Model setup

GeoSWMM model can be prepared by adding treatment to the runoff controls placed on the study site is the Tutorial_06_Initial.gdb. This geodatabase already contains the local LIDs defined in Tutorial 05 and the redesigned storage unit and outlet structures described in Description of Study Area. The land uses in geodatabase were re-calculated and re-assigned to each subcatchment based on the discretization employed in Tutorial 05. The washoff function for TSS is the same as EMC washoff function used in Tutorial 06. The land uses assigned to each subcatchment are listed in Table 2.5. The subcatchments treated by infiltration trenches or filter strips are marked in grey in this table. Figure 2.3 illustrate the model of Initial scenarios in GeoSWMM in ArcGIS Pro.

Both the Geodatabase and Model Object Panel contain network element information (e.g. rain gages, subcatchments, junction, conduits, orifice, weir, storage unit and outfalls). In addition, model object panel contains non-visual object information (e.g. time series, treatment and transect data).

Before the models are simulated, provide (or check) the following input data.

Rain Gage Properties

A rain gage provides precipitation or rainfall data to a GeoSWMM model. In this tutorial, rainfall in the study area is measured at a Rain gage. The Property Editor of this gage is shown below.

A 2-hour synthetic storm event with three different return periods i.e. 2-year, 10-year and 100-year has been assigned as the rainfall Data Source. To set rainfall data in the rain gage, the type of Data Source (e.g. time series or external file) and the Series Name need to be assigned. For tutorial, in the above figure, the Data Source is specified as TIMESERIES and the Series Name is specified as 2yr. This series is created under Time Series block in the Model Object Panel (see Time Series Data for details). To learn more on rainfall data types that can be assigned GeoSWMM, review the User’s Manual.

In GeoSWMM, every subcatchment must be linked to a rain gage for the model to run. However, no rainfall should be applied directly to subcatchments representing LID features like filter strips and infiltration trenches, since they are considered part of their parent subcatchments. To address this, a special Time Series named "Null" is created with zero rainfall values. A corresponding Rain Gage, also named "Null", is linked to this series and assigned to all LID subcatchments. Meanwhile, the main runoff-generating subcatchments remain connected to the actual rain gage (e.g., "Rain Gage") used in the model.
Moreover, two new hyetographs (same as Tutorial 06) are defined in GeoSWMM using its Time Series Editor. The names of the new rainfall series will be 0.1-in and 0.19-in, respectively. They will be used by the model’s single rain gage in addition to the 2-yr, 10-yr and 100-yr storms.

Subcatchment Properties

The study area is divided into sixteen subcatchments, each with a unique mix of land use types. Table 2.6 provides a summary of these land uses, with subcatchments treated by infiltration trenches or filter strips highlighted in grey. Figure 2.5 displays the Land Uses Editor used to define these classifications.

Other hydrologic properties (e.g. roughness, depression storage etc.) can be found in the Property Editor of these subcatchments. Most of these properties are calculated using the Tools of GeoSWMM. These properties are the same as Tutorial 05. For details on how to estimate these physical and hydrologic properties using GeoSWMM, see the User’s Manual.

Table 2.6 : Area and Land Uses for LID Subcatchments

Outfall Properties

The entire study catchment drains to the FREE type outfall O1 under both scenarios. Invert elevation of this outfall is 385.12 feet.

Time Series Data

Three rainfall time series datasets are provided under Time Series block in the Model Object Panel. These datasets represent a 2-hour synthetic storm event with three return periods i.e. 2-year, 10-year and 100- year as well as two new rainfall series namely 0.1-in and 0.19-in. Figure 2.6 shows the distribution of rainfall over 2 hours duration for the abovementioned storms.

Rainfall time series are assigned as a Rain Gage property in three storm event scenarios. The Rain Format is Intensity with Time Interval of 5 minutes. The Time Series Editor and the View options should appear as in Figure 2.7.

Additionally, under Time Series block, users can create, import or edit time series data for any object (e.g. node inflow). For details on working with time series data in GeoSWMM, see the User’s Manual.

LID Treatment

It is assumed that each filter strip and infiltration trench can provide 70% TSS removal for the runoff that passes over it. This is a typical removal observed for infiltration-based LIDs (Sansalone and Hird, 2003). A new land use, named “LID” is created with no TSS buildup function, with an EMC TSS washoff coefficient with 0 mg/L of TSS, and a TSS BMP efficiency of 70%. The land use assignment for each of the LID subcatchments (Highlighted in Table 2.6) is set to 100% LID. As a result, all of the runoff generated from upstream subcatchments that flow over these LID subcatchments receives 70% TSS removal. Table 2.7 illustrates the properties and input of the LID land use type.

Table 2.7 : Properties and values of LID in Land Use Editor

Detention Pond Treatment

The removal of TSS in the detention pond is simulated using the exponential model given by Equation 2.4. One can roughly estimate what the removal constant k in this expression must be so that a targeted level of pollutant removal is achieved within a 40-hour detention time for the 0.19 in. WQCV design storm. If Equation 2.3 were applied over a 40-hour period to achieve a target TSS reduction of 95%, then an estimate of k would be:

Where, is some representative value of the pond depth during the 40-hour release period. As can be verified later on, the average depth in the pond for the 0.19 in design storm over a 40-hour duration is 0.02 ft. Using this value in the expression for k yields an estimate of 0.01 ft/hr. This value is of the same order as the 0.03 ft/hr figure quoted in US EPA (1986) that represents the 20-th percentile of settling velocity distributions measured from 50 different runoff samples from seven urban sites in EPA’s Nationwide Urban Runoff Program (NURP).

Considering Equation (2.3), with the value k of 0.01 ft/hr and assuming a minimum residual TSS concentration C* of 5 mg/L, the following expression is entered into SWMM’s Treatment Editor for the storage unit SU1:

C = 5+ (TSS – 5) * EXP(-0.01 / 3600 / DEPTH * DT)

Where,

• 5 is the value assumed for C*
• TSS is the identifier given to the TSS concentration C for this model.
• 0.01/3600 is the value of k expressed in units of ft/s.
• DEPTH is the reserved word that SWMM uses for the water depth d in feet.
• DT is the reserved word that SWMM uses for the routing time step t in seconds.

N.B: When SWMM sees reserved words like DEPTH and DT within a treatment expression it knows to automatically insert their current values into the expression at each time step.

Loss Parameters

In a catchment hydrologic process, major water losses accounted are infiltration and evapotranspiration. To account for Infiltration loss from the subcatchments, Horton model has been applied. Parameter values used in the models are listed in Table 2.8. Horton’s parameters of filter strips and infiltration trenches are the same as Tutorial 06.

Table 2.8 : Horton Infiltration Model Parameters

Note that evapotranspiration and other loss properties are not assigned. For details on these loss parameters, review the User’s Manual.

Simulation Options Setting

The Options block in the Model Object Panel enables simulation settings. There are five tabs in Options editor. In this tutorial, a 48-hour long simulation has been carried out at 1 minutes Time Step to compare the peak-runoffs for both undeveloped and developed scenarios. Table 2.9 lists the primary simulation settings.