EPA/600/JA-03/261
2003
COSTS OF BEST MANAGEMENT PRACTICES AND ASSOCIATED LAND FOR URBAN
STORMWATER CONTROL
David J. Sample1, M. ASCE; James P. Heaney2, M. ASCE; Leonard T. Wright1;
Chi-Yuan Fan3, M. ASCE; Fu-Hsiung Lai3, F. ASCE ;and Richard Field3, M. ASCE
ABSTRACT
New methods are used to evaluate stormwater controls and Best Management Practices (BMPs) within a land
development context. Costs are developed using published literature and standard cost estimation guides. A method
is developed in which costs are determined for each parcel within a development for specific land uses. The effect
of including the opportunity cost of land in the analysis is evaluated. Costs attributable to stormwater controls are
allocated among purposes. A method is developed in which stormwater control costs are assigned at the parcel
level. Data gaps and research needs are then explored in the context of addressing this complex problem.
KEYWORDS
Costs, cost estimate, sewer, urban stormwater management
1 Doctoral student, Department of Civil, Environmental and Architectural Engineering, University of Colorado,
Boulder, CO 80309-0428. E-Mail: sample(@,colorado.edu. and wrightl(@,colorado.edu.
2 Professor, Department of Civil, Environmental and Architectural Engineering, University of Colorado, Boulder,
CO 80309-0428. E-Mail: heanev@colorado.edu.
3 Urban Watershed Management Branch, Water Supply and Water Resources Division, National Risk Management
Research Laboratory, US Environ. Protection Agency, Edison, NJ 08837-3679. E-Mail: fan.chi(g),epa.gov,
lai.dennis(@,epa.gov, and field.richard(g)epa.gov.
ID
-------
INTRODUCTION
Urbanization has resulted in increased stormwater flow into receiving waters, increases in flood peaks, and degraded
water quality. Initially, stormwater management focused on easing flooding. Conveyance pipes and detention
facilities were used for this purpose. With increased understanding of nonpoint source pollution, which has
traditionally included stormwater sources, a holistic design of urban stormwater management systems needs to
incorporate the multiple purposes of controlling major and minor floods, as well as stormwater pollution. These
controls and Best Management Practices (BMPs) can be expensive. For example, the present value of the cost of
retrofitting water quality controls for the Atlanta Metropolitan Area alone has been estimated to be about $1.5
billion (Law 2000).
It has been suggested that implementing source controls at the onset of development is more cost-effective.
Unfortunately, data on costs and performance of these land-intensive controls are lacking. Available cost
information indicates that a significant portion of the costs depend on site-specific factors. Additionally, the multi-
purpose wet-weather flow control systems may have somewhat conflicting goals. For example, a combined sewer
system provides the dual purposes of transporting wastewater and stormwater. Storm drainage systems provide
local flood control but also are a source of water quality problems and degrade downstream receiving waters.
Stormwater detention systems serve as both quantity and quality controls. Streets transport stormwater in addition
to their main function as transportation conduits. Heaney (1997) proposed apportioning the costs of a multi-purpose
facility by designing stormwater systems meeting each single-purpose goal, and dual and multiple purpose goals.
Then the cost of the cooperative facility that meets all required goals is apportioned among the purposes using
cooperative w-person game theory.
The life-cycle costs of a stormwater control facility include the initial capital costs and the present value of annual
operation and maintenance (O&M) costs that are incurred over time, less the present value of the salvage value at
the end of the service life. Capital costs include construction, easements and land acquisition, exploration, and
engineering planning and design costs, and any additional costs for environmental mitigation. Because the value of
land is site specific, typically only construction costs are included in the analysis. Moss and Jankiewicz (1982)
promote the use of life-cycle costing to determine the best type of storm sewer pipe to buy. They point out that the
2D
-------
length of service life is difficult to estimate due to the variable material durability, in-place structural durability,
abrasive characteristics of the drainage, and corrosive characteristics of both ground water and drainage. In a case
study for Winchester, Virginia, service lives for reinforced concrete, aluminized-steel corrugated pipe, and asphalt-
coated galvanized steel pipes were estimated to be 75, 25, and 20 years, respectively (Moss and Jankiewicz, 1982).
While life-cycle costing is an essential part of a cost-effective evaluation of various stormwater control systems, data
on O&M costs are seldom available on a comprehensive basis, and were not part of this study.
This paper focuses on the life-cycle costs of on-site stormwater control facilities. The opportunity costs of land are
also estimated and included in this analysis. Unless indicated otherwise, all costs are expressed in terms of January
1999 dollars based on the Engineering News Record Construction Cost Index (ENR CCI) of 6,000 (Engineering
News Record, 1999). The methods are applied to a hypothetical urban area called Happy Hectares.
COST ESTIMATION METHODS
The traditional way to summarize cost estimating data is to approximate the total cost using a single variable power
function as shown in Equation 1. This power function is linear in the log transform. The two parameters can be
estimated from a log-log graph or found using linear regression on the log-transformed data, or using nonlinear
regression on the untransformed data.
C = «0xa' (1)
where: C = total cost, $,
x = independent variable that is some measure of component size,
a0 = coefficient, and
al = exponent.
The exponent, al, represents the economies of scale factor. If al is less than 1.0, then unit costs decrease as size
increases. A generic economies of scale factor that has been used for years is al = 0.6 (Peters and Timmerhaus
1980). When al = 1, the power function simplifies to a linear relationship and no economies of scale are present.
If al >1, then diseconomies of scale exist.
3D
-------
A key reason for the popularity of the power function approximation is that it offers a way to replace a cost database
with a single equation. This feature was very important before the widespread use of computers. The negative side
of this simple approximation is that the fit may be inaccurate. Also, total cost is seldom a function of only one
explanatory variable. For a multiple explanatory variable case, the cost estimating problem can be expressed in a
general form as:
C = f(xl,x2,...xi,...xn) (2)
where: C = total cost, and
xt = 1th independent variable.
If a database of total costs as a function of n explanatory variables is available, an approximating equation can be
developed using a variety of multiple regression approaches. The drawback to this approach is that the relationship
of total cost to several explanatory variables is seldom this simple.
Recently, the U.S. EPA funded a research project in which the methods and data for estimating capital costs for
stormwater facilities were evaluated and data were collected (Fan et al. 2000, Heaney et al. 1999a). A summary of
the capital cost functions obtained from this analysis can be found in Table 1. Data for estimating capital costs for
pipes, manholes, and pavement were obtained from R.S. Means (1996a) and based on the methods described in
Dames and Moore (1978) and Grigg and O'Hearn (1976). Data on capital costs for storage facilities can be found in
U.S. EPA (1993). Data on capital costs for BMPs can be obtained from Young et al. (1995), Schueler et al. (1992),
and ASCE (2001). Of these sources, the data on BMPs is probably the least reliable. Some ongoing research
projects are underway to improve the database of life-cycle costs and performance of BMPs. An internet database
which describes the various BMPs can be found in ASCE (2000); a summary of the cost and performance data for
selected BMPs was recently published in ASCE (2001) The State of California Department of Transportation,
Caltrans, is sponsoring ongoing research in which individual BMP life-cycle costs and performance are intensively
tracked over a period of several years (Caltrans 2000). The results of these research projects may enable BMPs to
be more easily selected based upon site-specific factors, with a set of costs and performance criteria.
4D
-------
CASE STUDY-LAND COSTS
As part of the literature review, several case studies on urban stormwater design were evaluated. The selected case
study is contained in a book on sewer design by Tchobanoglous (1981). This textbook subdivision design problem
was adapted to allow for multiple land uses, and is shown in Figure 1. Tchobanoglous develops the calculations for
designing sanitary and storm sewers for the same study area. The total area is approximately 43 hectares. The
highest part of the drainage area is on the south side. All drainage ultimately goes to a local brook. The layout of
the storm sewer system is also shown in Figure 1. The entire study area is divided into 54 sub-areas that range from
0.3 to 1.4 hectares in size. A parcel-level Geographic Information System (GIS) was developed to assist in the
analysis (Sample et al. 2001). The GIS allows the designer to query information on the parcel level that can be used
in the cost and optimization analysis. The right of way characteristics are shown in Table 2. The attributes of the
residential, commercial, apartments, and school land uses are described in Table 3.
A spreadsheet was designed to incorporate all of the necessary information to perform trial and error design, based
upon the methods developed by Miles and Heaney (1988) and refined in Heaney et al. (1999b). The hydrologic
analysis used in this procedure is based upon the Natural Resources Conservation Service (NRCS) methodology,
which is easily adaptable to the GIS, by assigning curve numbers (CN) to appropriate land uses and control options.
Urban storm drainage designs are usually sized to handle a 5- or 10-year storm. Flood control systems are typically
designed to provide protection for the 100-year storm. For this example, a five-year recurrence interval is used for
the calculations. A key component of the capital cost of control options is the value of land. However, much of the
current cost information does not include this component (ASCE 2001). Including land cost presents unique
challenges in allocating costs appropriately by function. The following sections describe a method for including
land costs in selection of BMPs. For more detail, the reader is referred to Heaney et al. (1999a).
Parcel-level cost analysis
Real estate appraisers estimate market value, which can be defined as (Boyce 1981):
The highest price in terms of money which a property will bring in a competitive and open market
under all conditions requisite in a fair sale, to the buyer and seller each acting prudently,
knowledgeably, and assuming the price is not affected by undue stimulus.
5D
-------
The present value of a series of future annual income is: D
Where
PV = present value, $,
A = annual income, $/year,
n = number of years, and
/ = interest rate per year.
(3)
As n tends to infinity, equation 3 becomes:
PVC = —
'
(4)
Where
PVC = capitalized present value of an infinite stream of future benefits.
The present value of an infinite future stream of earnings is called the capitalized value of the future income stream.
For example, a detailed investigation of the rate of return for muck farms north of Lake Apopka in Florida revealed
an expected annual return of about $1,137 per hectare (Heaney 1994). Using a discount rate of 10 percent, the
expected present value of this land would be $11,367 per hectare. Detailed studies of comparable muck farm land
indicated an average selling price of $1 1,1 19 per hectare, very close to the farm budget analysis.
Urban land use does not have a metric similar to crop productivity. However, a reliable estimate of urban land value
can be obtained by viewing the urban development as an investment opportunity. The first step is to calculate the
investment in raw land and its improvements exclusive of the building. The next step is to assume a reasonable
return on investment, say 6 percent. Thus, the annual benefit of committing this parcel of land to this use is 6
percent of the investment. The land is assumed to hold its value over time. Therefore, the present value of the
future sales price equals the original purchase price. In summary, the urban cost of committing land to this use is the
6D
-------
opportunity cost, which is estimated as the investment cost times the rate of return.
Transportation Cost
Much of the cost of urban stormwater infrastructure can be attributed to providing automobile access. A relatively
large literature is directed at estimating the true costs of various forms of transportation, particularly automobile-
related transportation. Litman (1998) summarizes this literature and recommends methods for properly estimating
the transportation costs. Heaney et al. (1999c) quantify the impact of the automobile on urban land use in general,
and urban stormwater systems in particular. Accommodating the automobile requires committing a major portion of
contemporary urban systems for streets, driveways, parking lots, garages, etc. Some costs of providing land for
transportation are paid by external subsidies from the state and federal governments. Much of the costs of local
streets and parking systems are paid by property and sales taxes. Thus, virtually none of these costs are directly
assessed on the user. This assignment of transportation costs is in stark contrast to a water utility wherein the total
cost is assigned to the users. Much of the cost for water utilities is in the form of commodity, or demand charges, so
that consumers are aware of the full cost and have direct incentives to reduce their demand. For the purposes of this
discussion, assume that a transportation utility exists in the urban area. This utility is responsible for all aspects of
transportation and parking. It must pay full costs for its network including land costs and levies this cost directly on
the transportation users.
Litman (1998) defines roadway land value as follows:
Roadway land value costs include the value of land used for rights-of-way and other public
facilities dedicated for automobile use. This cost could also be defined as the rent that users
would pay for roadway land if it were managed as a utility, or at a minimum, the taxes that would
be paid if road rights-of-way were taxed.
Housing (Construction and Land) Cost
It is instructive to trace the development of raw land into housing or other uses and then estimate the investment in
raw and improved land. Dion (1993) provides a breakdown on the components of cost for a typical house built in
1991 as shown in Table 4. Finished land and labor/materials constitute 75 percent of the total cost. If the overhead
7D
-------
and financing are prorated to the land and the house, then the land cost constitutes about 29% of total costs. The
Urban Land Institute (1989) presents another breakdown of land development costs for 1984 and 1988 as shown in
Table 5. A rule of thumb in the home construction industry is that the cost of the house should be about twice the
cost of the land. Thus, land costs are assumed to be 50% of construction costs.
A breakdown of housing costs by function for a typical medium density residential house is shown in Table 6. The
total construction cost for the house is $87,900. The total land value is estimated to be 50 percent of the cost of the
house. Each component is then allocated its value based upon the proportion of area that it occupies. Unimproved
land is assumed to be 2/3 of the total land value. The costs of improvements for water, wastewater, and stormwater
are estimated for each functional unit. For example, all of the wastewater costs are assigned to the house.
Landscaping costs depend upon several factors, including opportunity costs, soil preparation costs including topsoil,
sod, and soil conditioners, and an irrigation system. In order to determine the opportunity cost, a land valuation
analysis must be performed for each land use. A land valuation analysis for a medium density residential lot is
presented in Table 6. The area of each component of the medium density lot is listed in column 2 of Table 6. The
percentage of each component is calculated in column 3. An estimate of the cost in $/m2 is found in column 4. By
multiplying the values in column 2 by column 4, the construction cost is calculated and is shown in column 5. Next,
the percentage in column 3 is multiplied by the total of column 6 to estimate the land cost breakdown in column 6.
Column 7, the unimproved land cost, is obtained by multiplying the values in column 6 by 2/3. Using this
calculation, the value of the 334.5 m2 of land for the yard function is $26,370.
Landscaping Cost
Next, the procedure for calculating opportunity costs for landscaping must be developed, as illustrated in Table 7.
The value of $26,370 is annualized, using an interest rate of 6%, and an infinite term (as in Equation 4), to obtain
$l,582/year. Then, the present worth of 25 years of annual costs of $l,582/year is calculated using an annual
discount rate of 6%, to obtain $20,226. Dividing this value by 334.5 m2 gives $60.50/m2. This value per square
meter is used for all grass types because the underlying land value is assumed to be constant regardless of the grass
type. Landscaping costs were developed from RS Means (1996b), updated to January 1999, and are included in
Table 7 (for a medium density residential lot). The initial capital investment consists of soil preparation costs
8D
-------
including sod, topsoil, and soil conditioners and an irrigation system. For the "good" lawn, the present value of the
initial landscaping investment is $23.90/m2. Costs for lesser quality lawns drop to $18.40/m2 and $10.20/m2 for fair
and poor quality lawns, respectively. For the good lawn system, operation and maintenance costs add an additional
$15.70/m2, bringing the total to $100.20/m2.
Portion of Total Cost Attributable to Stormwater Quality Control
These unit cost estimates are preliminary in that the proper definition of costs depends on the alternatives providing
"equivalent" levels of service. For example, consider the following three options for a 557.4 m2 lawn:
• Conventional lawn with a sprinkling system
• 278.7 m2 of conventional lawn and 278.7 m2 of forest
• 185.8 m2 of conventional lawn, 185.8 m2 of forest, and 185.8 m2 of swales.
While it is possible to estimate the cost of each of these three options, the customer must view these options as
providing the same level of service in order for them to be equivalent. If the customer strongly prefers the
conventional lawn, then it is inaccurate to select other options based on lower cost if they are not perceived to be
equivalent. Further work is needed to provide more accurate assessments of equivalent landscapes. For this
analysis, the customers are assumed to be indifferent regarding the available options and simply select the least cost
combination of BMP controls. An estimated 10 percent of this total cost is allocated to Stormwater management.
This number is admittedly arbitrary but it represents our best guess at a reasonable value to assign to Stormwater
management.
Similar estimates were made for "fair" and "poor" lawns. The resulting total costs attributable to Stormwater vary
from $7.50/m2 (poor) to $8.70/m2 (fair). Better lawns are preferable from the viewpoint of being able to store more
water. However, they also cost more. A linear programming model was then used to find the least costly mix of
alternatives (Heaney et al. 1999b).
Similar land valuation estimates were made for low density residential lots, commercial, apartments, and schools.
An analogous procedure was followed for these uses, except that the commercial, apartments, and schools were
aggregated into single lots, respectively. These valuations can be found in Table 8 for low density residential,
9D
-------
commercial, apartments, and schools.
Typically, land value is not included in the analysis of cost of WWF systems. However, it is essential to include this
cost. The amount to be charged should be based on the opportunity cost of the land. This land value is an essential
part of the analysis since most of the on-site or neighborhood BMPs are land intensive, e.g., detention systems,
functional landscapes. The incidence of these costs is also a critical factor as a possible incentive to customers for
providing on-site controls and as an accurate assessment of their fair share of the total cost.
The customers of the urban WWF system can be viewed as the individual parcels served by the system. However,
this taxonomy ignores perhaps the largest generator of urban WWFs, which is, runoff from transportation land use,
especially during micro storms. Right-of-way land use comprises about 25 percent of total land use. However, it
constitutes a disproportionately large amount of the directly connected impervious area that is the major source of
runoff from small storms. Transportation systems also constitute a major portion of the WWF quality loads. Thus,
transportation systems should be included as separate customers in order to evaluate their share of the cost of the
WWF system.
Pavement and concrete costs must be included in this calculation. These costs were estimated based upon unit costs
developed from R.S. Means (1996a) for various right-of-way widths, and per m2 costs which are $37.30/m2 for
curbs, $12.90/m2 for pavement, and $3.20/m2 for sidewalks and patios. Since the area of each paved surface is
known, the total cost can be obtained by multiplying estimates of the imperviousness area. Alternatively, the length
(for right-of-way uses) may be multiplied by the unit cost factors ($/unit length) for each right-of-way. The rights-
of-way identified in Figure 1 were assigned widths based upon the following criteria:
• most streets within the development have a 15.2-meter (50-foot) right-of-way,
• the minor arterial has a 18.2-meter (60-foot) right-of-way, and
• the major arterial has a 21.3 meter (70-foot) right-of-way.
The total right-of-way costs are not just a function of pavement costs. There is an opportunity cost to devoting land
for rights-of-way instead of to development. Several different methods could be used for determining the value of
ion
-------
the right-of-way; the one selected here is to use the lowest valued land use, which is the opportunity cost for
undeveloped land for low density residential use, or $37.60/m2. This method is consistent with marginal cost
analysis. Several street profiles were analyzed. Street 1 is a standard street with curb and gutter. Street 2 is a street
with porous pavement and curb and gutter. Street 3 is a standard pavement street with swales. Street 4 is a street
with porous pavement and swales. Because the right-of-way must remain constant, the travel lane was reduced in
the case of streets using swales. These costs are added to the opportunity cost and apportioned to stormwater as
shown for all rights of way in Table 9. Streets with better infiltration characteristics are more expensive. The same
LP model as presented in Heaney et al. (1999b) selects the least costly mix.
The costs of parking, sidewalks and patios, and driveways were determined using a similar procedure. Parking lots
were evaluated in the following forms: standard pavement, and three types of porous pavement of gradually
increasing permeability. The cost increases with the increase in permeability of the parking area. A ratio of 5% was
used to apportion the costs to stormwater. Two types of sidewalks and patios were evaluated, standard and porous.
The results of this analysis are summarized in Table 9. Again, as the infiltration performance increases, so does the
cost. The two types of driveways evaluated were standard and porous, as shown in Table 9. As with the sidewalks,
costs increase as the permeability increases.
Next, runoff volumes were calculated in terms of the difference in volume between the pre-and post- development
scenarios. BMP control costs are estimated in $/m2. These costs are assumed to be the incremental costs over and
above the costs of conventional systems.
Using the procedures developed previously, unit costs for various control options were estimated for eight different
land uses: low and medium density residential, commercial, schools, apartments, and 15.2, 18.2, and 21.3 meter (50,
60, and 70 foot) rights-of- way. These unit costs, including opportunity costs, are listed in Table 9. An alternative
analysis was performed excluding the effect of opportunity costs. These results are presented in Table 10.
Using $53.80/m2 in Tables 9 and 10 results in a spread of from $0.19-0.71 per liter of storage. Costs for aspen and
woods are estimated based upon typical landscaping costs, and the computed $/liter unit costs were compared with
11D
-------
Schueler (1992) and U.S. EPA (1993) for reasonableness. The incremental cost for roofs is based on the added cost
to direct this runoff to the appropriate pervious area, the value of which was checked for reasonableness. The values
computed fell within acceptable ranges from these references, adjusted upwards to bring them to the same time
scale.
The results of the optimization can be found in Heaney, et al. (1999b) and are summarized in Figures 2, neglecting
opportunity costs, and Figure 3, including opportunity costs. The optimal total system cost, including land
opportunity costs for Happy Hectares, is $4.2 million (this result has been adjusted to reflect costs in the
Denver/Boulder, Colorado area). The total system cost, neglecting opportunity costs, is $3.9 million. This
represents approximately between 14% (including the effect of opportunity costs) and 16% (neglecting opportunity
costs) of the total $26.6 million investment neglecting opportunity costs, and $30.8 million investment including
opportunity costs. A key issue is that the allocation of a fixed percentage of costs to stormwater control needs to be
further evaluated. More work needs to be done to estimate the appropriate allocation of costs to stormwater control
for a multi-use land feature element.
SUMMARY AND CONCLUSIONS
Virtually all of the cost estimates in the literature are based on the conventional approach of fitting regression
equations to cross-sectional data on as-builts. These approaches were the only viable alternative until the
widespread availability of microcomputers. Until recently, research in the site-specific design factors of BMPs and
other control options for wet weather flows has not been done. This paper presents a possible general methodology
to analyze and use this information on a development scale level, which the authors argue is an appropriate scale for
this analysis.
The following conclusions and recommendations are evident from this research:
l.DA process-oriented approach to cost-effectiveness evaluations appears to be more appropriate for
evaluating distributed stormwater controls than are methods based on simple power function cost curves.
Curve fitting approaches to cost estimating are usually based on a very limited number of explanatory
variables and do not reflect the wide variety of factors affecting total system costs for these systems.
12 D
-------
2. D The unit cost data provided by companies such as R.S. Means are a valuable source of the necessary cost
data and should be an integral part of the overall cost-effectiveness evaluations.
3. D This method needs to be expanded to include on-site controls such as infiltration. Such an analysis is not
simple since storage routing is required at the parcel level in addition to evaluating larger storage systems.
4. D A database of flow and quality monitoring for small (40 hectares or less) catchments is needed to evaluate
actual system response for small drainage areas. These catchments can be used for overall cost-
effectiveness evaluations.
5. D The benefits of urban stormwater systems need to be better quantified. Flood damages are relatively easy
to estimate. However, stormwater quality control benefits are more elusive.
6. D The overall system evaluation should include structural and non-structural BMPs as well as conventional
storm drainage systems.
7. D A defensible method is needed to allocate costs among the many uses of developed land, of which
distributed stormwater control systems is one. The land development costs attributable to stormwater
management is a research need.
8. D Downstream receiving water impacts should be included in the evaluations.
9. D A combined sewer design should be evaluated and its cost apportioned among wastewater and stormwater.
The effect of providing additional storage in the combined sewer should be evaluated.
lO.DThe cost optimization should be refined to take into account both the broader land use optimization and
cost allocation to the level of each land use, and subsequently to each parcel. Combined with GIS, this
analysis should be performed for several different scenarios (micro storms, minor storms, and major
storms).
ILDThe impact of streets and parking as integral parts of the urban stormwater system needs to be evaluated.
Streets and parking comprise the majority of the directly connected impervious areas for stormwater
systems. Hence, they are a major source of the problem. However, they also comprise an essential element
of the stormwater management system, especially during periods of very high runoff when the sewers are
overloaded. A significant part of the cost of streets and parking is for drainage. This cost needs to be
included in the overall cost of stormwater management systems. A preliminary attempt has been made here
to quantify deleterious impacts from micro storms. More work is needed to identify these impacts, and
13 D
-------
assess an allocated, life-cycle cost of mitigating them.
14 D
-------
REFERENCES
American Society of Civil Engineers (2000) National Stormwater Best Management Practices Database, developed
under contract from the U.S. EPA, information available at www.asce.org/peta/tech/nsbdO 1 .htm and
www.bmpdatabase.org.
American Society of Civil Engineers (2001) A Guide for Best Management Practice (BMP) Selection in Urban
Developed Areas, Reston, VA.
Boyce, B.N. (1981) Real Estate Appraisal Terminology. (ISBN 0884105970) Harper Business, New York.
Caltrans (2000) Caltrans BMP Evaluation, found at www.dot.ca.gov/hq/env/stormwater/publicat/project/100.htm.
Dames & Moore (1978) Construction Costs for Municipal Wastewater Conveyance Systems: 1973-1977, US EPA
Technical Report, Office of Water, EPA 430/9-77-014.
Dion, T.R. (1993,) Land Development for Civil Engineers. Wiley-Interscience, New York.
Engineering News Record (1999) Construction Cost Index, found at www.enr.org.
Fan, C-Y, Field, R., Lai, F., Heaney, J.P., Sample, D., and L. Wright (2000) Costs of Urban Stormwater Control,
Proc.ASCE 's Joint Conference on Water Resources Engineering, Minneapolis, MN.
Grigg, N.S. and O'Hearn, J. (1976) Development of Storm Drainage Cost Functions, Journal of the Hydraulics
Division, ASCE, 102 (8):
Heaney, J.P. 1994. Conceptually Sound and Operational Methods for Environmental Valuation. Water Resources
Update, Issue No. 96.
Heaney, J.P. (1997) Cost allocation, Chapter 13 in ReVelle, C. and McGarity, A.E., (Ed.) Design and Operation of
Civil and Environmental Engineering Systems, Wiley-Interscience, New York.
Heaney, J.P., Sample, D.J., and Wright, L.T. (1999a) Costs of Urban Stormwater Systems, Final Report to the US
EPA, National Risk Management Laboratory, Cincinnati, Ohio.
Heaney, J.P., Wright, L.T., and Sample, D.J., Field, R., and Fan, C.-Y. (1999b) Innovative Methods for the
Optimization of Gravity Storm Sewer Design. Proceedings of the 8th International Conference on Urban
Storm Drainage, IAHR/IAWQ, Sydney, Australia.
Heaney, J.P., Wright, L.T., and Sample, D.J. (1999c) Innovative Methods for Optimization of Urban Stormwater
Systems, Report to the US EPA, Edison, NJ.
Law (2000) Cost Analysis and Stormwater Financing Review, presented to the Clean Water Initiative Task Force, in
15 D
-------
coordination with the Atlanta Regional Commission and the Georgia Environmental Protection Division.
Litman, T. (1998) Transportation Cost Analysis: Techniques, Estimates and Implications. Victoria Transport Policy
Institute, Victoria, British Columbia.
Miles, S.W. and Heaney, J.P. (1988) Better than "Optimal" Method for Designing Drainage Systems, Journal of
Water Resources Planning and Management, ASCE, 114, (5): 477-499.
Moss, T. and Jankiewicz, EJ. (1982) What type sewer pipe is best? Life Cycle Cost Analysis Yields Answer, Civil
Engineering, October, p. 75-76.
Peters, M.S and K.D. Timmerhaus (1980) Plant Design and Economics for Chemical Engineers. McGraw-Hill, New
York.
R.S. Means (1996a) Heavy Construction Cost Data, 10th Annual Edition, R.S. Means Company, Inc., Kingston, MA.
R.S. Means (1996b) Landscaping Unit Cost Data, R.S. Means Company, Inc., Kingston, MA.
Sample, D.J., Heaney, J.P., Wright, L., and Koustas, R. (2001) Geographic Information Systems, Decision Support
Systems, and Urban Stormwater Management, Journal of Water Resources Planning and Management,
ASCE, 127(3): 155-161.
Schueler, T. R., Kumble, P., and Heraty, M. (1992)^4 Current Assessment of Urban Best Management Practices:
Techniques for Reducing Nonpoint Source Pollution in the Coastal Zone, Metropolitan Washington
Council of Governments, Washington, DC.
Tchobanoglous, G. (1981) Wastewater Engineering: Collection and Pumping of Wastew ater., McGraw-Hill, New
York, NY.
Urban Land Institute (1989J Project Infrastructure Development Handbook, Urban Land Institute, Washington, D.C.
US EPA (1993) Combined Sewer Control Manual, EPA 625/R-93-0007.
Young, G. K., Stein, S., Cole, P., Kammer, T., Graziano, F., Bank, F (1995) Evaluation and Management of
Highway Runoff Water Quality, Technical Report for the Federal Highway Administration, Washington,
DC.
16 D
-------
Table 1: Capital cost functions for selected wet-weather controls***
Item
CMP drainage
pipe
RCP pipe
Manholes
Surface storage
Deep tunnels
Detention basins
Retention basins
Infiltration
trenches
Infiltration basins
Sand filters*
Grassed swales**
Equation
C = 0.5262£>13024Z
C = 0.1368£>16259Z
C = 1458#09317
C = 1.515*106F0826
C = 2.162*106F0795
C = 2.195 *10V069
C = 2.247 *104F075
C = 1482.864F063
C = 178.967F069
C = K,A
C = K2L
Explanatory variable
D = Diameter in cm
L = Length in m
D = Diameter in cm
L = Length in m
H = manhole height in m
V = volume of storage in M
liters
V = volume of storage in M
liters
V = volume of storage in M
liters
V = volume of storage in M
liter
V= volume of voids in m3
V= volume of basin in m3
A = impervious surface in
hectares
L = length of swale in m
Source
R.S. Means (1996a)
R.S. Means (1996a)
R.S. Means (1996a)
U.S. EPA (1993)
U.S. EPA (1993)
Young etal. (1995)
Young etal. (1995)
Young etal. (1995)
Young etal. (1995)
ASCE(2001)
ASCE(2001)
*K, is a constant ranging from 27,700-55,300. D
**K2 is a constant ranging from 16.4-45.9. D
***Costs in 1/99 $. None include cost of land acquisition. D
Note: Significant figures are necessary due to metric conversion, and do not imply the equivalent accuracy in the D
result. D
Table 2: Right-of-way characteristics of Happy Hectares
R/W
m
15.2
18.3
21.3
Length,
m
8,741.8
342.6
835.5
Curb
m
1.2
1.2
1.2
Parking
m
2.4
4.9
4.9
Landscaping
strip, m
3.0
3.0
5.5
Sidewalk
m
2.4
2.4
2.4
Traffic
Lanes, m
6.1
6.7
7.3
17
-------
Table 3: Lot characteristics in Happy Hectares
Land Use (residential)
Medium Density
Residential (14.8-19.8
Dwelling Units/Hectare)
Low Density Residential
(4.9-12.4 Dwelling
Units/Hectare)
Land Use (non-
residential)
Apartments
Commercial
School
#of
Parcels
255
51
#of
Parcels
2
6
3
Roof
Area
m2
1,600
2,000
Stories
2
1
1
Patio
m2
18.6
37.2
Parcel
Area
m2
15,113.8
44,694.0
13,880.7
Driveway
m
55.7
74.3
Roof
Area
m2
4,359.8
14,199.6
6,417.9
Land-
scaping
m
334.5
910.5
Parking
Area
m2
6,975.6
28,306.2
4,813.1
Total
Area
m2
557.4
1,207.8
Land-
scaping
m2
3,778.5
2,188.2
2,649.8
Table 4: Breakdown of the cost of a typical house, 1991$ (Dion 1993).
Item
Overhead/Profit
Financing
Finished land
Labor/materials
Total
% of Total
20
5
22
53
100
Amount, $
$24,000
$6,000
$26,400
$63,600
$120,000
Note: For the convenience of the reader, the ENR Construction Cost Index (CCI) of 1991 is 4835.
Table 5: Breakdown of the cost of housing in 1984 and 1988 (Urban Land Institute 1989).
Item
Raw land
Land improvements
Financing
Labor
Marketing
Materials
Overhead
Profit
Advertising
Other
Total*
% of development
1988$
19.3
12.6
4.4
17.4
4.3
24.1
6.5
8.1
1.2
0.4
98.3
% of development
1984$
17
7
6
18
4
29
7
9
2
2
101
*Note: The totals do not sum to 100 in the source.
18
-------
Table 6: Land valuation for medium density lot, 1/99$.
Component
Roof-house
Roof-garage
Driveway
Yard
Patio
Total
Area
m2
111.5
37.2
55.7
334.5
18.6
557.4
%of
total
20.0%
6.7%
10.0%
60.0%
3.3%
100.0%
$/m2
5.20
3.20
0.40
0.10
0.40
Construction
Cost, $
$67,500
$13,600
$2,400
$3,600
$800
$87,900
Total Land $*
$8,790
$2,930
$4,395
$26,370
$1,465
$43,950
Unimproved
Land, $**
$5,860
$1,953
$2,930
$17,580
$977
$29,300
Total land value = 0.5*construction cost.
**Unimproved land value = (2/3)*total land value.
19
-------
Table 7: Cost analysis of landscaping for medium density lot, 1/99$
Item
A. Initial Capital Investment
1. Soil preparation
Initial cost of sod
Initial cost of topsoil, 15 cm
Spreading topsoil, 15 cm
Soil conditioners
Sprinkler system
Sub-total
2. Opportunity Cost of Land
Land investment cost, $
Opportunity cost investment rate
Annual cost, $/yr.
Interest rate per year
Present worth over 25 years, $
Present worth, $/m2,
Total of initial capital investment
B. Operation & Maintenance Costs, $
Lawn watering
cm per year
% of pervious area that is irrigated
Cost of water, $/1000L
Present worth factor
Present worth
Lawn maintenance
Weeks per year
$/week
Maintenance area, m2
Present worth
Sprinkler system maintenance
Total operation and maintenance costs
C. Total Cost
Portion attributable to stormwater control
Assumed %
D. Cost for Stormwater
Input
Data
$26,370
6%
$1,582
0.06
$20,226
129
$0.40
12.78
26
$8.46
267.6
10%
Good
$/m2
$4.60
$5.40
$6.90
$0.30
$6.70
$23.90
$60.50
$84.40
$2.60
$10.50
$2.70
$15.70
$100.20
$10.00
Fair
$/m2
$3.70
$4.30
$5.50
$0.20
$4.70
$18.40
$60.50
$78.90
$1.60
$5.40
$1.60
$8.60
$87.50
$8.70
Poor
$/m2
$2.80
$3.20
$4.10
$0.10
$0.00
$10.20
$60.50
$70.70
$1.00
$3.80
$0.00
$4.70
$75.50
$7.50
20 D
-------
Table 8: Summary of cost analysis for other land uses
Land Use
Low Density Residential
Commercial
Apartment
School
Good
$/m2
$7.40
$23.50
$13.90
$27.60
$7.40
Fair
$/m2
$6.50
$22.80
$13.10
$26.80
$6.50
Poor
$/m2
$5.30
$21.60
$11.90
$25.60
$5.30
Table 9: Unit costs for controls, including opportunity costs for land, 1/99$
ID
Aspen F
Aspen G
Driveway 1
Driveway 2
Grass F
Grass G
Grass P
Parking 1
Parking 2
Parking 3
Parking 4
Patio 1
Patio 2
Roof 1
Roof 2
Sidewalk 1
Sidewalk 2
Storage
Street 1
Street 2
Street 3
Street 4
Swales 1
Swales 2
Woods F
Woods G
LDRes
$/m2
$21.50
$32.30
$2.50
$2.70
$6.50
$7.40
$5.30
$2.50
$2.70
$2.80
$3.00
$2.00
$2.00
$0.00
$16.10
$2.00
$2.00
$53.80
$2.70
$2.80
$2.90
$3.00
$32.30
$64.60
$8.60
$15.10
MDRes
$/m2
$21.50
$32.30
$2.50
$2.70
$6.50
$7.40
$5.30
$2.50
$2.70
$2.80
$3.00
$2.00
$2.00
$0.00
$16.10
$2.00
$2.00
$53.80
$2.70
$2.80
$2.90
$3.10
$32.30
$64.60
$8.60
$15.10
Commercial
$/m2
$21.50
$32.30
$2.50
$2.70
$22.80
$23.50
$21.60
$2.50
$2.70
$2.80
$3.00
$2.00
$2.00
$0.00
$16.10
$2.00
$2.00
$53.80
$2.70
$2.80
$2.90
$3.00
$32.30
$64.60
$8.60
$15.10
School
$/m2
$21.50
$32.30
$2.50
$2.70
$26.80
$27.60
$25.60
$2.50
$2.70
$2.80
$3.00
$2.00
$2.00
$0.00
$16.10
$2.00
$2.00
$53.80
$2.70
$2.80
$2.90
$3.00
$32.30
$64.60
$8.60
$15.10
Apartments
$/m2
$21.50
$32.30
$2.50
$2.70
$13.10
$13.90
$11.90
$2.50
$2.70
$2.80
$3.00
$2.00
$2.00
$0.00
$16.10
$2.00
$2.00
$53.80
$2.70
$2.80
$2.90
$3.00
$32.30
$64.60
$8.60
$15.10
RW15.2
$/m2
$21.50
$32.30
$2.50
$2.70
$6.50
$7.40
$5.30
$2.50
$2.70
$2.80
$3.00
$2.00
$2.00
$0.00
$16.10
$2.00
$2.00
$53.80
$2.70
$2.80
$2.90
$3.00
$32.30
$64.60
$8.60
$15.10
RW18.2
$/m2
$21.50
$32.30
$2.50
$2.70
$6.50
$7.40
$5.30
$2.50
$2.70
$2.80
$3.00
$2.00
$2.00
$0.00
$16.10
$2.00
$2.00
$53.80
$2.70
$2.80
$2.90
$O 1 f\
3.10
$32.30
$64.60
$8.60
$15.10
RW21.3
$/m2
$21.50
$32.30
$2.50
$2.70
$6.50
$7.40
$5.30
$2.50
$2.70
$2.80
$3.00
$2.00
$2.00
$0.00
$16.10
$2.00
$2.00
$53.80
$2.60
$2.80
$2.90
$3.00
$32.30
$64.60
$8.60
$15.10
21
-------
Table 10: Unit costs for controls, excluding opportunity costs for land, 1/99$
ID
Aspen F
Aspen G
Driveway 1
Driveway 2
Grass F
Grass G
Grass P
Parking 1
Parking 2
Parking 3
Parking 4
Patio 1
Patio 2
Roof 1
Roof 2
Sidewalk 1
Sidewalk 2
Storage
Street 1
Street 2
Street 3
Street 4
Swales 1
Swales 2
Woods F
Woods G
LDRes
$/m2
$21.50
$32.30
$0.60
$0.90
$2.70
$3.70
$1.50
$0.60
$0.90
$1.00
$1.20
$0.20
$0.20
$0.00
$16.10
$0.20
$0.20
$53.80
$0.80
$1.00
$1.00
$1.10
$32.30
$64.60
$8.60
$15.10
MDRes
$/m2
$21.50
$32.30
$2.50
$2.70
$2.70
$4.00
$1.50
$2.50
$2.70
$2.80
$3.00
$2.00
$2.00
$0.00
$16.10
$2.00
$2.00
$53.80
$2.70
$2.80
$2.90
$3.00
$32.30
$64.60
$8.60
$15.10
Commercial
$/m2
$21.50
$32.30
$2.50
$2.70
$2.70
$3.40
$1.50
$2.50
$2.70
$2.80
$3.00
$2.00
$2.00
$0.00
$16.10
$2.00
$2.00
$53.80
$2.70
$2.80
$2.90
$3.00
$32.30
$64.60
$8.60
$15.10
School
$/m2
$21.50
$32.30
$2.50
$2.70
$2.70
$3.40
$1.50
$2.50
$2.70
$2.80
$3.00
$2.00
$2.00
$0.00
$16.10
$2.00
$2.00
$53.80
$2.70
$2.80
$2.90
$3.00
$32.30
$64.60
$8.60
$15.10
Apartments
$/m2
$21.50
$32.30
$2.50
$2.70
$2.70
$3.40
$1.50
$2.50
$2.70
$2.80
$3.00
$2.00
$2.00
$0.00
$16.10
$2.00
$2.00
$53.80
$2.70
$2.80
$2.90
$3.00
$32.30
$64.60
$8.60
$15.10
RW15.2
$/m2
$21.50
$32.30
$0.60
$0.90
$2.70
$3.70
$1.50
$0.60
$0.90
$1.00
$1.20
$0.20
$0.20
$0.00
$16.10
$0.20
$0.20
$53.80
$0.80
$1.00
$1.00
$1.10
$32.30
$64.60
$8.60
$15.10
RW19.2
$/m2
$21.50
$32.30
$0.60
$0.90
$2.70
$3.70
$1.50
$0.60
$0.90
$1.00
$1.20
$0.20
$0.20
$0.00
$16.10
$0.20
$0.20
$53.80
$0.80
$0.90
$1.10
$1.20
$32.30
$64.60
$8.60
$15.10
RW21.3
$/m2
$21.50
$32.30
$0.60
$0.90
$2.70
$3.70
$1.50
$0.60
$0.90
$1.00
$1.20
$0.20
$0.20
$0.00
$16.10
$0.20
$0.20
$53.80
$0.80
$0.90
$1.10
$1.20
$32.30
$64.60
$8.60
$15.10
22 D
-------
S Manholes
Storm sewers
Parcel boundaries
Land Uses:
Apartments
Commercial
LD Residential
MD Residential
School
Figure 1: Land use in Happy Hectares (adapted from Tchobanoglous 1981)
23
-------
Costs of BMP controls
without Opportunity Costs
16%
Lot Parking
Right-of-way Landscaping
4%
Right-of-way Paving
9%
Lot Landscaping
59%
Figure 2: Summary of cost distribution for Happy Hectares, neglecting opportunity costs
24
-------
Costs of BMP controls with
Opportunity Costs
14%
Right-of-way Opportunity
Costs
19%
Right-of-way Landscaping
3%D
Right-of-way Paving
8%
Lot Landscaping
46%
Figure 3: Summary of cost distribution for Happy Hectares, including opportunity costs
25
------- |