EPA SWMM GUI
5.3.0
Tutorials and User Guide
|
This tutorial provides an introduction to using EPA SWMM, Version 5, for modeling the quantity and quality of stormwater runoff produced from urban areas. The topics to be covered include:
In this tutorial we will model the drainage system serving a 12 acre residential area. The system layout is shown below and consists ofsubcatchment areas S1 through S3 Note, storm sewer conduits C1 through C4, and conduit junctions J1 through J4. The system discharges to a creek at the point labeled Out1. We will first go through the steps of creating the objects shown in this diagram on SWMM's Study Area Map and setting the various properties of these objects. Then we will simulate the water quantity and quality response to a 3-inch, 6-hour rainfall event, as well as a continuous, multi-year record.
You can click the View Map button that appears in each topic's header panel to refer to this drawing at any time. Use the [button_next] button to move to the next topic, the [button_prev] button to return to the previous topic, and the [button_main] button to return to the start of the tutorial. Side notes have been added to many of the topics that describe additional features of EPA SWMM. These can be viewed in a pop-up window by clicking on the word Note where it appears.
Our first task is to create a new project in EPA SWMM and make sure that certain default options are selected. Using these defaults will simplify the data entry tasks later on.
Rain Gages: | Gage |
Subcatchments: | S |
Junctions: | J |
Outfalls: | Out |
Conduits: | C |
ID Increment: | 1 |
This will make EPA SWMM automatically label new objects with consecutive numbers following the designated prefix.
Area: | 4 |
Width: | 400 |
% Slope: | 0.5 |
% Imperv: | 50 |
N-Imperv: | 0.01 |
N-Perv: | 0.10 |
Dstore-Imperv: | 0.05 |
Dstore-Perv: | 0.05 |
Zero-Imperv: | 25.0 |
Infil. Model <click to edit> | |
Method: | Modified Green-Ampt |
Suction Head: | 3.5 |
Conductivity: | 0.5 |
Initial Deficit: | 0.26 |
Node Invert: | 0 |
Node Max. Depth: | 4 |
Conduit Length: | 400 |
Conduit Geometry: <click to edit> | |
Shape: | Circular |
Max. Depth: | 1.0 |
Barrels: | 1 |
Conduit Roughness: | 0.01 |
Flow Units: | CFS |
Link Offsets: | DEPTH |
Routing Model: | Kinematic Wave |
Next we will set some map display options so that ID labels and symbols will be displayed as we add objects to the study area map, and links will have direction arrows.
Before placing objects on the map we should set its dimensions.
Finally, look in the status bar at the bottom of the main window and check that the Auto-Length feature is off. If it is on, then click the down arrow button and select "Auto-Length: Off" from the popup menu that appears. Also make sure that the Offsets option is set to Depth. If set to Elevation then click the down arrow button and select "Depth Offsets" from the popup menu that appears.
We are now ready to begin adding components to the Study Area Map. We will start with the subcatchments. Remember that you can click the View Map button of this tutorial at any time to see how we want our map to look eventually. Note[^1]
Observe how sequential ID labels are generated automatically as we add objects to the map.
[1] Drawing objects on the map is just one way of creating a project. For large projects is will be more convenient to first construct an EPA SWMM project file external to the program. The project file is a text file that describes each object in a specified format as described in the Users Manual. Data extracted from various sources, such as CAD drawings or GIS files, can be used to create the project file.
[2] If you right-click (or press Enter) after adding the first point of a subcatchment's outline, the subcatchment will be shown as just a single point.
Next we will add in the junction nodes and the outfall node that comprise part of the drainage network.
Now we will add the storm sewer conduits that connect our drainage system nodes to one another. (You must have created a link's end nodes as described in the previous topic before you can create the link.) We will begin with conduit C1 which connects junction J1 to J2.
Repeat steps 2 and 3 for conduits C2 through C4. Note[^1]
[1] Although all of our conduits were drawn as straight lines, it is possible to draw a curved link by left-clicking at intermediate points where the direction of the link changes before clicking on the end node.
To complete the construction of our study area schematic we need to add a rain gage.
At this point we have completed drawing the example study area. Your system should look like the one seen by pressing the View Map button above. If the rain gage, subcatchments or nodes are out of position you can move them around by
To re-shape a subcatchment's outline:
This same procedure can also be used to re-shape a link.
As visual objects are added to our project, SWMM assigns them a default set of properties. To change the value of a specific property for an object we must select the object into the Property Editor (shown below). There are several different ways to do this. If the Editor is already visible then you can simply click on the object or select it from the Project Browser. If the Editor is not visible then you can make it appear by one of the following actions:
- double-click the object on the map |
|
- right-click on the object and select Properties from the pop-up menu that appears | |
- select the object from the Project Browser and then click the Browser's button. |
Whenever the Property Editor has the focus you can press the F1 key to obtain a more detailed description of the properties listed.
Two key properties of our subcatchments that need to be set are the rain gage that supplies rainfall data to the subcatchment and the node of the drainage system that receives runoff from the subcatchment. Since all of our subcatchments utilize the same rain gage, Gage1, we can use a shortcut method to set this property for all subcatchments at once:
To set the outlet node of our subcatchments we have to proceed one by one, since these vary by subcatchment:
Finally, we wish to represent area S3 as being less developed than the others. Select S3 into the Property Editor and set its % Imperviousness to 25.
The junctions and outfall of our drainage system need to have invert elevations assigned to them. As we did with the subcatchments, select each junction individually into the Property Editor and set its Invert Elevation to the value shown in the table below. Note[^1]
Node | Invert |
J1 | 96 |
J2 | 90 |
J3 | 93 |
J4 | 88 |
Out1 | 85 |
Only one of the conduits in our example system has a non-default property value. This is conduit C4, the outlet pipe, whose diameter should be 1.5 ft. instead of 1 ft. To change its diameter, select conduit C4 into the Property Editor and set the Max. Depth value to 1.5.
[1] An alternative wau to move from one object of a given type to the next in order (or to the previous one) in the Property Editor is to hit the Page Down (or Page Up) key.
In order to provide a source of rainfall input to our project we need to set the rain gage properties. Select Gage1 into the Property Editor and set the following properties:
Rain Format: | INTENSITY |
Rain Interval: | 1:00 |
Data Source: | TIMESERIES |
Series Name: | TS1 |
As mentioned earlier, we want to simulate the response of our study area to a 3-inch, 6-hour design storm. A time series named TS1 will contain the hourly rainfall intensities that make up this storm. Thus we need to create a time series object and populate it with data. To do this:
0 | 0 |
1 | 0.5 |
2 | 1.0 |
3 | 0.75 |
4 | 0.5 |
5 | 0.25 |
6 | 0 |
[1] The time series editor can also be launched directly from the Rain Gage Property editor by selecting the editor's Series Name field and double clicking on it.
[2] Leaving off the dates for a time series means that SWMM will interpret the time values as hours from the start of the simulation. Otherwise, the time series follows the date/time values specified by the user.
Having completed the initial design of our example project it is a good idea to give it a title and save our work to a file at this point. To do this:
The project data is saved to the file in a readable text format. You can view what the file looks like by selecting Project | Details from the main menu. To open our project at some later time, we would select the Open command from the File menu.
Before analyzing the performance of our example drainage system we need to set some options that determine how the analysis will be carried out. To do this:
We are now ready to run the simulation. To do so, select Project | Run Simulation on the main menu (or simply click the button).
If there was a problem in running the simulation, a Status Report will appear describing what errors occurred.
Upon successfully completing a run, there are numerous ways in which to view the results of the simulation. We will illustrate just a few here:
The Status Report contains useful information about the quality of a simulation run, including a mass balance on rainfall, infiltration, evaporation, runoff, and inflow/outflow for the conveyance system. To view the report, select Report | Status (or click the button on the Main Toolbar and then select Status Report from the drop down menu).
For the system we just analyzed the report indicates the quality of the simulation is quite good, with negligible mass balance continuity errors for both runoff and routing (-0.39% and 0.03%, respectively, if all data were entered correctly). Also, of the 3 inches of rain that fell on the study area, 1.75 infiltrated into the ground and essentially the remainder became runoff.
The Summary Report contains tables listing summary results for each subcatchment, node and link in the drainage system. Total rainfall, total runoff, and peak runoff for each subcatchment, peak depth and and hours flooded for each node, and peak flow, velocity, and depth for each conduit are just some of the outcomes included in the summary report.
To view the Summary Report select Report | Summary from the main menu (or click the button on the Main Toolbar and then select Summary Report from the drop down menu). The report's window has a drop down list from which you select a particular report to view. For our example, the Node Flooding Summary table indicates there was internal flooding in the system at node J2. Note[^1] The Conduit Surcharge Summary table shows that Conduit C2, just downstream of node J2, was at full capacity and therefore appears to be slightly undersized.
[1] In EPA SWMM, flooding will occur whenever the water surface at a node exceeds the maximum defined depth. Normally, such water will be lost from the system. The option also exists to have this water pond atop the node and be re-introduced into the the drainage system when capacity exists to do so.
Simulation results (as well as some design parameters, such as subcatchment area, node invert elevation, link maximum depth) can be viewed in color-coded fashion on the study area map. To view a particular variable in this fashion:
To generate a time series plot of a simulation result:
For our example, this dialog can be used to graph the flows in conduits C1 and C2 as follows:
After a plot is created you can:
SWMM can generate profile plots showing how water surface depth varies across a path of connected nodes and links. Let's create such a plot for the conduits connecting junction J1 to the outfall Out1 of our example drainage system. To do this:
As you move through time using the Map Browser or with the Animator control, the water depth profile on the plot will be updated. Observe how node J2 becomes flooded between hours 2 and 3 of the storm event.
The appearance of a profile plot can be customized or it can be copied or printed using the same procedures as for a time series plot.
In the analysis just run we chose to use the Kinematic Wave method of routing flows through our drainage system. This is an efficient but simplified approach that cannot deal with such phenomena as backwater effects, pressurized flow, flow reversal, and non-dendritic layouts. SWMM also includes a Dynamic Wave routing procedure that can represent these conditions. This procedure, however, requires more computation time, due to the need for smaller time steps to maintain numerical stability.
Most of the effects mentioned above would not apply to our example. However we had one conduit, C2, that flowed full and caused its upstream junction to flood. It could be that this pipe is actually being pressurized and could therefore convey more flow than was computed using Kinematic Wave routing. We would now like to see what would happen if we apply Dynamic Wave routing instead.
To run the analysis with Dynamic Wave routing:
If you look at the Status Report for this run you will see that there is no longer any flooding and that the peak flow carried by conduit C2 has been increased from 3.52 cfs to 4.04 cfs.
[1] Normally when running a Dynamic Wave analysis, one would also want to reduce the routing time step (on the Time Steps page of the dialog) and perhaps select the Variable Time Step option (on the Dynamic Wave page of the dialog) as well.
In the next phase of this tutorial we will add water quality analysis to our example project. SWMM has the ability to analyze the buildup, washoff, transport and treatment of any number of water quality constituents. The steps needed to accomplish this are:
We will now apply each of these steps, with the exception of number 5, to our example project. Note[^1]
[1] Aside from direct runoff, EPA SWMM allows pollutants to be introduced into the nodes of a drainage system through: user defined time series of direct inflows, dry weather inflows, groundwater interflow, or rainfall derived inflow/infiltration.
For this tutorial example we will define two runoff pollutants; total suspended solids (TSS), measured as mg/L, and total Lead, measured in ug/L. In addition, we will specify that the concentration of Lead in runoff is a fixed fraction (0.25) of the TSS concentration. To add these pollutants to our project:
In SWMM, pollutants associated with runoff are generated by specific land uses assigned to subcatchments. In our example, we will define two categories of land uses: Residential and Undeveloped. To add these land uses to the project:
Next we need to define buildup and washoff functions for TSS in each of our land use categories. Functions for Lead are not needed since its runoff concentration was defined to be a fixed fraction of the TSS concentration. Normally, defining these functions requires site specific calibration.
In this example we will assume that suspended solids in Residential areas builds up at a constant rate of 1 pound per acre per day until a limit of 50 lbs per acre is reached. For the Undeveloped area we will assume that buildup is only half as much. For the washoff function, we will assume a constant event mean concentration of 100 mg/L for Residential land and 50 mg/L for undeveloped land. To define these functions for the Residential land use:
Now do the same for the Undeveloped land use category, except use a maximum buildup of 25, a buildup rate constant of 0.5, a buildup power of 1, and a washoff EMC of 50.
The final step in our water quality example is to assign a mixture of land uses to each subcatchment area:
Before we simulate the runoff quantities of TSS and Lead from our study area, an initial buildup of TSS should be defined so it can be washed off during our single rainfall event. We can either specify the number of antecedent dry days prior to the simulation or directly specify the initial buildup mass on each subcatchment. We will use the former method:
Now run the simulation by selecting Project | Run Simulation or by clicking on the Main Toolbar.
When the run is completed, view its Status Report. Note that two new sections have been added for Runoff Quality Continuity and Quality Routing Continuity. From the Runoff Quality Continuity table we see that there was an initial buildup of 47.5 lbs of TSS on the study area and an additional 2.2 lbs of buildup added during the dry periods of the simulation. About 48 lbs were washed off during the rainfall event. The quantity of Lead washed off is a fixed fraction (0.25 times 0.001 to convert from mg to ug) of the TSS as was specified.
If you plot the runoff concentration of TSS for subcatchment S1 and S3 together on the same time series graph you will see the difference in concentrations resulting from the different mix of land uses in these two areas. You can also see that the duration over which pollutants are washed off is much shorter than the duration of the entire runoff hydrograph (i.e., 1 hour versus about 6 hours). This results from having exhausted the available buildup of TSS over this period of time.
As a final exercise in this tutorial we will demonstrate how to run a long term continuous simulation using a historical rainfall record and how to perform a statistical frequency analysis on the results. The rainfall record will come from a file named sta310301.dat which was included with the example data sets provided with EPA SWMM. Note[^1] It contains several years of hourly rainfall beginning in January 1998. The data are stored in the National Climatic Data Center's DSI 3240 format which SWMM can automatically recognize.
To run a continuous simulation with this rainfall record:
[1] The example data sets can be found in My Documents | EPA SWMM Projects | Sample Projects
After our continuous simulation is completed we can perform a statistical frequency analysis on any of the variables produced as output. For example, to determine the distribution of rainfall volumes within each storm event over the two-year period simulated:
Object Category: | System |
Variable Analyzed: | Precipitation |
Event Time Period: | Event-Dependent |
Statistic: | Total |
Event Thresholds: | |
Rainfall | 0 |
Event Volume | 0 |
Inter-Event Hours | 6 |
This will identify the rainfall volume from each event which is separated by 6 or more hours without rainfall.
The results of this request will be a Statistics Report form containing four tabbed pages: a Summary page, a page containing a rank-ordered listing of each event, a page containing a histogram of the occurrence frequency versus event magnitude, and a Frequency Plot page that plots event magnitude versus cumulative frequency.
The Summary page shows that there was a total of 213 rainfall events. The Events page shows that the largest rainfall event had a volume of 3.35 inches and occurred over a 24- hour period. There were no events that matched the 3-inch, 6-hour design storm event used in our previous single-event analysis which had produced some internal flooding. In fact, the status report for this continuous simulation indicates that there were no flooding or surcharge occurrences over the simulation period.
We have only touched the surface of SWMM's capabilities. Some additional features of the program that you will find useful include:
You can find more information on these and other features in the SWMM User's Manual.
HAPPY SWMMING!