AWRA 2003 International Congress
June 29 - July 2. 2003
BMP COST ANALYSIS FOR SOURCE WATER PROTECTION
Richard N. Koustas and Ariamalar Selvakumar**
ABSTRACT: Cost equations are developed to estimate capital, and operations and maintenance (O&M) costs for
commonly used best management practices (BMPs). Total BMP volume and/or surface area is used to predict these
costs. ENR construction cost index was used to adjust cost data to December 2000 value. Construction costs
equations were adjusted for regional differences using ENR 20 city cost index. O&M costs were further adjusted for
rainfall bias based on regional rainfall zones. Cost functions for the determination of BMP costs follow the single
determinant equation: C = aPb, where a and b are a constant coefficient and exponent, respectively, and P is the
determinant variable. The regression results showed that the capital costs of all BMPs correlate well with basin
volume (R2 > 0.70) and reasonably well with basin area (R2 range from 0.43 to 0.95) except for wet detention ponds
and grass swales. The O&M costs correlate well with basin area for grassed swales and constructed wetlands (R2 >
0.86) and not so well for wet detention ponds (R2 = 0.49). These cost equations, in conjunction with BMP
performance models, will become a useful decision-making tool for evaluation of BMPs cost-effectiveness and
placement for protection of drinking water sources.
KEY TERMS: best management practices; capital cost; O&M cost; detention ponds.
INTRODUCTION
Stormwater pollution is an extensive problem throughout the United States. Constituents frequently found
as pollutants in stormwater include visible matter, infectious (pathogenic) microorganisms, oxygen demanding
materials, suspended solids, nutrients, and toxicant (e.g., heavy metals, pesticides, and petroleum hydrocarbons).
The 1987 amendment to the Clean Water Act mandated U.S. Environmental Protection Agency (USEPA) to develop
a strategy for the National Pollution Discharge Elimination System (NPDES) for stormwater discharges. Soon after,
USEPA promulgated regulations governing permits for stormwater discharges. The Stormwater Phase II Final Rule
(USEPA, 1999a) covers stormwater discharges associated with industrial activity and discharge from storm sewer
system. Operators of the regulated activities are required to apply for NPDES permit coverage and implement
stormwater discharge management controls (e.g., BMPs) that effectively reduce or prevent discharge of pollutants
into receiving waters. No single BMP will satisfy all the stormwater control objectives. Therefore, consideration
should be given to cost-effective combinations of measures that will achieve the overall objectives of a particular
site. In order to implement effective BMPs, the operators of regulated activities need proper design guidance and
tools to estimate capital, and operating and maintenance (O&M) costs.
This paper presents the cost data and methods for calculating capital, operation and maintenance cost and
cost adjustments for commonly used urban storm runoff BMPs. The costs for retention/detention ponds, swales, and
constructed wetlands are presented in this paper.
Base Capital Costs (Construction Costs)
Base capital costs primarily refer to BMP construction costs that include expenditures for labor, materials,
and equipment. Capital cost includes costs for engineering, excavation and grading, control structure, sediment
control, landscaping, and appurtenances. Base capital costs of BMPs can vary significantly depending on site
drainage characteristics and site conditions (USEPA, 1999b). Capital costs also vary based on regional cost of living
(Ruggles, 1997). These regional differences can be found in the section on inflation and regional costs.
Cost per total BMP volume can be used to predict the overall construction costs and may be reported as
cost per total basin volume (USEPA, 1999b). Costs are calculated based on formulas that vary based on the total
* Environmental Engineers, Urban Watershed management Branch (MS-104), United States Environmental
Protection Agency, 2890 Woodbridge Avenue, Edison, NJ 08837. Phone: (732) 321 6639; Fax: (732) 321 6640; E-
Mail: koustas.richard@epa.gov

-------
volume of the storage facility. These models are typically based on cost functions. Cost functions typically follow
the single determinant formula (Novotny and Chesters, 1981):
C = aPb
where: C is the cost unit ($); P is total pond storage volume (the magnitude of the primary determinant); and a, b are
statistical coefficients determined from regression analysis.
Design, Permitting, and Contingency Costs
Other costs to consider during the construction phase include: design (engineering and administrative),
permitting, and contingency costs. Design and permitting costs include costs for site investigations, surveys,
designing, and planning of a BMP (Novotny and Chesters, 1981). Contingency costs are simply any unexpected
costs incurred during the development of a BMP. These costs are usually expressed as a fraction of the base capital
costs and have been considered as a uniform percentage for BMPs (USEPA, 1999b).
Operation and Maintenance (O&M) Costs
O&M costs incorporate the value of materials and labor needed to ensure proper operation and functionality
of a BMP facility. O&M activities are typically divided into two categories: aesthetic and functional (USEPA,
1999b). Functional maintenance includes activities needed for performance and safety; aesthetic maintenance is
primarily for public acceptance. O&M costs may include costs for site maintenance, chemicals, insurance, real
estate taxes, and plant supplies (Novotny and Chesters, 1982). O&M costs are calculated on an annual basis
throughout the life of a BMP facility. O&M costs have been expressed as a fraction of the base capital costs.
Inflation and Regional Cost Adjustments
All dollar figures expressed in various studies are presented in different year dollars. For example, the
Sears et al. (199) study dollar value is in 1995 dollars, where as those of Brown and Schueler (1997) are in 1997
dollars. In order to accurately assess the costs data used for developing the cost equation, all cost data are adjusted
for inflation and reported in December 2000 dollars. To adjust for inflation, the Engineering News Record (ENR)
construction cost index history data is used to adjust BMP cost data to December 2000 dollars.
Total capital costs for the construction of a BMP facility vary upon the region of the county or state in
which facility is located. In this study cost data is adjusted for regional cost variation by using the ENR 20 city
construction cost indexes for December 2000 (ENR, 2000). The cost data adjusted for inflation is multiplied by the
20 city index factor (city index/national index for December 2000) of the city closest to the region of study to adjust
for the regional variation.
O&M costs are also adjusted for inflation and regional variation using the ENR indexes. Since the amount
of regional rainfall may impact O&M costs, further adjustment to O&M cost is made based on a methodology
presented by the American Public Works Association (APWA, 1992).
RESULTS
All of the capital and O&M cost data used in this study came from five main sources - BMP Cost
Effectiveness Database by Tom Schueler and Whitney Brown (Brown and Schueler, 1997); Southeastern Wisconsin
Regional Planning Commission Database (SWRPC, 1991); North American Wetland Database (NAWD) (USEPA,
1993); the EPA Design Manual for Wetlands (USEPA, 1998); and Cost data Format for the Nationwide Urban
Runoff Program (NURP) Projects (MRI, 1980). These cost data are first adjusted for inflation and reported in
December 2000 dollars. They are then adjusted for regional cost variation using the methodology described above
using the nearest city index (from the 20 city ENR December 2000 index) closest to the region of study. The O&M
cost data is further adjusted for the rainfall bias using the rainfall regions described in the previous section. The cost
data from these databases adjusted for inflation and regional variation is the basis of regression analysis to develop
the coefficients for the capital and O&M cost equations. The cost functions for the determination of BMP costs
typically follow the single determinant equation: C = aPb, where a and b are a constant coefficient and exponent,
respectively, and P is the determinant variable. Linear regression for area, volume, and total cost was performed on
a log-log scale to obtain the R2 value and the relevant cost equation. A summary of results is provided in Table 1.

-------
Table 1. Regression Results and Data Points
BMP
Description
Equation
R2 Value
N
Dry Detention Ponds
Capital cost vs. area
Y = 1.50 A1180
0.86
16
Capital cost vs. volume
Y = 20.4V0'757
0.94
13
Wet Detention Ponds
Capital cost vs. area
Y = 3821A028
0.22
24
Capital cost vs. volume
Y= 385.8V0515
0.56
17
O&M cost vs. area
Y = 46.40A0'40
0.49
8
Grassed Swales
Capital cost vs. area
Y = 16.64A0 69
0.23
4
O&M cost vs. area
Y = 38.9A32
0.99
4
Wetland Data
FWS Wetlands
Capital cost vs. area
Y = 121.5A057
0.48
12
Capital cost vs. volume
Y = 53.1V0'67
0.70
5
O&M cost vs. area
Y = 2.2A0'65
0.86
6
SF Wetlands
Capital cost vs. area
Y = 1016.9A049
0.43
20
Capital cost vs. volume
Y= 55.7V068
0.70
12
Where: A = Area in ft2, V = Volume in ft3, and N = number of data points.
The cost equation developed shows that the capital costs of all BMPs correlate well (R2 > 0.70) with basin volume
and reasonably well (R2 range from 0.43 to 0.95) with basin area, except for wet detention ponds (R2 = 0.56 and
0.22) and grassed swales (R2 =0.23). The O&M costs of wet detention ponds correlate well with basin area (R2 =
0.61) but not so well for wetlands (R2 = 0.22). The O&M costs show good correlation with grass swale area (R2 =
0.99).
CONCLUSIONS
The equations predict significant construction cost decreases per unit volume or area with increasing basin
or facility size for detention/retention basins and constructed wetlands. The base capital costs of grassed swales tend
to be lower that the costs for the three other detention practices. This reduced cost can be attributed to the less
invasive nature of grassed swale construction and cheaper material costs (grass seed). The cost equation developed
on actual cost data is a rough estimate of the actual cost of the BMP=s capital and O&M costs. O&M costs are
annual costs. All O&M costs for the life of the BMPs must be reduced to present value and added to the capital cost
to obtain the total cost of the BMP over its life. This method can be used in screening process for selecting the
appropriate BMP. The actual cost of constructing any BMP is variable and depends largely on the site conditions
and drainage areas. For example, BMP constructed in a very rocky soil, the excavation cost may increase the
construction cost substantially beyond those predicted by the equation. Actual capital cost may include cost of land
acquisition which is not included here. Adjustment for inflation using the ENR cost index is a good methodology.
Regional variations in cost can be estimated using the techniques presented here. However, further studies are
needed to predict cost variation accurately within regions. Effort should be made to collect capital and O&M cost
data along with performance. A larger cost database will help to develop a more accurate cost equation.
REFERENCES
American Public Works Association (APWA). 1992. A Study of Nationwide Costs to Implement Municipal Storm
Water Best Management Practices. Southern California Chapter. Water Resource Committee.

-------
Brown, W. and T. Schueler. 1997. The Economics of Storm Water BMPs in the Mid-Atlantic Region. Center for
Watershed Protection, Elliot City, MD, 41 pp.
Engineering News Record (ENR). 2000. www.enr.com.
Midwest Research Institute (MRI). 1980. Cost Data Format for the Nationwide Urban Runoff Program (NURP)
Projects. US EPA, November.
North Virginia Regional Commission (NVRC). 1979. Guidebook for Screening Urban Nonpoint Pollution
Management Strategies.
Novotny, V. and G. Chesters. 1981. Handbook of Nonpoint Pollution Sources and Management. Van Nostrand and
Reinhold Environmental Engineering Series, New York, NY (p. 511).
Ruggles, J. 1997. Costs Considerations of Stormwater Management Facilities B the Case of VDOT. Undergraduate
Thesis in Civil Engineering, University of Virginia, Charlottesville, VA.
Sears, T.R., J.S. Bays and G.W. Medley. 1997. Development of Cost-Effective Stormwater Treatment Alternatives.
Proceedings from the Watershed 96 Conference, A National Conference on Watershed Management, Baltimore,
Maryland.
Southeastern Wisconsin Regional Planning Commission (SWRPC). 1991. ACosts of Urban Nonpoint Source Water
Pollution Control Measures@, Technical Report No. 31 Waukesha, WI. .
United States Environmental Protection Agency. 1993. North American Wetland Database.
United States Environmental Protection Agency. 1998. EPA Design Manual for Wetlands.
United States Environmental Protection Agency. 1999a. Guidance Manual For Implementing Municipal Storm
Water Management Programs: Planning and Administration. EPA-600-R-99-0121. United States Environmental
Protection Agency, Washington, D.C.
United States Environmental Protection Agency. 1999b. Preliminary Data Summary of Urban Stormwater Best
Management Practices. EPA-821-R-99-012. United States Environmental Protection Agency, Washington, D.C.

-------