Environmental Review Toolkit
Water, Wetlands, and Wildlife

 Previous Table of Contents Next     

Stormwater Best Management Practices in an Ultra-Urban Setting: Selection and Monitoring

6.4 Evaluation Phase

The evaluation phase provides a more detailed process to evaluate the ability of structural and nonstructural BMPs to meet management objectives. This process results in a final list of BMP options that can be ranked to optimize selection. Questions that should be asked during the evaluation phase include:

  • Do any structural or nonstructural BMPs have limited constituent removal effectiveness?

  • Is there one (or more) structural or nonstructural BMPs that outperform other BMPs?

  • Is a combination of structural and nonstructural BMPs warranted?

6.4.1 Evaluation of Structural BMPs

In the evaluation phase, a detailed analysis of the physical characteristics of the site will generate criteria for evaluating the constructability and feasibility of various structural BMPs. The effectiveness of the BMP in meeting management objectives for water quantity or quality control can be evaluated, along with any requirements for pretreatment that may increase the cost of the BMP.

Criteria for constructability include:

  • Site soils and topographic features. (Infiltration trenches typically require soils with an infiltration rate at least 12.7 mm/h [0.5 in/h].)

  • Depth to groundwater and bedrock. (Bioretention facilities require at least 1.8 m [6 ft].)

  • Land area commitments and availability of open space. (Detention ponds typically require 10 to 20 percent of the total drainage area.)

  • Site slope and hydraulics features. (Wetlands require very mild slopes to facilitate long residence times.)

  • Existing stormwater drainage pathways, storm drains, swales, etc. (Inverts of existing drainage pathways establish the minimum discharge elevation of any retrofit BMP.)

Effectiveness criteria include:

  • Stormwater quantity control and constituent removal effectiveness.

  • Effective integration with other BMPs and stormwater management efforts.

BMP Selection Based on Physical Site and Treatment Effectiveness

A prototype of a compost stormwater treatment facility (CSF) was constructed in Washington County, Oregon. The avenue was widened in 1987 to five lanes with additional bike lanes and sidewalks. An existing water quality swale that was constructed prior to the road widening was determined inadequate to treat the additional stormwater runoff (the swale acted as a pretreatment for a wetland pond that drained into Beaverton Creek). The compost filter was selected based on its ability to be retrofitted into the existing water quality swale, and its widely reported ability for adsorbing heavy metal, oils, greases, nutrients, and organic toxins (Stormwater Management, 1994).

There are several good sources available for detailed design and construction procedures and information, which can assist in evaluation of structural BMPs. These sources include Design of Stormwater Filtering Systems (Claytor and Schueler, 1996); Evaluation and Management of Highway Runoff Water Quality (Young et al., 1996); Urban Drainage Design Manual Hydraulic Engineering Circular 22 (Brown et al., 1996); and Retention, Detention, and Overland Flow for Pollutant Removal from Highway Stormwater Runoff; Volume II: Design Guidelines (Dorman et al., 1996b). Design specifications for new, commercially available BMPs can be obtained from their manufacturers along with information on installation or construction procedures.


Of the two broad categories of evaluation, criteria related to construction will probably have a stronger influence on the selection of structural BMPs in an ultra-urban setting. Screening potential BMPs based on construction issues requires significant effort to collect and then to evaluate site-specific information on the proposed site. Most of the required information is available from sources like USDA soil surveys, USGS topographic maps, and roadway feasibility studies. To evaluate BMPs and determine the feasibility of their use for a specific site requires information regarding:

  • The topography of the drainage area and the proposed general location of the BMP (vertical contours of less than 1.5 m [5 ft] is preferred).

  • Estimates of the volume of stormwater runoff to be managed for water quality and quantity.

  • Local information on any existing stormwater drainageways, including, where possible, information on open channel and storm drain size and inverts.

  • Information on soils in the vicinity of the proposed BMP.

  • Depth to bedrock and seasonal high groundwater levels.

  • General information on utility rights-of-way and land ownership.

This information can be used to eliminate structural BMPs that are not feasible based on the existing site conditions. It also can be used to estimate the preliminary design size for each remaining candidate BMP. This preliminary Federal Highway Administration design size may then be used to confirm whether the BMP will meet management objectives and whether a detailed design for the BMP should be considered. This highly iterative process may entail visits to the BMP site, division and redivision of the drainage area, and evaluation of local features and their spatial relationships in order to determine the feasibility of different structural BMP options.

It is recommended that the preliminary size of candidate BMPs be determined based on assumptions about the runoff volume from the estimated impervious surface area contributing to the BMP. For water quality BMPs, this usually means management of a volume resulting from either the water quality storm (e.g., the 1-year event) or a designated water quality depth (e.g., 12.7 mm [0.5 in] of runoff). For BMPs focused on water quantity management, this means management of a runoff volume originating from a design rainfall event (e.g., the 10-year rainfall event). For water quality and quantity BMPs, a tentative size can be estimated by assuming that all rainfall contacting the estimated impervious surface area will become runoff, all of which must be managed in the BMP. With the water quality or quantity volume, it is possible to make a first-cut estimate of the surface footprint of the BMP. For example, the surface area of a detention pond or infiltration trench can be estimated based on the runoff volume and other site information such as the depth to bedrock. Or, the surface area of an infiltration basin can be established based on the runoff volume and the general infiltration rate of the local soils. Concurrently, the surface footprint of candidate BMPs can be used to judge whether the required surface relief is present to incorporate the BMP into existing drainage ways or storm drains. Many BMPs require several feet of hydraulic drop between the BMP and any downstream outlet (Table 47). Too much or too little surface relief will prevent installation of a BMP either because the BMP cannot be tied into the existing network or because it physically cannot be constructed on the site. With approximate dimensions of candidate BMPs, it is possible to see which BMPs will fit within the existing drainage way and utility rights-of-way.

Evaluation of readily available information on the candidate project and on-site visits to assess conditions usually are sufficient to determine which BMPs are feasible. In considering a BMP, it is important to recognize that incorporating access for BMP maintenance and standard design slopes for earthwork (e.g., 3 vertical to 1 horizontal) will probably increase the required BMP surface area over that estimated as part of the preliminary design.


Effectiveness issues considered in the evaluation phase relate to management of both water quantity and water quality. In assessing constructability, it might be determined that no single BMP can fully satisfy the management objectives for both runoff quantity and quality. As a result, consideration of off-site management (e.g., management in a regional facility) might be necessary. Evaluation of site characteristics might demonstrate that it is best to manage stormwater runoff water quality on site and provide stormwater control for flood protection off site. For this reason, BMPs are frequently used in combinations of two or more; designers must determine how candidate BMPs will interact with other existing stormwater controls. For example, the designer would need to consider how a section of porous pavement will decrease the required size of a downstream infiltration trench, which also affects inflows to a downstream wet pond.

To determine which candidate BMPs will effectively remove stormwater constituents of concern, an in-depth review of prior monitoring study reports would be beneficial. In this way candidate BMPs with a history of solid performance for the pollutants of concern in locations or situations similar to the proposed project can be identified. A review of applicable monitoring studies will also help identify which BMPs may require special features (e.g., flow equalization basins) that will impose an additional cost on implementation.

The Fact Sheets provided in Chapter 3 will help designers identify any pretreatment recommendations or requirements, or seasonal limits that exist for candidate BMPs. For example, a BMP might have limited effectiveness without forebays or a presettling area or might fail to work during prolonged periods of drought or extreme climatic conditions. An in-depth understanding of the operational environment of candidate BMPs will probably decrease the number of potential candidates requiring further evaluation.

For example, assume initial screening of structural BMP options for an ultra-urban site has resulted in a list of possible BMPs that include an underground sand filter, a vegetated rock filter, and a multi-chambered treatment train (MCTT). The primary constituents of concern for water quality treatment are total suspended solids (TSS), total phosphorous (TP), and zinc. A review of ultra-urban BMP monitoring studies identifies the removal efficiencies reported in the literature for the constituents and BMPs of interest (Table 54).

Table 54. Mass-Balance Pollutant Removal Efficiencies for Constituents and BMPs of Interest
BMP Study Location Land Use Mass-Balance Removal Efficiency (%)
TSS P Zinc
Underground Sand Filter Virginia1 Parking Lot 79 63 91
Vegetated Rock Filter Florida2 Industrial 924 874 634
MCTT Alabama3 Parking Lot 83 N/A 91
1 Bell et al (1995).
2 DRMP (1995).
3 Pitt (1996).
4 Median of several reported values.

Assume the objective is to achieve a minimum of 50 percent removal efficiency for all constituents. Though the performance of the MCTT for phosphorus removal has not been established it is projected to be in the range of 70 percent. Based on this, the MCTT might be eliminated from further consideration or weighted differently than the other BMPs that have a proven record of performance.

As another example, assume a 4 ha (10 ac) redevelopment site located in the downtown of a major metropolitan area. The receiving water is a "brackish" waterbody that is phosphorus limited. The site is located over historic tidal wetlands (now fill) and the site terrain is flat. The proposed redevelopment project will cover 90 percent of the site with building and parking (surface parking is proposed for 10 percent of the site area). State water quality regulations require 80 percent removal of total suspended solids and 50 percent removal of total phosphorus.

The initial scoping analysis determined that water quantity (flooding and channel protection) was not a management goal, that nutrient load and sediment removal were the primary management objectives for controlling nonpoint source pollutants, and that the site was not considered a "hotspot." The BMPs eliminated during the broad scoping analysis were the ponds, wetlands, and the underground storage tank. The remaining BMP groups are the swales, bioretention, sand filters, organic media filters, vegetated filter strip, infiltration trench, oil/grit separator, catch basin inserts, and some of the manufactured systems.

Because the project site lies over an historic tidal wetland that was filled in the last century, infiltration is not feasible. The site's open space area is severely restricted by the proposed project so the BMPs that consume larger surface areas would not be practical. This site restriction excludes the swales, surface sand filter, bioretention, and vegetated filter strip. The remaining BMPs are the underground sand filter, the organic media filter, oil/grit separator, catch basin inserts, and some of the manufactured systems.

The pollutant removal capability of the oil/grit separator, catch basin inserts, and the manufactured systems will eliminate them from further consideration for a stand-alone BMP. However, these practices can be considered for pre-treatment for other BMPs. The remaining structural practices include the underground sand filter and the organic media filter.

6.4.2 Evaluation of Nonstructural BMPs

In the evaluation phase, limited information is available to characterize the improvements nonstructural controls may provide to the quality of receiving waterbodies. Although there is not an extensive body of information demonstrating the effectiveness of nonstructural BMPs, a significant reduction in nonpoint source constituent loads can often be achieved by controlling their sources.

Information does, however, exist for some nonstructural BMPs such as streetsweeping (see Chapter 3 ) while other nonstructural BMPs can be evaluated through programmatic assessments. For example, an annual analysis of the frequency of vehicle maintenance, reduction in metric tons of salt applied per snow event, and percent of loading docks covered in a drainage area can all be used to determine the "effectiveness" of nonstructural BMP implementation. Further analysis of nonstructural BMPs must by necessity rely on qualitative assessments. Table 55 provides qualitative constituent removal information on selected nonstructural BMPs and on their ability to improve the functioning of structural BMPs. Information from this table can be used to further refine the list of feasible nonstructural BMPs by eliminating those that do not provide stormwater constituent removal for particular constituents of concern. For example, if the primary constituents of concern generated from a drainage area are oil and grease, nonstructural BMPs such as landscaping and ground maintenance programs and covered raw material storage, among others, can be eliminated from further consideration.

6.4.3 Multiple BMP Treatment Train

A multiple BMP treatment train is a combination of BMPs to treat water quality problems. The ability of a nonstructural BMP to remove specific constituents prior to contamination of stormwater makes them ideal for combining with and enhancing the effectiveness of structural BMPs. Nonstructural and structural BMPs can be used together and structural BMPs can be combined in an ultra-urban environment to optimize pollution control. Once the structural BMP selection process has produced a narrowed set of options, the feasibility of their combination with selected nonstructural BMPs can be evaluated. The result of this analysis might be a BMP or group of BMPs specifically grouped to address the ultra-urban area in question. Nonstructural BMPs can enhance the performance of structural BMPs by preventing the entry of constituents that are difficult for structural BMPs to remove, and/or reducing the structural BMP maintenance requirements. As illustrated in Table 55, only 10 out of the listed 25 nonstructural BMPs can provide the added benefit of reducing structural BMP maintenance.

To determine whether one BMP or multiple BMPs are necessary, a few management questions should be addressed:

  • Does any one BMP meet all of the stormwater management objectives?

  • Does a structural BMP meet all of the objectives?

  • To what degree does a structural BMP meet the objectives?

  • Can an existing structural BMP be retrofitted to increase its effectiveness?

  • Does a nonstructural BMP meet all of the objectives?

  • To what degree does a nonstructural BMP meet the objectives?

  • Can an existing nonstructural BMP or nonstructural BMP management program be modified or enhanced to increase its effectiveness?

  • Does a nonstructural BMP improve the performance of a structural BMP?

  • Does a nonstructural BMP reduce the maintenance requirements of a structural BMP?

The preferred management strategy is to use nonstructural BMPs to prevent stormwater contamination rather than have to manage and treat stormwater runoff; source elimination is equal to 100 percent removal effectiveness. If the pollutant source cannot be controlled, however, the effectiveness of an appropriate nonstructural management technique to reduce its losses can be evaluated. The type of nonstructural BMP and associated pollutants controlled will dictate the need for additional pretreatment or treatment controls provided through structural BMPs. For example, if a nonstructural BMP, such as parking lot sweeping, effectively prevents some particulates from entering stormwater, a structural pretreatment facility like a vegetated filter strip might not be necessary. As another example, implementing a streetsweeping program according to specifications (refer to Chapter 3), may reduce 55 to 93 percent of dust, dirt, and particulate build-up on ultra-urban roadways and paved surfaces (NVPDC, 1992), thereby reducing the need for structural BMPS that employ sedimentation as a primary removal mechanism.

Table 55 lists a large range of nonstructural practices that can be used in conjunction with structural BMPs or by themselves to reduce stormwater pollution. A lack of monitoring information on their effectiveness in combination with structural BMPs, however, makes recommendations concerning the use of combinations of various nonstructural and structural BMPs difficult.

Table 55. Nonstructural BMP Constituent Removal Effectiveness
BMP May Provide Stormwater Constituent Removal1 Provides Direct Pre-Treatment for Structural BMP Reduces Maintenance Requirements for Structural BMP
1 2 3 4 5 6 7
Ultra-Urban Areas
Litter and Debris Removal  
Property maintenance Y Y
Proper dumpster placement         Y Y
Stream Clean-ups       N N
Frequent storm drain maintenance   Y Y
Parking lot sweeping Y Y
Education and Training  
Storm drain stenciling Y N
Employee education         N N
Landscaping and Vegetated Practices  
Landscaping/Groundskeeping programs     Y Y
Containment and Diversion  
Covered fueling stations     Y N
Covered raw material storage         Y Y
Elimination of non-stormwater discharges and connections Y N
Loading dock covers and proper location     Y N
Chemical Handling and Storage  
Hazardous materials and chemical storage       Y N
Proper hazardous materials and chemical storage       Y N
Spill control plans       Y N
Litter and Debris Control  
Road maintenance       Y Y
Streetsweeping   Y Y
Adopt-A-Road program             Y Y
Adopt-A-Stream program         N N
Pesticide and Fertilizer Application  
Pesticide application control           Y N
Landscaping and Vegetation Practices  
Mowing reduction         N N
Chemical Handling and Storage  
Road salt application and storage           Y N
Municipal fleet maintenance           Y N
Proper hazardous materials use and storage         Y N
Containment and Diversion  
Sediment and erosion control         Y Y
1(1) Nutrients, (2) Suspended solids, (3) Trash, (4)Trace metals, (5) Oil and grease, (6) Trace organics, (7) Other.

Back to top

Questions and feedback should be directed to Susan Jones (Susan.Jones@dot.gov, 202-493-2139) and Marcel Tchaou (Marcel.Tchaou@dot.gov, 202-366-4196).

HEP Home Planning Environment Real Estate

Federal Highway Administration | 1200 New Jersey Avenue, SE | Washington, DC 20590 | 202-366-4000