment mass
(kg) entering each user-defined junction during each time period. The value reported for
each time period is assumed to remain constant for the duration of that time period.
17. Storage_evaporation (m3/s). This variable represents the water loss from evaporation
expressed as a flow rate (m3/s) for each reservoir during each time period. The value
reported for each time period is assumed to remain constant for the duration of that time
period.
18. Downstream flow (m3/s). This variable represents the water flow rate (m3/s) that enters
the reach (channel) or reservoir immediately downstream of the reservoir of interest each
time period. At diversion reservoirs, this value does not include the diverted flow rate. The
value reported for each time period is assumed to remain constant for the duration of that
time period. This variable is only reported for hydrologic simulations that are conducted
using the SedSim model.
19. Turbine flow (m3/s). This variable represents the water flow rate (m3/s) that is discharged
through the turbines of the hydropower plant at the reservoir in each time period. Turbine
flow is constrained by the discharge capacity of the outlet works that feed the turbines, and
by the maximum power (MW) capacity of the turbines. The value reported for each time
period is assumed to remain constant for the duration of that time period. This variable is
only reported for hydrologic simulations that are conducted using the SedSim model.
20. Flow_bypass (m3/s). This variable represents the water flow rate (m3/s) that is discharged
into every reservoir bypass, if a reservoir bypass exists. The remainder of flow is assumed to
enter the reservoir.
21. Spilled Flow (m3/s). This variable represents any water flow rate (m3/s) that is discharged
from the reservoir during each time period without contributing to hydropower production.
Wild et al. (2019)

SedSim User Manual

P a g e | 97

Any outlet that does not generate hydropower contributes flow to the spill rate. The value
reported for each time period is assumed to remain constant for the duration of that time
period. This variable is only reported for hydrologic simulations that are conducted using
the SedSim model. Spill flows include overflows.
22. Overflow (m3/s). This variable represents any water flow rate (m3/s) that is discharged from
the overflow (spillway) outlet of a reservoir each time period, which generally separates the
top of the active storage zone from the flood storage zone. The value reported for each
time period is assumed to remain constant for the duration of that time period. This
variable is only reported for hydrologic simulations that are conducted using the SedSim
model.
23. Diversion flow (m3/s). This variable represents any water flow rate (m3/s) that is discharged
from the diversion outlet during each time period. The value reported for each time period
is assumed to remain constant for the duration of that time period. This variable is only
reported for hydrologic simulations that are conducted using the SedSim model.
24. Controlled flow (m3/s). This variable represents any water flow rate (m3/s) that is
discharged from the controlled outlet of a reservoir during each time period. The value
reported for each time period is assumed to remain constant for the duration of that time
period. This variable is only reported for hydrologic simulations that are conducted using
the SedSim model.
25. Low level flow (m3/s). This variable represents any water flow rate (m3/s) that is discharged
from the low-level outlet of a reservoir during each time period. The value reported for
each time period is assumed to remain constant for the duration of that time period. This
variable is only reported for hydrologic simulations that are conducted using the SedSim
model.
26. Power Production (MW). This variable represents the hydropower production at each
reservoir during each time period. The value cannot exceed the user-supplied hydropower
plant capacity (in the "Reservoir Specifications" worksheet in the input data file).
27. Energy Production (MWH). This variable represents the energy production at each
hydropower reservoir during each time period.
28. Suspended Sediment Mass Inflow (kg). This variable represents the mass of suspended
sediment (kg) that enters a reach or reservoir during each time period.
29. Suspended Sediment Mass Outflow (kg). This variable represents the mass of suspended
sediment (kg) that exits a reach or reservoir during each time period.

Wild et al. (2019)

SedSim User Manual

P a g e | 98

30. Bypass Suspended Sediment Mass (kg). This variable represents the suspended sediment
mass (kg) contained in the flow that is discharged into each reservoir's sediment bypass, if a
sediment bypass exists for the reservoir.
31. Trapping Efficiency (fraction). This calculated value is the fraction of suspended sediment
that remains in the reservoir in a simulated time period. Available sediment includes
sediment that remains in suspension from a previous time period and sediment that enters
the reservoir from an upstream element during the time period.
32. Residence Time (years). This variable represents the residence time (years) of water in a
reservoir. This value currently is currently calculated based on the average storage (m3) over
the last 365 days, and the total reservoir outflow volume during the last 365 days (m3).
33. Settled Sediment Mass (kg). This variable represents the sediment mass (kg) that is held in
bottom storage in the element of interest. This includes sediment that has settled in the
dead and active storage zones. In reservoirs, the bottom sediment mass includes only mass
that has been trapped according to the Brune (1953) curve trap efficiency. The user may
specify that some bottom sediment mass exists in reservoirs at the beginning of simulation.
In reaches, a calibrated rating function such as appears in Eq. (2.3) is responsible for
determining the discharge of sediment from a reach. Any sediment that exists in suspension
(either as a remainder from the previous simulation period or as inflow from an upstream
reach or reservoir) in excess of the carrying capacity mass and that is not discharged from
the reach must settle in the reach. This settled mass contributes to the bottom sediment
mass. In reaches, this bottom sediment mass can be re-suspended in future time periods
when not enough sediment exists in suspension in a reach to satisfy the sediment mass
discharge specified by the calibrated carrying capacity function. Each reach begins with a
pre-specified, finite amount of initial sediment mass, and is therefore by definition
exhaustible. In reservoirs, a variety of sediment management practices are capable of
removing the settled sediment.
34. Suspended Sediment Mass (kg). This variable represents the mass (kg) of sediment in
suspension in a reach or reservoir. In reservoirs, suspended sediment mass is any mass that
is not trapped due to sedimentation or discharged from the reservoir outlets. The same is
true of reaches, in that suspended sediment is sediment that did not settle or get
discharged from the reach.
35. Total Sediment Mass (kg). This variable represents the sum (kg) of the bottom sediment
mass and suspended sediment mass, which were defined previously.
36. Total Sed. Surplus Deficit (kg). This variable represents the total sediment mass (kg) that
exists in a reach or reservoir that is in excess (or deficit) of the amount of sediment that
existed in the element at the beginning of simulation. As reservoirs do not begin with initial
sediment mass availability in the bottom sediment mass pool, the value of this variable is
Wild et al. (2019)

SedSim User Manual

P a g e | 99

always a surplus value in reservoirs, and is equal to the total sediment mass stored in the
reservoir. Conversely, reaches are initialized with a certain amount of sediment mass. For
this reason, the TS_surplus_deficit value is useful reach-related information to review,
because it indicates what total quantity of mass (including suspended and settled mass)
resides in a reach that is different from the value assumed to exist at the simulation start
date.

8.3. Statistics Output File
The statistics output file contains many more worksheets than the time series output
file. They are listed below together with a short description. Many of the worksheets are only
slightly different from one another. For this reason, every worksheet contained in this output
file will not be separately described. Instead, this section will provide a description of the six
different types of statistical manipulation of the time series data that are provided, and a
summary of which variables are included within of these six categories. Before the six
categories are presented, it will first be helpful to provide a more detailed description of the
first two worksheets that appear in this output file (assuming a regulated system simulation):
1. Power Reliability. For every reservoir, this worksheet provides the reliability for various
levels of hydropower production, where power reliability for a particular level of power
production is defined as the fraction of the simulation days during which power production
at the reservoir exceeded the power production level of interest. Reliability results for each
reservoir are shown for 50 different levels of power production, where each interval is
equally sized at 1/50 of the facility's power production capacity (MW).
2. Energy Reliability. For every reservoir, this worksheet provides the reliability for various
levels of hydropower facility energy production, where energy reliability for a particular
level of power production is defined as the fraction of the simulation days during which
energy production at the reservoir exceeded the energy production level of interest.
Reliability results for each reservoir are shown for 50 different levels of energy production,
where each interval is equally sized at 1/50 of the facility's energy production capacity
(MWH).
The following six categories of statistical calculations are applicable to many of the
worksheets. For some of the variables, only four of the six categories are applicable, whereas all
of the categories apply for a few of the variables. After the categories are presented, a list of
the variables to which each category applies is provided.
A. Annual Statistics. For every reach and reservoir, this worksheet presents the mean,
standard deviation, maximum, minimum, and median of all the daily variable values
contained within each year. For example, for a 50-year simulation, for every reservoir the

Wild et al. (2019)

SedSim User Manual

P a g e | 100

water inflow rate (m3/s) worksheet would include 50 mean annual flow rates, 50 standard
deviations, etc.
B. Annual Sum Statistics. For every reach and reservoir, this worksheet presents the annual
sum of all the daily variable values contained within each year. For example, for a 50-year
simulation, for every reservoir the water inflow volume (m3) worksheet would include 50
annual inflow volume values (one for each simulation year). In this case, the point of the
worksheet is to report the annual sum, which means statistics are not relevant. Note that
Sum-based manipulations of the time series are not appropriate for all variables. For
example, taking the sum of daily inflow rates over the course of one simulation year would
not be meaningful information.
C. Monthly Statistics. For every reach and reservoir, this worksheet presents the mean,
standard deviation, maximum, minimum, and median of all the daily variable values
contained within each month. For example, for a 50-year simulation, for every reservoir the
water inflow rate (m3/s) worksheet would include 12*50=600 mean monthly flow rates,
50*12 monthly standard deviations of flow rate, etc.
D. Monthly Sum Statistics. For every reach and reservoir, this worksheet presents the monthly
sum of all the daily variable values contained within each month. For example, for a 50-year
simulation, for every reservoir the water inflow volume (m3) worksheet would include
12*50 monthly inflow volume values (one for each simulation month). In this case, the
point of the worksheet is to report the monthly sum, which means statistics are not
relevant. Note that Sum-based manipulations of the time series are not appropriate for all
variables. For example, taking the sum of daily inflow rates over the course of one month in
a particular simulation year would not be meaningful information.
E. Mean Monthly Statistics. For every reach and reservoir, this worksheet presents the mean
of the mean, standard deviation, maximum, minimum, and median of all the daily variable
values contained within each month. For example, for a 50-year simulation, for every
reservoir the water inflow rate (m3/s) worksheet would include 12 means (each of the 12
taken over the 50 mean monthly flow rates for each month), 12 means of the standard
deviations of flow rates, etc.
F. Mean Monthly Sum Statistics. For every reach and reservoir, this worksheet presents the
mean, standard deviation, maximum, minimum and median of the monthly sum of all the
daily variable values contained within each month. For example, for a 50-year simulation,
for every reservoir the water inflow volume (m3) would include 12 (one for each month)
mean monthly inflow volume values (taken over the 50 sums), 12 (one for each month)
standard deviations of the monthly inflow volume values (taken over the 50 sums), etc.
Note that Sum-based manipulations of the time series are not appropriate for all variables.
For example, taking the sum of daily inflow rates over the course of one month in a
particular simulation year would not be meaningful information.
Wild et al. (2019)

SedSim User Manual

P a g e | 101

For categories A, C and E above, variables for which corresponding worksheets are provided are
the following:
i.
ii.
iii.
iv.
v.
vi.
vii.
viii.
ix.
x.
xi.
xii.
xiii.
xiv.
xv.
xvi.
xvii.

Water Storage
Suspended Sediment Mass
Settled Sediment Mass
Total Sediment Mass
Total Sediment Surplus Deficit
Outflow Rate
Inflow Rate
Outflow Volume
Inflow Volume
Sediment Mass Outflow
Sediment Mass Inflow
Water Surface Elevation
Trap Efficiency
Residence Time
Power
Energy
Spilled Flow

For categories B, D and F above, variables for which corresponding worksheets are provided are
the following:
vii.
viii.
ix.
x.
xi.

Outflow Volume
Inflow Volume
Sediment Mass Outflow
Sediment Mass Inflow
Energy

Wild et al. (2019)

SedSim User Manual

P a g e | 102

9. Assumptions, Limitations and Caveats
These comments apply to the application of the SedSim model to the Mekong Basin:
•

Modeling sediment even using 3D hydrodynamic models is a difficult task. Hydraulic
studies in laboratories typically are based on a much more detailed knowledge of
the distribution of sediment sizes, densities, and even sediment shapes. Thus to
think that these daily simulations using a 1D mass balance model are in some sense
accurate would be a mistake. We view this SedSim model as strictly a screening tool,
and even then the focus should be on the relative changes in sediment loads and
depositions rather than the actual ones. In addition, averages of these daily results
over longer time periods are probably more reliable than the daily results
themselves.

•

The way sediment is currently modeled does not distinguish between suspended
and bed loads, or among different sediment size classes. Once data availability
justifies the distinction between bed and suspended loads, the accuracy of the
predicted sediment load simulations could be improved.

•

The main purpose of this modeling exercise is to assess the impact of changes in the
hydrologic and sediment regimes on the ecosystems in the basin, especially in the
Tonle Sap and Delta regions. We need to be aware of the uncertainties of such
predictions even assuming our hydrologic and sediment inputs are without errors,
and not waste time and resources perfecting our sediment predictions if the added
precision does not improve the accuracy of the ecological predictions.

Wild et al. (2019)

SedSim User Manual

P a g e | 103

10.

References

Atkinson, E. 1996. The Feasibility of Flushing Sediment from Reservoirs, TDR Project R5839, Rep. OD 137.
HR Wallingford.
Brune, G.M. (1953). Trap Efficiency of Reservoirs. Transactions of the American Geophysical Union 34,
407-418.
Churchill, M.A. (1948). Discussion of “Analysis of Use of Reservoir Sedimentation Data,” by L.C.
Gottschalk, pp. 139-140. Proc. Federal Inter-Agency Sedimentation Conf., Denver, CO, USA.
Fan. J., 1986. "Turbid Density Currents in Reservoirs," Water International, 11 (3): 107-116.
Fan, J., 1991. "Density Currents in Reservoirs," Workshop on Management of Reservoir Sedimentation,
New Delhi.
Habib-ur-Rehman, M. A. Chaudhry and N. Akhtar, 2009. Assessment of Sediment Flushing Efficiency of
Reservoirs. Pakistan Journal of Science 61(3):181-187.
Kawashima, S., Johndrow, T.B., Annandale, G.W., and Shah, F., 2003. Reservoir conservation volume I:
the RESCON approach, economic and engineering evaluation of alternative strategies for managing
sedimentation in storage reservoirs. A contribution to promote conservation of water storage assets
worldwide, The International Bank for Reconstruction and Development /The World Bank, Washington,
DC, USA. 52 pp.
Kondolf, G.M. (1997). Hungry water: Effects of dams and gravel mining on river channels.
Environmental Management 21(4):533-551.
Kondolf, G.M., Alford, C., and Rubin, Z. (2011). Cumulative Effects of Tributary Dams on the Sediment
Loads and Channel Form in the Lower Mekong River: Progress Report through 30 September 2011.
Department of Landscape
Architecture and Environmental Planning, University of California,
Berkeley.
Kummu, M. and Varis, O. (2007). Sediment-related impacts due to upstream
Lower Mekong River. Geomorphology 85:275-293.

reservoir trapping, the

Kummu, M., Lu, X.X., Wang, J.J., Varis, O. (2010). Basin-wide sediment trapping efficiency of emerging
reservoirs along the Mekong. Geomorphology 119:181–197
Meybeck, M., Laroche, L., Durr, H.H., Syvitski J.P.M. (2003). Global variability of daily total suspended
solids and their fluxes in rivers. Global and Planetary
Change 39(1-2): 65-93.
Migniot, C. (1981). Erosion and Sedimentation in sea and river, La Pratique des sols et des fondations,
editions Le Moniteur 1981.

Wild et al. (2019)

SedSim User Manual

P a g e | 104

Milliman, J.D., Meade, R.H. (1983). World-wide delivery of river sediment to the oceans. Journal of
Geology, 91: 1-21.
Morehead, M.D., Syvitski, J.P., Hutton, E.W.H., Peckham, S.D. (2003). Modeling the temporal variability
in the flux of sediment from ungaged river basins. Global and Planetary Change, 39(1-2): 95-110.
Morris, G.L. and Fan, J. (1998). Reservoir Sedimentation Handbook, McGraw Hill, New York, USA.
Palmieri A., Shah F., Annandale G.W., Dinar A. 2003. Reservoir conservation volume I: the RESCON
approach economic and engineering evaluation of alternative strategies for managing sedimentation in
storage reservoirs. A contribution to promote conservation of water storage assets worldwide, The
International Bank for Reconstruction and Development /The World Bank, Washington, DC, USA. 101
pp.
Strand, R. I., and Pemberton, E. L., 1987. "Reservoir Sedimentation," In Design of Small Dams. U.S.
Bureau of Reclamation, Denver.
Vörösmarty, C.J., Meybeck, M., Fekete, B., Sharma, K., Green, P., Syvitski, J.P.M., 2003. Anthropogenic
sediment retention: major global impact from registered river impoundments. Global and Planetary
Change 39 (1–2), 169–190.
Walling, D.E., Webb, B.W. (1983). Patterns of Sediment Yield. In: Gregory, K.J. (Ed.), Background to
Pelaeohydrology. Wiley, New York, NY, pp. 69-100.
Walling, D.E. (2009). The Sediment Load of the Mekong River. In Ian Campbell (Ed.), Mekong:
Biophysical Environment of an International River Basin (113-142). New York, NY: Academic Press.
Washburn, E.W. (1928). International Critical Tables of Numerical Data, Physics, Chemistry and
Technology, National Research Council of USA. McGraw-Hill, New York.
White, W.R. (2000). “Flushing of Sediments from Reservoirs.” Thematic Review IV.5. World Commission
on Dams.
White, W. R. (2001). Evacuation of Sediments from Reservoirs. London: Thomas Telford.
Wild, T.B. and Loucks, D.P. (2014). Managing Flow, Sediment and Hydropower Regimes in the Sre Pok, Se
San and Se Kong Rivers of the Mekong Basin. Water Resour. Res., 50, 5141-57.
Wild, T.B., Loucks, D.P., Annandale, G.W., and Kaini, P. (2015a). Maintaining Sediment Flows through
Hydropower Dams in the Mekong River Basin. J. Water Resour. Plann. Manage, 142(1), 05015004.
Wild, T.B. and Loucks, D.P. (2015b). Mitigating Dam Conflicts in the Mekong River Basin. Chapter 2 in
Conflict Resolution in Water Resources and Environmental Management, edited by K.W. Hipel et al.,
Springer (Heidelberg), pp. 25-48.
Wild, T.B. and Loucks, D.P. (2015c). An Approach to Simulating Sediment Management in the Mekong
Wild et al. (2019)

SedSim User Manual

P a g e | 105

River Basin. Chapter 12 in Sediment Matters, edited by P. Heininger and J. Cullman, Springer
(Heidelberg), pp. 187-99.
Wild, T.B., Reed, P.M., Loucks, D.P., Mallen-Cooper, M., Jensen, E.D. (2018). Balancing Hydropower and
Ecological Impacts in the Mekong: Tradeoffs for Sambor Mega Dam. J. Water Resour. Plann. Manage.
DOI: 10.1061/(ASCE)WR.1943-5452.0001036.
Wild, T.B., Loucks, D.P. and Annandale, G.W. (in review). SedSim: A River Basin Simulation Screening
Model for Reservoir Management of Sediment, Water, and Hydropower. Journal of Open Research
Software.
Xue, Z., Liu, J.P., DeMaster, D., Nguyen, L.V., and Ta, T.K.O. (2010). Late Holocene Evolution of the
Mekong Subaqueous Delta, Southern Vietnam. Marine Geology, 269: 45-60.

Wild et al. (2019)

SedSim User Manual

P a g e | 106

Appendix A: Access to Model Software
When in the main model workbook, simultaneously pressing the Alt and F11 keys (AltF11) will open up the windows that allow access to the visual basic application software that
executes the SedSim model. This will open the Microsoft Visual Basic editor. Next, double-click
on the category titled "VBAProject (SedSim.xlsm)". You will then be required to enter the
password you have been given to access the model's source code. Next, click on the "Modules"
folder, and double click on "SedimentModel". If you wish to close the visual basic editor and
return to the SedSim model interface, press Alt-F11 again. As long as you keep the
“SedSim.xlsm” file open, you can continue to access the source code you have already opened
without needing to enter a password every time. The authors of this program suggest saving
the model as it is before making any changes, in case those changes do not perform as
expected. We are not going to be able to debug different versions of this program, and would
rather have users send us suggestions on what might be changed or added and we can try to do
that.

Figure A.0.1 A display of the beginning of the SedSim program VBA code resulting from selecting Alt + F11 when the
main model workbook is open.

Wild et al. (2019)

SedSim User Manual

P a g e | 107

Appendix B: Equations for Flow, Sediment and Hydropower Simulation
SedSim simulates the mass balance of water and sediment in each reach (or river
channel) and reservoir, as well as the hydropower production associated with the water
released during each time period.
For reaches and reservoirs, the inflows of water and sediment mass are determined in the same
way, by summing all of the inflows into the reach or reservoir of interest. In symbolic form, the
water and sediment mass balances are given by the following:
Water:
IQ(j,t) = Incremental inflow (m3/s) to reach j in period t

Qin(j,t) = total inflow (m3/s) from upstream reaches and/or reservoirs, diversions, and
incremental inflow into reservoir j in period t.
=
 outflow over all incoming reaches k to reach j +
reservoirs z to reach j +
 over all incoming d
inflows into reach j
= ∑𝑘𝑘 Q 𝑜𝑜𝑜𝑜𝑜𝑜 (k, t) + ∑𝑧𝑧 Q 𝑜𝑜𝑜𝑜𝑜𝑜 (z, t) + ∑𝑑𝑑 Q 𝑜𝑜𝑜𝑜𝑜𝑜 (d, t) + IQ(j,t)
Sediment:

IS(j,t) = Incremental mass (kg) of sediment to reach j in period t

SMin(j,t) = total inflow (kg) from upstream reaches and/or reservoirs, diversions, and
incremental inflow into reservoir j in period t.
=
 outflow over all incoming reaches k to reach j +
reservoirs z to reach j +
 over all incoming d
inflows into reach j
= ∑𝑘𝑘 SM𝑜𝑜𝑜𝑜𝑜𝑜 (k, t) + ∑𝑧𝑧 SM𝑜𝑜𝑜𝑜𝑜𝑜 (z, t) + ∑𝑑𝑑 SM𝑜𝑜𝑜𝑜𝑜𝑜 (d, t) + IS(j,t)
Next, assumptions specific to reaches and reservoirs are discussed, respectively.

Reach components
Water
Internal hydrologic simulation requires river reach flow routing. Note that sediment simulation
and flow simulation in reaches are completely separated in SedSim, in that the extent of
deposition/scour/concentration does not impact flow. Each reach is assumed to be similar to a
lake in which the outflow rate is a function of the storage (elevation) in the reach. In the
SedSim model the release of water from each reach in each time period is assumed to be
Wild et al. (2019)

SedSim User Manual

P a g e | 108

defined as a non-linear function of the total initial storage volume plus inflow minus losses in
each period minus the volume that would remain in ponds in the reach if all remaining water
were suddenly withdrawn. Letting S(t) be the initial storage volume (m3); I(t) the inflow volume;
L(t) the losses from seepage and evaporation; and PS the ponding storage, or the reach storage
volume below which no outflow occurs; the outflow Q out (t) (m3/s) in period t will be
Q out (t) =δ[S(t) – min(PS, S(t)) + I(t) – L(t)]γ

for all periods t

(1)

Parameters PS, δandγwill differ for each reach. Both δandγwill usually be less than 1 but
never outside the range from 0 to 1. If the value of the expression in brackets is less than 1,
then γ is assumed to be 1.
The final reach storage volume is:
S(t+1) = S(t) + I(t) – L(t) – Q out (t)Δt

for all periods t

(2)

Alternatively, the user can choose to specify that reach outflow equals reach inflow (steady
state).
Sediment
SM(j,t) = mass of sediment in reach j at beginning of period t
SM out (k,t) = sediment mass (kg) outflow from reach k in period t
= Min{a(j) * Q out (j,t)b(j) * Q out (j,t) * Δt, Sediment available in reach as bed or
suspended sediment}
where a(j) and b(j) are user-defined sediment carrying capacity constants for reach j.
SM(j,t) + SM in (j,t) – SM out (j,t) = SM(j,t+1)
Reservoir components
Flows:
The outflow rate (m3/s) from reservoir j in time period t, Q out (j,t), is given by the following
relationship:
Qout(j,t) = Min{Max{R ST (j,t),R Env (j,t)},K R (j,t),Max{0,(S(j,t) + V in (j,t) – Evap(j,t) * Δt)/ Δt }}}
where R ST (j,t) is the water release rate (m3/s) required to meet the storage target (m3) at
reservoir j at the end of time period t (or beginning of time period t+1); R Env (j,t) is the userWild et al. (2019)

SedSim User Manual

P a g e | 109

established minimum downstream environmental flow (m3/s) for reservoir j during time period
t; K R (j,t) is the capacity (m3/s) of the outlets at reservoir j to release flow during time period t;
and V in (j,t) is the water inflow volume (m3) to reservoir j during time period t; S(j,t) is the water
storage volume (m3) at reservoir j at the beginning of time period t; and Evap(j,t) is the
evaporation rate (m3/s) from reservoir j during time period t.
The storage target, S Target (j,t), represents the storage target (m3) in reservoir j to be met at the
end of period t (i.e., the storage target for the beginning of period t+1). This can be the user’s
pre-established storage target for the current date, the storage target corresponding to the
user’s pre-established water surface elevation target, or a storage target that is established
internally in SedSim to achieve sediment management-related goals (e.g. to initiate reservoir
drawdown for flushing).
Note that R ST (j,t) is defined as follows:
𝑠𝑠𝑠𝑠𝑠𝑠
(𝑗𝑗, 𝑡𝑡) - S Target (j,t)
R ST (j,t) = S(j,t) + (Q in (j,t) – Evap(j,t)) -𝑉𝑉𝑠𝑠𝑠𝑠𝑠𝑠

𝑠𝑠𝑠𝑠𝑠𝑠
where 𝑉𝑉𝑠𝑠𝑠𝑠𝑠𝑠
(𝑗𝑗, 𝑡𝑡) is the volume of sediment that settles, or is trapped, in reservoir j during time
period t. Note that the inflowing water volume (m3) in a time period is reduced by an amount
equal to the volume of sediment that settles, as SedSim assumes the estimation of water
volume flowing into the reservoir included suspended sediment in the estimation. Any
sediment volume that remains in suspension is technically part of the water volume until it
settles.

Evaporation (m3/s), or Evap(j,t), is given by the following relationship:
Evap(j,t) = (E m (j)/D m )A s (j,t)/86400
where E m (j) is the average monthly (for the corresponding month time period t is in)
evaporation depth (mm) at reservoir j; D m (j) is the number of days in the month time period t
is in; and A s (j,t) is the water surface area of reservoir j during time period t.
The water storage balance is given by the following:
𝑠𝑠𝑠𝑠𝑠𝑠
(𝑗𝑗, 𝑡𝑡)/86400 = S(j,t+1)
S(j,t) + (Q in (j,t) - Qout(j,t) - Evap(j,t)) * Δt - 𝑉𝑉𝑠𝑠𝑠𝑠𝑠𝑠

Every reservoir j begins with an initial total storage capacity, K(j), but this capacity declines as a
result of sediment accumulation. That is, the reservoir water storage capacity at the end of
each timer period, K(j,t+1), is given by the following:
K(j,t+1) = K(j,t) - SSM(j,t+1)/ρ

Wild et al. (2019)

SedSim User Manual

P a g e | 110

Where SSM(j,t+1) is the mass of sediment that has settled in the reservoir through the end of
time period t, and ρ is the user-defined density of deposited sediment (kg/m3). SSM can
increase as sediment settles in the reservoir, but can decrease if sediment is removed from the
reservoir via sediment management techniques.
Sediment
Under normal operating conditions, the sediment released from a reservoir is given by the
following
SM out (k,t) = sediment mass outflow from reservoir k in period t
= Q out (j,t)*C(j,t) + S rem (j,t) + S dc (j,t)
Where Q out (j,t) is the outflow rate from hydropower, spillway and mid-level outlets at reservoir
j during time period t; C(j,t) is the concentration of sediment stored in suspension in the
reservoir j during time period t, assumed to be uniform throughout the reservoir’s storage
space; S rem (j,t) is the sediment mass (kg) removed from reservoir j during time period t via
flushing or general sediment removal; and S dc (j,t) is the sediment mass discharged through lowlevel outlets at reservoir j during time periods t when density current venting is taking place.
The mass of sediment that has settled in the reservoir through the end of time period t,
SSM(j,t+1) is described by the following relationship:
SSM(j,t+1) = SSM(j,t) + TE(j,t) * SM in (j,t) - S rem (j,t)
where TE(j,t) = trapping efficiency of reservoir j in period t based on initial conditions, and
SM in (j,t) is the sediment mass (kg) inflow to reservoir j during time period t.
Note that the approach to computing TE(j,t) changes depending on whether the reservoir is
being operated normally or if reservoir operations-based sediment management techniques
are being performed (e.g., flushing, sluicing, bypassing and density current venting). See
previous sections of the user manual for the approach we take for each sediment management
technique.
Hydropower
HP(j,t) = Hydropower production (MW) at reservoir j in period t
= (9.81/1000) * e(j) * h(j,t) * Q out (j,t)
where h(t,j) is the head above the turbines at reservoir j in timer period t, and e(j) is the
efficiency (fraction) of the turbines at reservoir j, assumed not to vary over time.

Wild et al. (2019)

SedSim User Manual

P a g e | 111

The hydropower calculations assume there are no losses in water quantity or in hydraulic head
as the water is transmitted from the reservoir to the powerhouse.

Wild et al. (2019)

SedSim User Manual

P a g e | 112



---

Source Exif Data:

File Type                       : PDF
File Type Extension             : pdf
MIME Type                       : application/pdf
PDF Version                     : 1.5
Linearized                      : Yes
Author                          : TomWild
Company                         : Hewlett-Packard Company
Create Date                     : 2019:02:25 11:02:03-05:00
Modify Date                     : 2019:02:25 11:02:18-05:00
Source Modified                 : D:20190225155759
Language                        : EN-US
Tagged PDF                      : Yes
XMP Toolkit                     : Adobe XMP Core 5.6-c015 81.157285, 2014/12/12-00:43:15
Metadata Date                   : 2019:02:25 11:02:18-05:00
Creator Tool                    : Acrobat PDFMaker 15 for Word
Document ID                     : uuid:07a97e90-9061-483c-b440-bd9a1cacd5eb
Instance ID                     : uuid:3f4ccbd2-5249-40b7-ac62-d8a40195f267
Subject                         : 768
Format                          : application/pdf
Title                           : 
Creator                         : TomWild
Producer                        : Adobe PDF Library 15.0
Page Layout                     : OneColumn
Page Count                      : 112

EXIF Metadata provided by EXIF.tools