EPA SWMM Infiltration Methods

Last updated on July 14, 2026

Infiltration is the process by which rainfall or surface water is absorbed into the soil through pervious surfaces within a subbasin. In stormwater modeling, accurately representing infiltration is critical because the volume of water that enters the soil directly affects how much becomes surface runoff. This runoff response, in turn, drives the peak flow rates, runoff volumes, and hydrograph shapes that engineers use to size pipes, design detention ponds, and evaluate flood risk.

Because infiltration has such a direct impact on runoff results, selecting the right infiltration method is an important modeling decision. Each method makes different assumptions about soil behavior, rainfall response, and moisture recovery, which affects how well it applies to a given land use, soil type, and storm event. Choosing an inappropriate method can lead to overestimated or underestimated runoff, resulting in drainage infrastructure that is either over-designed, under-designed, or not well aligned with field conditions.

In GeoSTORM, the following EPA SWMM infiltration methods are available:

Green Ampt
Horton
Modified Green Ampt
Modified Horton
SCS Curve Number

This article describes each EPA SWMM infiltration method available in GeoSTORM.

Selecting an Infiltration Method

Follow the steps below to select an infiltration method:

From the Input ribbon menu, click the Scenario Manager dropdown menu, and select the Scenario Manager command.
The Scenario Manager dialog box will be displayed, as shown below.
From the EPA SWMM infiltration method dropdown combo box, select the infiltration method.

Green Ampt Method

The Green Ampt method is a physically based infiltration method that represents the movement of water into soil using a simplified wetting-front concept. The method assumes that, as water infiltrates, a sharp wetting front advances downward through the soil. This wetting front separates a saturated or nearly saturated zone above from soil at its initial moisture content below. The rate at which water infiltrates depends on the capillary suction pulling water toward the dry soil, the hydraulic conductivity of the saturated zone, and the current depth of the wetting front. Under sustained rainfall or ponded conditions, the wetting front moves deeper into the soil, the hydraulic gradient gradually decreases, and the infiltration rate approaches the saturated hydraulic conductivity.

Because this method is based on Darcy’s Law and soil-water movement concepts, its parameters can be estimated from measurable soil properties, soil surveys, or published values for different soil texture classes. This makes it a strong choice when soil data is available or when a more physically meaningful infiltration method is preferred.

In GeoSTORM, the Green Ampt infiltration method utilizes the following parameters:

Soil capillary suction head
Soil hydraulic conductivity
Soil initial deficit

The following table summarizes the main pros and cons of the Green Ampt infiltration method:

Pros	Cons
Parameters are related to soil properties (i.e., hydraulic conductivity, porosity, and initial moisture condition) that can be measured, estimated from soil surveys, or obtained from published tables. Works well for many urban, agricultural, and disturbed-soil catchments where the sharp wetting-front assumption is reasonable. Published parameter values are available for many USDA soil texture classes, making the method practical for engineering applications.	Best suited to surface-runoff modeling where infiltration from surface water into the soil controls runoff generation; less applicable in forested or highly permeable watersheds where subsurface flow often dominates. Less accurate for soils with highly irregular layering or macropores (e.g., cracked clays, soils with root channels). Requires soil-property estimates that may be uncertain if site-specific testing or reliable soil survey data are not available

Pros

Cons

Parameters are related to soil properties (i.e., hydraulic conductivity, porosity, and initial moisture condition) that can be measured, estimated from soil surveys, or obtained from published tables.
Works well for many urban, agricultural, and disturbed-soil catchments where the sharp wetting-front assumption is reasonable.
Published parameter values are available for many USDA soil texture classes, making the method practical for engineering applications.

Best suited to surface-runoff modeling where infiltration from surface water into the soil controls runoff generation; less applicable in forested or highly permeable watersheds where subsurface flow often dominates.
Less accurate for soils with highly irregular layering or macropores (e.g., cracked clays, soils with root channels).
Requires soil-property estimates that may be uncertain if site-specific testing or reliable soil survey data are not available

Horton Method

The Horton method is an empirical method for describing how a soil’s infiltration capacity changes during a rainfall event. Under sustained rainfall where the rainfall supply exceeds the soil’s ability to absorb water, the infiltration capacity decreases from an initial maximum rate toward a long-term minimum rate. At the beginning of a storm, the soil infiltrates water quickly. As the soil wets up, air becomes trapped, pores close, and the infiltration rate decreases exponentially until it reaches a long-term minimum. Horton captured this behavior with a simple exponential decay equation relating the current infiltration rate to the initial maximum rate, the final minimum rate, and a decay constant controlling how quickly the rate decreases.

Because the method is empirical, it is relatively simple to apply and is useful where detailed soil-property data are limited, but general infiltration behavior can be estimated from land use, soil type, or calibration data.

In GeoSTORM, the Horton infiltration method utilizes the following parameters:

Maximum infiltration rate
Minimum infiltration rate
Horton decay constant
Maximum infiltration volume
Saturated soil drying time

The following table summarizes the main pros and cons of the Horton infiltration method:

Pros	Cons
Well-suited for urban and agricultural catchments where overland flow is common, and infiltration capacity is reduced by soil compaction or impervious cover. Relatively straightforward to parameterize using published literature values for different land uses and soil types. Uses an intuitive conceptual model that is easy to explain to regulatory reviewers.	Empirically based, its parameters are not directly tied to measurable physical soil properties. Less suitable for forested, highly permeable, or subsurface-flow-dominated watersheds where overland flow is rare. Tracks infiltration based on elapsed time, which can be less accurate during events with long dry periods or variable intensities.

Modified Green Ampt Method

The Modified Green-Ampt method refines the original Green-Ampt approach for storms that begin with a prolonged period of light rainfall before rainfall intensity exceeds the soil’s saturated hydraulic conductivity. In the original Green-Ampt method, the moisture deficit in the upper soil layer can be depleted too early during these low-intensity periods. The modified method preserves the moisture deficit in the shallow surface layer during the early portion of the storm, resulting in more realistic infiltration estimates when heavier rainfall occurs later.

This modification is especially useful for storm patterns where light rainfall continues for an extended period before the peak rainfall intensity arrives, such as many frontal storm events. In these cases, the modified method can allow more infiltration before runoff begins and may produce lower runoff volumes than the original Green-Ampt method. For short-duration, high-intensity storms where ponding begins almost immediately, the original and modified methods often produce similar results.

In GeoSTORM, the Modified Green Ampt infiltration method utilizes the following parameters:

Soil capillary suction head
Soil hydraulic conductivity
Soil initial deficit

The following table summarizes the main pros and cons of the Modified Green Ampt infiltration method:

Pros	Cons
Produces more realistic infiltration estimates for storms with prolonged low-intensity rainfall before peak rainfall occurs. Uses the same physically meaningful parameter set as the original Green-Ampt method. Parameters can be estimated from soil surveys, published soil texture tables, laboratory testing, or calibration data. Better represents soil moisture dynamics in the shallow surface layer where ponding and runoff initiation occur.	Adds modeling complexity compared to the original Green-Ampt method, with limited benefit for short-duration, high-intensity storms where ponding begins quickly. Requires reliable estimates of soil suction, hydraulic conductivity, and initial moisture deficit. Still works within a surface-runoff modeling framework, making it less applicable where subsurface flow dominates runoff response. Less accurate for strongly layered soils, cracked clays, macropores, root channels, or other preferential flow conditions.

Pros

Cons

Produces more realistic infiltration estimates for storms with prolonged low-intensity rainfall before peak rainfall occurs.
Uses the same physically meaningful parameter set as the original Green-Ampt method.
Parameters can be estimated from soil surveys, published soil texture tables, laboratory testing, or calibration data.
Better represents soil moisture dynamics in the shallow surface layer where ponding and runoff initiation occur.

Adds modeling complexity compared to the original Green-Ampt method, with limited benefit for short-duration, high-intensity storms where ponding begins quickly.
Requires reliable estimates of soil suction, hydraulic conductivity, and initial moisture deficit.
Still works within a surface-runoff modeling framework, making it less applicable where subsurface flow dominates runoff response.
Less accurate for strongly layered soils, cracked clays, macropores, root channels, or other preferential flow conditions.

Modified Horton Method

The Modified Horton method addresses a limitation of the original Horton formulation. In the original Horton method, infiltration capacity decreases along a time-based decay curve during a rainfall event. This can reduce the available infiltration capacity even during periods when rainfall intensity is low, and the soil is not infiltrating water at the full potential Horton rate.

The modified version replaces elapsed time along the Horton curve with cumulative infiltration in excess of the minimum infiltration rate as the state variable. In practical terms, the infiltration curve advances based on how much water actually infiltrates above the minimum rate, rather than simply how much time has passed during rainfall. This provides a more realistic infiltration estimate during low-intensity rainfall and variable-intensity storm events.

Similar to the original Horton method, the Modified Horton method also accounts for soil drying and recovery between storm events through the saturated soil drying time parameter. This makes it useful for continuous simulations with multiple rainfall events, especially where storms include alternating periods of light rainfall, heavier rainfall, and dry intervals.

In GeoSTORM, the Modified Horton infiltration method utilizes the following parameters:

Maximum infiltration rate
Minimum infiltration rate
Horton decay constant
Maximum infiltration volume
Saturated soil drying time

The following table summarizes the main pros and cons of the Modified Horton infiltration method:

Pros	Cons
Simple and computationally efficient, making it suitable for long-term simulations with multiple storm events. Provides more realistic infiltration estimates during low-intensity rainfall than the original Horton method. Better handles variable-intensity rainfall events where infiltration capacity should respond to actual infiltrated volume rather than elapsed rainfall time alone.	Still empirically based, so its parameters are not directly derived from first-principles soil physics. Requires reasonable estimates of maximum infiltration rate, minimum infiltration rate, decay constant, maximum infiltration volume, and drying time. Less accurate for rapid, intense infiltration changes caused by highly variable storm patterns or extreme events with sharply shifting rainfall intensity.

Pros

Cons

Simple and computationally efficient, making it suitable for long-term simulations with multiple storm events.
Provides more realistic infiltration estimates during low-intensity rainfall than the original Horton method.
Better handles variable-intensity rainfall events where infiltration capacity should respond to actual infiltrated volume rather than elapsed rainfall time alone.

Still empirically based, so its parameters are not directly derived from first-principles soil physics.
Requires reasonable estimates of maximum infiltration rate, minimum infiltration rate, decay constant, maximum infiltration volume, and drying time.
Less accurate for rapid, intense infiltration changes caused by highly variable storm patterns or extreme events with sharply shifting rainfall intensity.

SCS Curve Number Method

The SCS Curve Number method, now commonly referred to as the NRCS Curve Number method, was developed by the U.S. Soil Conservation Service and is documented in NRCS hydrology guidance such as TR-55 and TR-20. Rather than modeling the infiltration process directly, it estimates direct runoff from a storm event based on cumulative rainfall depth and a single lumped parameter: the Curve Number (CN). The CN represents the combined effect of land use, hydrologic soil group, cover conditions, imperviousness, and antecedent moisture conditions.

Within EPA SWMM, the Curve Number method is adapted for continuous rainfall-runoff simulation. The method assumes that a soil’s total infiltration capacity can be estimated from its tabulated Curve Number. During a rainfall event, this capacity is depleted as a function of cumulative rainfall and the remaining infiltration capacity. As the storm continues, the infiltration rate can approach zero, which is an important behavioral difference from methods such as Horton and Green-Ampt, which retain a minimum or physically based limiting infiltration rate.

Curve Numbers are commonly tabulated by land use or cover type and hydrologic soil group: A, B, C, or D. In GeoSTORM, composite Curve Numbers can be computed from land use/land cover data and hydrologic soil group data for each subbasin.

In GeoSTORM, the SCS Curve Number infiltration method utilizes the following parameters:

Curve number
Saturated soil drying time

The following table summarizes the main pros and cons of the SCS Curve Number infiltration method:

Pros	Cons
Most effective for the types of hydrologic problems it was originally designed to solve, such as evaluating land use changes and conservation measures on direct runoff. Widely accepted by regulatory agencies and commonly used throughout the U.S. and internationally. Can be applied in forested areas, where the Green Ampt and Horton methods are less appropriate. Requires only two infiltration parameters, both of which are relatively easy to estimate from published NRCS tables. Easy to explain, review, and audit in agency submittals because CN values are widely tabulated by land use and hydrologic soil group.	May produce inaccurate results when computed runoff depth is less than 0.5 inches — the method was not validated for very small runoff volumes. Not recommended when the watershed-weighted CN is below 40 (highly permeable soils or heavily forested catchments). Infiltration rate approaches zero during long storm events rather than maintaining a constant minimum rate, which can overestimate runoff in prolonged storms. Does not account for rainfall intensity — a 1-inch rain in 1 hour and a 1-inch rain over 24 hours produce the same total infiltration loss. Not suitable for estimating runoff from snowmelt or rain on frozen ground.

Pros

Cons

Most effective for the types of hydrologic problems it was originally designed to solve, such as evaluating land use changes and conservation measures on direct runoff.
Widely accepted by regulatory agencies and commonly used throughout the U.S. and internationally.
Can be applied in forested areas, where the Green Ampt and Horton methods are less appropriate.
Requires only two infiltration parameters, both of which are relatively easy to estimate from published NRCS tables.
Easy to explain, review, and audit in agency submittals because CN values are widely tabulated by land use and hydrologic soil group.

May produce inaccurate results when computed runoff depth is less than 0.5 inches — the method was not validated for very small runoff volumes.
Not recommended when the watershed-weighted CN is below 40 (highly permeable soils or heavily forested catchments).
Infiltration rate approaches zero during long storm events rather than maintaining a constant minimum rate, which can overestimate runoff in prolonged storms.
Does not account for rainfall intensity — a 1-inch rain in 1 hour and a 1-inch rain over 24 hours produce the same total infiltration loss.
Not suitable for estimating runoff from snowmelt or rain on frozen ground.

Infiltration Method Quick Comparison

The following table provides a quick reference for comparing each infiltration method.

Method	Best For	Key Parameters	Key Limitation
Green Ampt	Sites with reliable soil-property data, especially hydraulic conductivity, suction head, and initial moisture condition	Capillary suction head, hydraulic conductivity, initial moisture deficit	Works within a surface-runoff modeling framework; less suitable where subsurface flow, macropores, or strong soil layering dominate
Horton	Urban and agricultural catchments with compacted, disturbed, or relatively low-infiltration soils	Maximum/minimum infiltration rates, decay constant, maximum infiltration volume, drying time	Empirically based; not recommended for forested or high-permeability soils
Modified Green Ampt	Storms with prolonged low-intensity rainfall before peak rainfall occurs, such as many frontal storm patterns	Capillary suction head, hydraulic conductivity, initial moisture deficit	Still assumes overland flow; less accurate for highly permeable or forested areas
Modified Horton	Continuous simulations, multi-event rainfall records, and variable-intensity storms	Maximum/minimum infiltration rates, decay constant, max infiltration volume, drying time	Empirically based; less accurate for rapid changes from intense storms
SCS Curve Number	Land use change evaluation, standard NRCS-based practice, agency submittals, and forested areas where appropriate CN values are selected	Curve number, saturated soil drying time	Runoff-volume method is driven mainly by cumulative rainfall depth; less accurate for runoff < 0.5 in, weighted CN < 40, snowmelt, frozen ground, or watersheds with major subsurface flow

Choosing the Right Infiltration Method

In practice, no single infiltration method is universally best. The right choice depends on the project’s land use, available soil data, storm type, regulatory requirements, and the level of physical detail needed for the analysis. For many urban stormwater projects that follow NRCS design standards, the SCS Curve Number method is a common starting point because it is widely understood by reviewers and easy to document.

When in doubt, consider running the model with more than one infiltration method and comparing the results. In GeoSTORM, infiltration methods are selected through the Scenario Manager when EPA SWMM is used as the hydrology analysis engine. This allows separate scenarios to be configured with different infiltration methods so peak flows, runoff volumes, and system performance can be compared without rebuilding the model setup.

Conclusion

EPA SWMM infiltration methods help GeoSTORM estimate how rainfall is divided between soil infiltration and surface runoff. Green-Ampt and Modified Green-Ampt are best suited when reliable soil-property data are available, Horton and Modified Horton provide practical empirical options, and SCS Curve Number supports widely accepted NRCS-based rainfall-runoff workflows. For best results, choose the infiltration method early, use input parameters that match the selected method, and review the runoff response carefully before using the model results for design decisions.