Infiltration is the process by which rainfall enters the soil surface. In EPA SWMM, the selected infiltration model and its parameters affect the computed runoff from pervious areas. This tutorial describes the five options available in SWMM 5.2 and the checks needed when selecting their inputs.
Overview of SWMM Infiltration Methods
SWMM 5.2 offers five infiltration options. The three method families are:
- Horton - Empirical time-decay model
- Green-Ampt - Physically-based soil physics model
- Curve Number - SWMM’s capacity-based adaptation of the SCS/NRCS Curve Number concept
The other two options are Modified Horton and Modified Green-Ampt. Modified Horton changes the state variable used during low rainfall; Modified Green-Ampt changes how the surface layer’s moisture deficit is depleted during an initial low-intensity period. These definitions come from the EPA SWMM 5.2 User’s Manual.
The Horton Method
Horton’s equation describes how infiltration capacity decreases exponentially from an initial maximum rate to a minimum equilibrium rate as the soil becomes saturated.
Governing Equation
Where:
- f_p(t) = Infiltration capacity at time t (in/hr or mm/hr)
- f_0 = Maximum (initial) infiltration rate
- f_infinity = Minimum (equilibrium) infiltration rate
- k = Decay coefficient (1/hr)
- t = Time since infiltration began
Horton Parameters
| Parameter | Description | Units |
|---|---|---|
| Max Rate (f_0) | Initial infiltration capacity | in/hr or mm/hr |
| Min Rate (f_infinity) | Final equilibrium rate | in/hr or mm/hr |
| Decay Constant (k) | Rate of decrease | 1/hr |
| Drying Time | Time to fully regenerate capacity | days |
| Max Volume | Maximum infiltration volume (optional) | in or mm |
Supporting Horton Inputs
Soil texture alone does not determine SWMM’s initial rate, final rate, and decay coefficient. Support those inputs with site observations, a traceable study applicable to the modeled soil and condition, or calibration to measured rainfall-runoff data. Document the source and test the sensitivity of the decision-relevant runoff results.
Horton Recovery During Dry Periods
Do not model recovery by applying a separate exponential directly to the maximum infiltration rate. The EPA solver converts the entered drying time to a regeneration constant:
where T_d is the entered drying time, converted to seconds by the solver. SWMM defines that
input as the time to reach 98 percent recovery on its exponential drying curve. During a dry step,
the solver uses k_r (including any monthly evaporation recovery factor) to move its internal
equivalent time on the Horton curve back toward the fully recovered state. This is a state update,
not the displayed f_0(t) equation that is sometimes used in simplified summaries.
The Green-Ampt Method
The Green-Ampt method is based on soil physics and assumes a sharp wetting front that moves downward through the soil profile. It is more physically based than Horton but requires additional soil parameters.
Governing Equation
Where:
- f = Infiltration rate (in/hr or mm/hr)
- K_s = Saturated hydraulic conductivity
- psi = Suction head at wetting front
- Delta theta = Change in moisture content (porosity - initial moisture)
- F = Cumulative infiltration depth
Green-Ampt Parameters
| Parameter | Description | Units |
|---|---|---|
| Suction Head (psi) | Capillary suction at wetting front | in or mm |
| Conductivity (K_s) | Saturated hydraulic conductivity | in/hr or mm/hr |
| Initial Deficit | Porosity minus initial moisture content | fraction |
Typical Green-Ampt Parameter Values
| Soil Texture | K_s (in/hr) | Suction Head (in) | Porosity | Field Capacity |
|---|---|---|---|---|
| Sand | 4.74 | 1.93 | 0.437 | 0.062 |
| Loamy Sand | 1.18 | 2.40 | 0.437 | 0.105 |
| Sandy Loam | 0.43 | 4.33 | 0.453 | 0.190 |
| Loam | 0.13 | 3.50 | 0.463 | 0.232 |
| Silt Loam | 0.26 | 6.69 | 0.501 | 0.284 |
| Sandy Clay Loam | 0.06 | 8.66 | 0.398 | 0.244 |
| Clay Loam | 0.04 | 8.27 | 0.464 | 0.310 |
| Silty Clay Loam | 0.04 | 10.63 | 0.471 | 0.342 |
| Sandy Clay | 0.02 | 9.45 | 0.430 | 0.321 |
| Silty Clay | 0.02 | 11.42 | 0.479 | 0.371 |
| Clay | 0.01 | 12.45 | 0.475 | 0.378 |
Modified Green-Ampt in SWMM
Modified Green-Ampt is a separate selectable option. EPA describes it as preserving moisture deficit in the top surface layer during an initial period when rainfall intensity is below saturated hydraulic conductivity; it is not a multiple-soil-layer model.
The Curve Number Method
SWMM uses the Curve Number to establish a maximum soil infiltration capacity. Its time-step
solver does not apply the NRCS event-runoff equation with an assumed initial abstraction of
I_a = 0.2S. Do not expect a SWMM run to reproduce a separate TR-55 event calculation solely
because both use a Curve Number input.
The solver converts Curve Number to maximum capacity using:
During rainfall, SWMM updates potential cumulative infiltration from cumulative rainfall P and
the effective remaining capacity S_e:
Actual infiltration is limited by available rainfall, runon and ponded water. Capacity is depleted
by infiltration and regenerates between events according to the drying-time input. These details
follow the official EPA solver’s curvenum_getInfil implementation.
Curve Number Parameters in SWMM
| Parameter | Description | Range |
|---|---|---|
| Curve Number | SCS/NRCS curve number; EPA solver bounds the input | 10-99 |
| Conductivity | Deprecated field; no longer used by SWMM 5.2 | — |
| Drying Time | Time to fully regenerate capacity | days |
Typical Curve Numbers by Land Use
| Land Use | HSG A | HSG B | HSG C | HSG D |
|---|---|---|---|---|
| Open Space (Good) | 39 | 61 | 74 | 80 |
| Open Space (Fair) | 49 | 69 | 79 | 84 |
| Residential (1/2 ac) | 54 | 70 | 80 | 85 |
| Residential (1/4 ac) | 61 | 75 | 83 | 87 |
| Commercial (85% imp) | 89 | 92 | 94 | 95 |
| Industrial (72% imp) | 81 | 88 | 91 | 93 |
| Paved | 98 | 98 | 98 | 98 |
| Woods (Good) | 30 | 55 | 70 | 77 |
| Pasture (Good) | 39 | 61 | 74 | 80 |
Selecting the Right Method
Consider Horton When:
- Parameters can be supported by observations, literature, and calibration
- Its empirical capacity-decay representation fits the study objective
Consider Green-Ampt When:
- Suction head, saturated conductivity, and initial moisture deficit can be supported
- A sharp wetting-front representation is appropriate for the modeled pervious area
Consider Curve Number When:
- The governing workflow requires curve-number inputs
- The difference between SWMM’s infiltration-capacity implementation and an event runoff calculation is documented
Parameter Estimation Strategies
From Soil Surveys
The NRCS Web Soil Survey can provide soil-map context, texture, and saturated hydraulic conductivity data that may support development of Green-Ampt inputs. Map-unit data are not site measurements: confirm the mapped component, depth interval, units, and whether grading or compaction changed the soil. Suction head and initial moisture deficit still need a defensible source or calibration appropriate to the modeled condition.
Do not translate Web Soil Survey texture or hydrologic soil group directly into Horton f_0,
f_infinity, or decay values. Those empirical parameters require separate supporting evidence and,
where observations are available, calibration.
From Literature Values
When site-specific data is unavailable, use traceable published values based on soil texture. The soil-property examples above reproduce the parameter types tabulated in Appendix A.2 of EPA’s User’s Manual; they remain starting values, not site measurements.
From Calibration
For models with observed flow data, calibrate infiltration parameters by:
- Start with literature values
- Adjust to match runoff volumes
- Fine-tune to match hydrograph timing
- Verify with independent storm events
Impervious vs. Pervious Areas
SWMM applies infiltration only to pervious portions of subcatchments. Set these properties correctly:
- Percent Impervious: Fraction of subcatchment that is impervious
- Percent Zero-Imperv: Fraction of impervious area with no depression storage
Infiltration parameters apply only to the pervious fraction (1 - %Imperv).
Accounting for Compaction
Urban soil behavior can differ substantially from an undisturbed map-unit description. Use field testing, construction records, or calibrated observations where compaction is material; do not apply a universal reduction factor.
Common Modeling Mistakes
1. Using Undisturbed Soil Values for Developed Sites
Do not use undisturbed soil values automatically for developed areas. Determine whether grading, fill, or compaction changed the modeled parameters and document the evidence.
2. Substituting an Antecedent-Moisture Lookup for SWMM State
Do not silently replace SWMM’s recovery state with an event-method antecedent-moisture adjustment. If a governing method requires adjusted Curve Numbers or initial deficits, document how that requirement is represented and test the resulting initial condition.
3. Treating a SWMM Method as Automatic Regulatory Consistency
Confirm the infiltration method and parameters accepted by the reviewing jurisdiction. A jurisdiction may require a separate event calculation or governing design-storm method; selecting SWMM’s Curve Number infiltration option does not by itself reproduce that method or establish consistency with it.
4. Neglecting Parameter Recovery
For continuous simulation, ensure drying time parameters are realistic. Unrealistic recovery can over- or under-estimate runoff in storm sequences.
Summary
Proper infiltration modeling in SWMM requires:
- Selecting the appropriate method for your application
- Using realistic parameter values based on soil type
- Accounting for urban soil compaction
- Considering antecedent moisture conditions
- Validating results against observed data when possible
The choice between Horton, Green-Ampt, and Curve Number methods depends on data availability, regulatory requirements, and modeling objectives. Parameterization alone does not prove accuracy; review continuity, sensitivity, and calibration or validation evidence appropriate to the study.
References
-
Rossman, L. A. (2022). Storm Water Management Model User’s Manual Version 5.2. U.S. Environmental Protection Agency, EPA/600/R-22/030. Official PDF.
-
U.S. Environmental Protection Agency. Storm Water Management Model Solver,
infil.c. Official source code. -
Rawls, W. J., Brakensiek, D. L., & Miller, N. (1983). Green-Ampt infiltration parameters from soils data. Journal of Hydraulic Engineering, 109(1), 62-70.
-
Natural Resources Conservation Service. (2004). National Engineering Handbook, Part 630: Hydrology, Chapter 9. U.S. Department of Agriculture.
-
Pitt, R., Chen, S. E., Clark, S. E., Swenson, J., & Ong, C. K. (2008). Compaction’s impacts on urban storm-water infiltration. Journal of Irrigation and Drainage Engineering, 134(5), 652-658.
-
Horton, R. E. (1941). An approach toward a physical interpretation of infiltration-capacity. Soil Science Society of America Journal, 5(C), 399-417.