Skip common site navigation and headers
United States Environmental Protection Agency
National Pollutant Discharge Elimination System (NPDES)
Begin Hierarchical Links EPA Home > OW Home > OWM Home > NPDES Home > Stormwater > Menu of BMPs End Hierarchical Links
Menu of BMPs Home
 
BMP Background
 
Public Education & Outreach on
Stormwater Impacts
 
Public Involvement/
Participation
 
Illicit Discharge Detection & Elimination
 
Construction Site Stormwater Runoff Control
 
Post-Construction Stormwater Management in New Development & Redevelopment
 
Pollution Prevention/Good Housekeeping for Municipal Operations
 
Stormwater Home
 
 
 

 

 
Search BMPs

Filter by Minimum Measure
GO Browse Fact Sheets Search Help

Grassed Swales

Minimum Measure: Post-Construction Stormwater Management in New Development and Redevelopment

Subcategory: Infiltration

Photo Description:  Grassed swales can be used along roadsides and parking lots to collect and treat stormwater runoff. Description

In the context of BMPS to improve water quality, the term swale (a.k.a. grassed channel, dry swale, wet swale, biofilter, or bioswale) refers to a vegetated, open-channel management practices designed specifically to treat and attenuate stormwater runoff for a specified water quality volume. As stormwater runoff flows along these channels, it is treated through vegetation slowing the water to allow sedimentation, filtering through a subsoil matrix, and/or infiltration into the underlying soils. Variations of the grassed swale include the grassed channel, dry swale, and wet swale. The specific design features and methods of treatment differ in each of these designs, but all are improvements on the traditional drainage ditch. These designs incorporate modified geometry and other features for use of the swale as a treatment and conveyance practice.

Applicability

Grassed swales can be applied in most situations with some restrictions. Swales are well suited for treating highway or residential road runoff because they are linear practices. Swales are also useful as one of a series of stormwater BMPs or as part of a treatment train, for instance, conveying water to a detention pond and receiving water from filter strips. Furthermore, swales are highly recommended by the proponents of design approaches such as Low Impact Development and Better Site Design (Low Impact Development (LID) and Other Green Designs fact sheet).

Regional Applicability

Grassed swales can be applied in most regions of the United States. In arid and semi-arid climates, however, the value of these practices needs to be weighed against the water needed to irrigate them.

Ultra-Urban Areas

Ultra-urban areas are densely developed urban areas with little pervious surface. Grass swales may not be well suited to ultra-urban areas because they require a relatively large area of pervious surfaces.

Stormwater Hot Spots

Stormwater hot spots are areas where land use or activities generate highly contaminated runoff, with concentrations of pollutants exceeding those typically found in stormwater. A typical example is a gas station or convenience store. With the exception of the dry swale design (see Design Variations), hot spot runoff should not be directed toward grassed channels. These practices either infiltrate stormwater or intersect the ground water, making use of the practices for hot spot runoff a threat to ground water quality.

Stormwater Retrofit

A stormwater retrofit is a stormwater management practice (usually structural) put into place after development has occurred to improve water quality, protect downstream channels, reduce flooding, or meet other specific objectives such as reducing loadings to comply with a TMDL waste load allocation. One retrofit opportunity using grassed swales modifies existing drainage ditches. Ditches have traditionally been designed only to convey stormwater. In some cases, it may be possible to incorporate features to enhance pollutant removal or infiltration such as check dams (i.e., small dams along the ditch that trap sediment, slow runoff, and reduce the effective longitudinal slope). Since grassed swales cannot treat a large area, using this practice to retrofit an entire watershed would be expensive because of the number of practices needed to manage runoff from a significant amount of the watershed's land area.

Cold Water (Trout) Streams

Grassed channels are a good treatment option within watersheds that drain to cold water streams. These practices do not pond water for a long period and often induce infiltration. As a result, standing water will not typically be subjected to solar warming.

Siting and Design Considerations

In addition to the broad applicability concerns described above, designers need to consider site conditions. In addition, they need to incorporate design features to improve the longevity and performance of the practice while minimizing the maintenance burden.

Siting Considerations

In addition to considering the restrictions and adaptations of grassed swales to different regions and land uses, designers need to ensure that this management practice is feasible at the site in question because some site conditions (i.e., steep slopes, highly impermeable soils) might restrict the effectiveness of grassed channels.

Drainage Area

Grassed swales should generally treat runoff from small drainage areas ( less than 5 acres). If used to treat larger areas, the flows through the swale become too large to produce designs to treat stormwater runoff in addition to conveyance.

Slope

Grassed swales should be used on sites with relatively flat slopes of less than 4 percent slope; 1 to 2 percent slope is recommended. When site conditions require installing the swales in areas with larger slopes, check dams can be used to reduce the influence of the slope. Runoff velocities within the channel become too high on steeper slopes. This can cause erosion and does not allow for infiltration or filtering in the swale.

Soils / Topography

Grassed swales can be used on most soils, with some restrictions on the most impermeable soils. In the dry swale (see Design Variations) a fabricated soil bed replaces on-site soils in order to ensure that runoff is filtered as it travels through the soils of the swale.

Ground Water

The required depth to ground water depends on the type of swale used. In the dry swale and grassed channel options, the bottom of the swale should be constructed at least 2 ft above the ground water table to prevent a moist swale bottom or contamination of the ground water. In the wet swale option, treatment is provided by creating a standing or slow flowing wet pool, which is maintained by intersecting the ground water.

Design Considerations

Although there are different design variations of the grassed swale (see Design Variations), there are some design considerations common to all designs. An overriding similarity is the cross-sectional geometry. Swales often have a trapezoidal or parabolic cross section with relatively flat side slopes (flatter than 3:1), though rectangular and triangular channels can also be used. Designing the channel with flat side slopes increases the wetted perimeter. The wetted perimeter is the length along the edge of the swale cross section where runoff flowing through the swale contacts the vegetated sides and bottom. Increasing the wetted perimeter slows runoff velocities and provides more contact with vegetation to encourage sorption, filtering, and infiltration. Another advantage to flat side slopes is that runoff entering the grassed swale from the side receives some pretreatment along the side slope.

Another similarity among designs is the type of pretreatment needed. In all design options, a small forebay should be used at the front of the swale to trap incoming sediments. A pea gravel diaphragm, a small trench filled with river-run gravel, should be constructed along the length of the swale and used as pretreatment for runoff entering the sides of the swale. Other features designed to enhance the performance of grassed swales are a flat longitudinal slope (generally between 1 percent and 2 percent) and a dense vegetative cover in the channel. The flat slope helps to reduce the flow velocity within the channel. The dense vegetation also helps reduce velocities, protects the channel from erosion, and acts as a filter to treat stormwater runoff. During construction, it is important to stabilize the channel while the vegetation is becoming established, either with a temporary grass cover or with natural or synthetic erosion control products. In addition to treating runoff for water quality, grassed swales must convey runoff from larger storms safely. Typical designs allow the runoff from the 2-year storm (i.e., the storm that occurs, on average, once every two years) to flow through the swale without causing erosion. Swales should also have the capacity to pass larger storms (typically a 10-year storm) safely.

Design Variations

The following discussion identifies three different variations of open channel practices, including the grassed channel, the dry swale, and wet swale.

Grassed Channel

Of the three grassed swale designs, grassed channels are the most similar to a conventional drainage ditch, with the major differences being flatter side slopes and longitudinal slopes, and a slower design velocity for water quality treatment of small storm events. Of all of the options, grassed channels are the least expensive but also provide the least reliable pollutant removal. An excellent application of a grassed channel is as pretreatment to other structural stormwater practices. A major difference between the grassed channel and many other structural practices is the method used to size the practice. Most stormwater management water quality practices are sized by volume. This method sets the volume available in the practice equal to the water quality volume, or the volume of water to be treated in the practice. The grassed channel, is a flow-rate-based design. Based on the peak flow from the water quality storm (this varies regionally, but a typical value is the 1-inch/ 24-hr storm), the channel should be designed so that runoff takes, on average, 10 minutes to flow from the top to the bottom of the channel. A procedure for this design can be found in Design of Stormwater Filtering Systems (CWP, 1996).

Dry Swales

Dry swales are similar in design to bioretention areas (see Bioretention (Rain Gardens) fact sheet). These designs incorporate a fabricated soil bed into their design. The native soil is replaced with a sand/soil mix that meets minimum permeability requirements. An underdrain system is installed at the bottom of the soil bed. This underdrain is a gravel layer that encases a perforated pipe. Stormwater treated in the soil bed flows into the underdrain, which routes this treated stormwater to the storm drain system or receiving waters. Dry swales are a relatively new design, but studies of swales with a native soil similar to the man-made soil bed of dry swales suggest high pollutant removal.

Wet Swales

Wet swales intersect the ground water and behave similarly to a linear wetland cell (see Stormwater Wetland fact sheet). This design variation incorporates a shallow permanent pool and wetland vegetation to provide stormwater treatment. This design also has potentially high pollutant removal. Wet swales are not commonly used in residential or commercial settings because the shallow standing water may be a potential mosquito breeding area.

Regional Variations

In cold or snowy climates, swales may serve a dual purpose by acting as both a snow storage/treatment and a stormwater management practice. This dual purpose is particularly relevant when swales are used to treat road runoff. If used for this purpose, swales should incorporate salt-tolerant vegetation, such as creeping bentgrass.

Arid Climates

In arid or semi-arid climates, swales should be designed with drought-tolerant vegetation, such as buffalo grass. As pointed out in the Applicability section, the value of vegetated practices for water quality needs to be balanced against the cost of water needed to maintain them in arid and semi-arid regions.

Limitations

Grassed swales have some limitations, including the following:

  • Grassed swales cannot treat a very large drainage area.
  • Wet swales may become a nuisance due to mosquito breeding.
  • If designed improperly (e.g., if proper slope is not achieved), grassed channels will have very little pollutant removal.
Maintenance Considerations

Maintenance of grassed swales mostly involves litter control and maintening the grass or wetland plant cover. Typical maintenance activities are included in Table 1.

Table 1. Typical maintenance activities for grassed swales (Source: Adapted from CWP, 1996)

Activity

Schedule

  • Inspect pea gravel diaphragm for clogging and correct the problem.
  • Inspect grass along side slopes for erosion and formation of rills or gullies and correct.
  • Remove trash and debris accumulated in the inflow forebay.
  • Inspect and correct erosion problems in the sand/soil bed of dry swales.
  • Based on inspection, plant an alternative grass species if the original grass cover has not been successfully established.
  • Replant wetland species (for wet swale) if not sufficiently established.

Annual
(semi-annual the first year)

  • Rototill or cultivate the surface of the sand/soil bed of dry swales if the swale does not draw down within 48 hours.
  • Remove sediment build-up within the bottom of the swale once it has accumulated to 25 percent of the original design volume.

As needed
(infrequent)

  • Mow grass to maintain a height of 3–4 inches

As needed
(frequent seasonally)

Effectiveness

Structural stormwater management practices can be used to achieve four broad resource protection goals. These include flood control, channel protection, ground water recharge, and pollutant removal. Grassed swales can be used to meet ground water recharge and pollutant removal goals.

Ground Water Recharge

Grassed channels and dry swales can provide some ground water recharge as infiltration is achieved within the practice. Wet swales, however, generally make little, if any, contributions to ground water recharge. Infiltration is impeded by the accumulation of debris on the bottom of the swale.

Pollutant Removal

Few studies are available regarding the effectiveness of grassed channels (Table 2). The data suggest relatively high removal rates for some pollutants, negative removals for some bacteria, and fair performance for phosphorous. One study of available performance data (Schueler, 1997) estimates the removal rates for grassed channels as:

Total Suspended Solids: 81%
Total Phosphorous: 29%
Nitrate Nitrogen: 38%
Metals: 14% to 55%
Bacteria: -50%

Table 2. Grassed swale pollutant removal efficiency data

Removal Efficiencies (% Removal)

Study

TSS

TP

TN

NO 3

Metals

Bacteria

Type

Goldberg 1993

67.8

4.5

-

31.4

42–62

-100

grassed
channel

Seattle Metro and Washington Department of Ecology 1992

60

45

-

-25

2–16

-25

grassed
channel

Seattle Metro and Washington Department of Ecology, 1992

83

29

-

-25

46–73

-25

grassed
channel

Wang et al., 1981

80

-

-

-

70–80

-

dry
swale

Dorman et al., 1989

98

18

-

45

37–81

-

dry
swale

Harper, 1988

87

83

84

80

88–90

-

dry
swale

Kercher et al., 1983

99

99

99

99

99

-

dry
swale

Harper, 1988.

81

17

40

52

37–69

-

wet
swale

Koon, 1995

67

39

-

9

-35 to 6

-

wet
swale

Occoquan Watershed Monitoring Lab, 1983

-100

-100

-100

-

-100

-

drainage
channel

Yousef et al., 1985

-

8

13

11

14–29

-

drainage
channel

Occoquan Watershed Monitoring Lab, 1983

-50

-9.1

-18.2

-

-100

-

drainage
channel

Yousef et al., 1985

-

-19.5

8

2

41–90

-

drainage
channel

Occoquan Watershed Monitoring Lab, 1983

31

-23

36.5

-

-100 to 33

-

drainage
channel

Welborn and Veenhuis, 1987

0

-25

-25

-25

0

-

drainage
channel

Yu et al., 1993

68

60

-

-

74

-

drainage
channel

Dorman et al., 1989

65

41

-

11

14-55

-

drainage
channel

Pitt and McLean, 1986

0

-

0

-

0

0

drainage
channel

Oakland, 1983

33

-25

-

-

20–58

0

drainage
channel

Dorman et al., 1989

-85

12

-

-100

14–88

-

drainage
channel

While it is difficult to distinguish between different designs based on the small amount of available data, grassed channels generally have poorer removal rates than wet and dry swales, although wet swales may export soluble phosphorous (Harper, 1988; Koon, 1995). It is not clear why swales export bacteria. One explanation is that bacteria thrive in the warm swale soils. Another explanation is that studies have not accounted for some sources of bacteria, and like any open BMP, swales likely receive inputs from wildlife. Another possible explanation is that local residents might walk dogs within the grassed swale area. Signs identifying swales as a stormwater BMP leading to local receiving waters might encourage some pet owners to clean up after their pets.

Cost Considerations

Little data are available to estimate the difference in cost between various swale designs. One study (SWRPC, 1991) estimated the construction cost of grassed channels at approximately $0.25 per ft2. This price does not include design costs or contingencies. Brown and Schueler (1997) estimate these costs at approximately 32 percent of construction costs for most stormwater management practices. For swales, however, these costs would probably be significantly higher since the construction costs are so low compared with other practices. A more realistic estimate would be a total cost of approximately $0.50 per ft2, which compares favorably with other stormwater management practices.

Costs to construct swales should be taken in context. With most development designs, some conveyance structure must be constructed as part of the development. The construction of grass swales is less expensive than concrete ditches or sewers. Hence, the use of grass swales is often a less expensive alternative than traditional design approaches.

References

Center for Watershed Protection (CWP). 1996. Design of Stormwater Filtering Systems. Prepared for the Chesapeake Research Consortium, Solomons, MD, and USEPA Region V, Chicago, IL, by the Center for Watershed Protection, Ellicott City, MD.

Brown, W., and T. Schueler. 1997. The Economics of Stormwater BMPs in the Mid-Atlantic Region. Prepared for the Chesapeake Research Consortium, Edgewater, MD, by the Center for Watershed Protection, Ellicott City, MD.

Dorman, M.E., J. Hartigan, R.F. Steg, and T. Quasebarth. 1989. Retention, Detention and Overland Flow for Pollutant Removal From Highway Stormwater Runoff. Vol. 1. FHWA/RD 89/202. Federal Highway Administration, Washington, DC.

Goldberg. 1993. Dayton Avenue Swale Biofiltration Study. Seattle Engineering Department, Seattle, WA.

Harper, H. 1988. Effects of Stormwater Management Systems on Groundwater Quality. Prepared for Florida Department of Environmental Regulation, Tallahassee, FL, by Environmental Research and Design, Inc., Orlando, FL.

Kercher, W.C., J.C. Landon, and R. Massarelli. 1983. Grassy swales prove cost-effective for water pollution control. Public Works, 16: 53-55.

Koon, J. 1995. Evaluation of Water Quality Ponds and Swales in the Issaquah/East Lake Sammamish Basins. King County Surface Water Management, Seattle, WA, and Washington Department of Ecology, Olympia, WA.

Oakland, P.H. 1983. An evaluation of stormwater pollutant removal through grassed swale treatment. In Proceedings of the International Symposium of Urban Hydrology, Hydraulics and Sediment Control, Lexington, KY. pp. 173-182.

Occoquan Watershed Monitoring Laboratory. 1983. Final Report: Metropolitan Washington Urban Runoff Project. Prepared for the Metropolitan Washington Council of Governments, Washington, DC, by the Occoquan Watershed Monitoring Laboratory, Manassas, VA.

Pitt, R., and J. McLean. 1986. Toronto Area Watershed Management Strategy Study: Humber River Pilot Watershed Project. Ontario Ministry of Environment, Toronto, ON.

Schueler, T. 1997. Comparative Pollutant Removal Capability of Urban BMPs: A reanalysis. Watershed Protection Techniques 2(2):379-383.

Seattle Metro and Washington Department of Ecology. 1992. Biofiltration Swale Performance: Recommendations and Design Considerations. Publication No. 657. Water Pollution Control Department, Seattle, WA.

Southeastern Wisconsin Regional Planning Commission (SWRPC). 1991. Costs of Urban Nonpoint Source Water Pollution Control Measures. Technical report no. 31. Southeastern Wisconsin Regional Planning Commission, Waukesha, WI.

Wang, T., D. Spyridakis, B. Mar, and R. Horner. 1981. Transport, Deposition and Control of Heavy Metals in Highway Runoff. FHWA-WA-RD-39-10. University of Washington, Department of Civil Engineering, Seattle, WA.

Welborn, C., and J. Veenhuis. 1987. Effects of Runoff Controls on the Quantity and Quality of Urban Runoff in Two Locations in Austin, TX. USGS Water Resources Investigations Report No. 87-4004. U.S. Geological Survey, Reston, VA.

Yousef, Y., M. Wanielista, H. Harper, D. Pearce, and R. Tolbert. 1985. Best Management Practices: Removal of Highway Contaminants By Roadside Swales. University of Central Florida and Florida Department of Transportation, Orlando, FL.

Yu, S., S. Barnes, and V. Gerde. 1993. Testing of Best Management Practices for Controlling Highway Runoff. FHWA/VA-93-R16. Virginia Transportation Research Council, Charlottesville, VA.

Information Resources

Maryland Department of the Environment (MDE). 2000. Maryland Stormwater Design Manual. [Maryland Stormwater Design Manual Exit EPA Site]. Accessed May 22, 2001.

Reeves, E. 1994. Performance and Condition of Biofilters in the Pacific Northwest. Watershed Protection Techniques 1(3):117-119.

Seattle Metro and Washington Department of Ecology. 1992. Biofiltration Swale Performance. Recommendations and Design Considerations. Publication No. 657. Seattle Metro and Washington Department of Ecology, Olympia, WA.

USEPA 1993. Guidance Specifying Management Measures for Sources of Nonpoint Pollution in Coastal Waters. EPA-840-B-92-002. U.S. Environmental Protection Agency, Office of Water. Washington, DC.

Watershed Management Institute (WMI). 1997. Operation, Maintenance, and Management of Stormwater Management Systems. Prepared for U.S. Environmental Protection Agency, Office of Water. Washington, DC, by the Watershed Management Institute, Ingleside, MD.

 

Office of Water | Office of Wastewater Management | Disclaimer | Search EPA

Begin Site Footer

EPA Home | Privacy and Security Notice | Contact Us

Last updated on July 23, 2012
URL:http://cfpub.epa.gov/npdes/stormwater/menuofbmps/index.cfm