EPA SWMM Water Quality Modeling: Pollutant Buildup, Washoff, and BMP Treatment

Water quality modeling in EPA SWMM enables engineers to predict pollutant loads from urban catchments and evaluate the effectiveness of stormwater best management practices (BMPs). This tutorial covers the fundamentals of SWMM’s water quality routines, including pollutant definition, buildup and washoff processes, treatment functions, and modeling common pollutants like total suspended solids (TSS), nutrients, and heavy metals.

SWMM Water Quality Framework

SWMM’s water quality module tracks pollutant mass through three main processes:

1. Buildup - Accumulation of pollutants on land surfaces during dry periods
2. Washoff - Removal of accumulated pollutants by stormwater runoff
3. Treatment - Removal of pollutants by BMPs, LIDs, and storage facilities

The model routes pollutant mass through the drainage network using the same kinematic or dynamic wave routing used for flow.

Defining Pollutants

Pollutant Properties

Each pollutant in SWMM requires the following properties:

Common Stormwater Pollutants

Typical pollutants modeled in urban stormwater include:

Pollutant Buildup Functions

SWMM offers four buildup function types that describe how pollutants accumulate on surfaces during dry weather.

Power Function

Where B is buildup mass per area, t is antecedent dry days, C_1 is the buildup rate constant, and C_2 is the time exponent.

Exponential Function

Where B_max is the maximum buildup and k is the buildup rate constant. This is the most commonly used function.

Saturation Function

Where C_3 is the half-saturation constant (days to reach half of maximum buildup).

External Time Series

Buildup can also be specified directly using a time series, useful for modeling known pollutant loading schedules (e.g., road salting).

Recommended Buildup Parameters

Pollutant Washoff Functions

Washoff functions describe how pollutants are removed from surfaces and enter runoff during storm events.

Exponential Washoff

Where W is washoff rate (mass/time), q is runoff rate per unit area, B is remaining buildup, C_1 is the washoff coefficient, and C_2 is the washoff exponent.

This function produces the classic “first flush” behavior where initial runoff has higher pollutant concentrations.

Rating Curve Washoff

Where Q is the runoff flow rate. This function relates washoff directly to flow regardless of remaining buildup.

Event Mean Concentration (EMC)

The simplest approach - a constant concentration throughout the event. Use when first flush effects are not important or EMC is the only available data.

Recommended Washoff Parameters

Land Use and Pollutant Loading

Defining Land Uses

Land uses in SWMM link surfaces to their buildup and washoff characteristics:

[LANDUSES]
;;Name           Sweeping   Availability   Last
;;               Interval
Commercial       7          0.5            0
Residential      0          0.0            0
Industrial       3          0.7            0

The sweeping interval and availability parameters model street sweeping effects:

• Sweeping Interval - Days between sweeping events
• Availability - Fraction of buildup available for sweeping (not trapped in cracks)
• Last Swept - Days since the last sweep at simulation start

Assigning Land Uses to Subcatchments

Each subcatchment can have multiple land uses assigned by area fraction:

[COVERAGES]
;;Subcatchment   LandUse          Percent
S1               Commercial       60
S1               Residential      30
S1               Industrial       10

BMP Treatment Functions

SWMM provides multiple approaches for modeling pollutant removal by BMPs, storage facilities, and LID controls.

Removal Equations

Treatment can be specified as a removal function applied to inflow concentrations:

Where R is the removal fraction (0-1).

More sophisticated treatment functions can reference:

• Hydraulic Residence Time (HRT) - Time water spends in facility
• Depth - Water depth in facility
• Area - Facility surface area
• Flow - Current flow rate

Example Treatment Functions

Settling-based TSS removal:

[TREATMENT]
;;Node           Pollutant        Function
Pond1            TSS              R = 1 - EXP(-0.5*HRT)
Pond1            TP               R = 1 - EXP(-0.3*HRT)

Flow-based removal (wetland):

[TREATMENT]
;;Node           Pollutant        Function
Wetland1         TSS              C = 20 + (Cin - 20)*EXP(-0.1*HRT)

LID Treatment

LID controls in SWMM automatically apply pollutant removal based on the processes occurring in each layer:

You can override default LID treatment by specifying removal fractions in the LID definition.

Modeling Specific Pollutants

Total Suspended Solids (TSS)

TSS is often the primary pollutant of concern and serves as a carrier for other pollutants:

Key considerations:

• Use exponential washoff to capture first flush
• Settling is the primary removal mechanism
• Particle size distribution affects settling velocity
• TSS can serve as a co-pollutant for metals and phosphorus

Typical model setup:

[POLLUTANTS]
;;Name           Units  Rain   GW    I&I   Decay  Snow  CoPoll  CoFrac
TSS              MG/L   0      0     0     0      NO    *       0

[BUILDUP]
;;Land Use       Pollutant   Function   Coeff1   Coeff2   Coeff3   Normalizer
Commercial       TSS         EXP        50       0.5      0        CURBLENGTH
Residential      TSS         EXP        25       0.3      0        CURBLENGTH

[WASHOFF]
;;Land Use       Pollutant   Function   Coeff1   Coeff2   Ecleaning  Bmp
Commercial       TSS         EXP        0.1      1.0      0          0
Residential      TSS         EXP        0.08     1.0      0          0

Nutrients (Nitrogen and Phosphorus)

Nutrients are critical for receiving water eutrophication:

Total Phosphorus:

• 40-80% is particulate (associated with TSS)
• Dissolved fraction is biologically available
• Model particulate P as a fraction of TSS

Total Nitrogen:

• Includes organic, ammonia, nitrite, and nitrate forms
• More complex transformations than phosphorus
• Consider atmospheric deposition in rainfall

Using co-pollutant relationships:

[POLLUTANTS]
;;Name           Units  Rain   GW    I&I   Decay  Snow  CoPoll  CoFrac
TSS              MG/L   0      0     0     0      NO    *       0
ParticulateP     MG/L   0      0     0     0      NO    TSS     0.002
DissolvedP       MG/L   0.02   0     0     0      NO    *       0

Heavy Metals

Heavy metals like copper, zinc, and lead are primarily associated with particulate matter:

Key sources:

• Zinc: Galvanized surfaces, tire wear
• Copper: Brake pads, roofing materials
• Lead: Historical deposits, paint

Modeling approach:

[POLLUTANTS]
;;Name           Units  Rain   GW    I&I   Decay  Snow  CoPoll  CoFrac
TotalZinc        UG/L   0      0     0     0      NO    TSS     0.8
TotalCopper      UG/L   0      0     0     0      NO    TSS     0.15

Bacteria and Pathogens

Fecal indicator bacteria require special considerations:

• First-order decay - Bacteria die off over time (use decay coefficient)
• High variability - Concentrations span orders of magnitude
• Temperature dependence - Decay rates vary with temperature
• Source identification - Pet waste, wildlife, sanitary sewer overflows

[POLLUTANTS]
;;Name           Units  Rain   GW      I&I     Decay  Snow  CoPoll  CoFrac
Ecoli            #/L    0      10000   100000  0.5    NO    *       0

Calibration and Validation

Required Monitoring Data

Water quality calibration requires:

1. Flow data - Concurrent flow measurements
2. Composite samples - Flow-weighted event composites for EMC
3. Grab samples - Time-series data for pollutograph shape
4. Dry weather data - Baseflow concentrations
5. Rainfall data - Event depths and antecedent dry period

Calibration Approach

Step 1: Calibrate hydrology first Water quality results are meaningless without accurate flow predictions.

Step 2: Adjust buildup parameters Match predicted annual loads to measured or literature values.

Step 3: Adjust washoff parameters Match event EMCs and pollutograph shapes.

Step 4: Validate treatment Compare predicted and measured BMP effluent concentrations.

Performance Metrics

Limitations and Best Practices

Model Limitations

1. Simplified chemistry - No speciation or transformation modeling
2. Empirical functions - Buildup/washoff functions are curve fits, not process-based
3. Lumped parameters - Spatial variability within subcatchments is not captured
4. Limited treatment - First-order removal does not capture complex BMP processes
5. No sediment transport - In-pipe deposition and resuspension not modeled

Best Practices

1. Start simple - Begin with TSS and add pollutants as needed
2. Use local data - Literature values are starting points only
3. Consider uncertainty - Report ranges, not single values
4. Validate predictions - Compare to monitoring data when available
5. Document assumptions - Record data sources and parameter derivations

References

1. Rossman, L.A. & Huber, W.C. (2016). Storm Water Management Model Reference Manual Volume III - Water Quality. EPA/600/R-16/093. U.S. Environmental Protection Agency.

2. Pitt, R., Maestre, A., & Clary, J. (2018). The National Stormwater Quality Database (NSQD), Version 4.02.

3. International Stormwater BMP Database. (2020). BMP Database Summary Statistics. Water Research Foundation.

4. Selbig, W.R. (2016). Evaluation of leaf removal as a means to reduce nutrient concentrations and loads in urban stormwater. Science of the Total Environment, 571, 124-133.

5. Wong, T.H.F., Fletcher, T.D., Duncan, H.P., & Jenkins, G.A. (2006). Modelling urban stormwater treatment - A unified approach. Ecological Engineering, 27(1), 58-70.