Long-Term Performance and Life-Cycle Costs of Stormwater Best Management Practices (2014)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

130 C H a P T E r 9 This chapter discusses the use of the BMP Evaluation Tool (located on the CD-ROM that accompanies this report) that can be used for planning-level estimates of BMP treatment performance and whole life costs. The tool is packaged as a collection of Excel workbooks (one for each BMP type) that can be used for BMP evaluation or to optimize BMP selection. An example is provided to show how the tool can be used to quickly compare the perfor- mance and cost of candidate BMPs to treat runoff from a highway site. 9.1 BMP Evaluation Tool Overview This section provides an overview of the functions, calcu- lation methodology, inputs, and results and interpretations that are common to each BMP workbook. 9.1.1 Tool Assessment Functions The tool assessment functions are to provide stormwater volumes, stormwater pollutant loads and concentrations, and costs. Stormwater Volumes. Provides an estimate of key storm- water volumes, including: • Annual stormwater runoff volume generated by the drain- age area to the BMP; • Stormwater runoff volume that bypasses the BMP; • Stormwater runoff that is captured, reduced, and released as treated effluent by the BMP; and • Total combined stormwater volume discharged to the receiving water body. Figure 9-1 illustrates a typical BMP and the relationship of these key stormwater volumes to the BMP. Stormwater Pollutant Loads and Concentrations. Pro- vides an estimate of key stormwater pollutant loads and con- centrations, including: • Annual stormwater runoff pollutant load generated by the drainage area to the BMP; • Stormwater runoff pollutant load that bypasses the BMP; • Stormwater runoff pollutant load captured, reduced, and released as treated effluent by the BMP; • Total combined stormwater pollutant load discharged to the receiving water body; • Total annual stormwater pollutant load reduction; and • Annual influent, treated, and combined effluent concentrations. Costs. Provides an estimate of whole life costs, including: • Direct and associated capital costs of designing and install- ing the BMP, • Regular and corrective maintenance costs of the BMP, and • Annualized whole life costs per annual load removed. 9.1.2 Tool Inputs The tool inputs include user-specific climate data based on closest available rain gauge, highway tributary area character- istics, and the treatment BMP design features/configuration. Rain gauges are selected based on the NCDC climate divi- sions (Figure 9-2). User-friendly features of the tool include a navigation bar to navigate to key input forms via one button click, a color-coded key to identify cell content application (i.e., instructions, headings, user data, and reference data), drop-down menus for select inputs, and built-in guidance information located directly adjacent to design values for ease of customization. Default values for climate and BMP design parameters are provided for ease of use. Section 9.3 discusses how most DOT BMP Planning Tool

131 Figure 9-2. NCDC climate divisions for tool rain gauge selection. Figure 9-1. General BMP stormwater volume routing schematic.

132 defaults are customizable by the user to adapt to site-specific needs. Appendix F: Planning Tool Handbook provides detailed information on tool organization, project set up, entering project data, and general information such as saving, editing, and printing multiple scenarios. 9.1.3 Tool Results and Interpretations Tool results are presented in a single worksheet and include: • Summary of the modeled scenario (tributary area, BMP type, rain gauge location, and precipitation depth), • Summary of design parameters (BMP type and configura- tion data), • Summary of whole life costs (capital and maintenance costs as well as WLC per load removed), • Tabular and graphical summary of volume performance (see Section 9.1.3.1), • Tabular and graphical summary of pollutant load perfor- mance (see Section 9.1.3.2), and • Tabular summary of water quality concentrations (see Sec- tion 9.1.3.3). Appendix F: Planning Tool Handbook provides detailed information on viewing and interpreting results. 9.1.3.1 Volume Performance Results The following volume performance results are provided by the tool: • Baseline Average Annual Runoff Volume—The total vol- ume of annual runoff for the site (highway) based on cli- matic region/subregion, drainage area, imperviousness, and soil type. • BMP Captured Volume—The volume of annual runoff captured by the BMP. • BMP Effluent Volume—The volume of annual runoff that is treated and released from the BMP. • Runoff Bypassed (Overflow) Volume—The volume of annual runoff not captured by the treatment BMP that bypasses or overflows directly to the receiving water body. Note that the tool conservatively assumes that overflow receives no treatment even though some limited treatment of this volume may occur. • Total Discharge Volume—The volume of annual runoff discharged to the receiving water body. This is calculated by adding the bypassed and effluent volumes. • Total Volume Reduction—The volume of annual runoff lost by the BMP through infiltration and ET. 9.1.3.2 Pollutant Load Performance Results The following pollutant load performance results are pro- vided by the tool: • Baseline Average Annual Runoff Load—The total annual pollutant load for the site (highway). This is calculated by multiplying total annual runoff volume by the character- istic highway runoff mean concentration. • BMP Captured Load—The annual pollutant load captured by the treatment BMP. This is calculated as the difference between the baseline average annual runoff load and the bypassed load. • BMP Effluent Load—The annual pollutant load from the BMP to the receiving water body. This is calculated by mul- tiplying the BMP effluent volume by the treatment BMP pollutant mean effluent concentration (computed based on influent–effluent concentration relationship). • BMP Load Reduction—The total annual pollutant load removed by the BMP. This is calculated by subtracting the BMP effluent load from the BMP captured load. • Bypassed Load—The annual pollutant load not captured by the treatment BMP and discharged directly to the receiving water body. This is calculated by multiplying the BMP bypassed volume by the characteristic highway run- off mean concentration. • Percent Annual BMP Load Removal—The percentage of annual pollutant load removed by the BMP. This is cal- culated by dividing the total BMP load reduction by the baseline average annual runoff load. • Total Discharge Load—The total annual pollutant load to the receiving water body. This is calculated by adding the bypassed load to the BMP effluent load. • Total Volume Reduction Load—The annual pollutant load removed via infiltration and ET. This is calculated by mul- tiplying the baseline average annual runoff load by the per- centage of total annual volume reduced. • Treatment Reduction Load—The annual pollutant load removed by the BMP by non-volume loss treatment pro- cesses that reduce concentrations, including adsorption, filtration, settling, decomposition, and plant uptake. This is calculated by subtracting both the total volume reduction load and the BMP effluent load from the BMP captured load. 9.1.3.3 Water Quality Concentrations The following water quality concentrations are provided by the tool: • Influent Concentration—The pollutant concentration in the BMP influent, given as default highway runoff concen- trations unless modified by the user.

133 • Treated Effluent Concentration—The pollutant concen- tration in the BMP effluent, calculated using influent/ effluent performance curves. • Whole Effluent Concentration—The pollutant concen- tration for the total discharge to the receiving water body, calculated by dividing the total discharge load by the total discharge volume. 9.1.4 Tool Supporting Data The tool provides underlying supporting data used to pro- duce the hydrologic and water quality estimates. For example, nomographs that summarize the long-term continuous sim- ulation model results specific to the user-selected rain gauge are provided. These nomographs could be used outside of the tool for additional BMP sizing and assessment purposes. Appendix F: Planning Tool Handbook provides information on viewing supporting data. 9.2 Worked Example of Tool This section applies the BMP Evaluation Tool to evaluate seven types of BMPs for a hypothetical site. Since physical site properties and climatic data are needed, the hypothetical site is assumed to be located in Daytona Beach, FL. The tool will be used to evaluate and compare seven different kinds of the BMPs assumed to be treating a 1,150-ft stretch of state high- way. The design criteria are based on mitigating the differ- ence between the pre- and post-development, 2-year, 24-hour rainfall event, which is assumed to follow an NRCS Type III distribution. The relevant physical characteristics of the drain- age area are shown in Table 9-1. All BMPs are assumed to be installed in the median between the east- and westbound travel lanes with the exception of the PFC and filter strip BMPs. A 1,150-ft section of SR 400 (Beville Road) just south of the Daytona Beach International Airport constitutes the drain- age area for this example. The drainage area is approximately 100-ft wide on average and consists of four 12-ft travel lanes and two approximately 5-ft-wide shoulders and a median of variable width, located between the east- and westbound travel lanes. For the purposes of this example, the average width of the median is approximated as 40 ft. The site has an assumed average slope of 3%. The combined pre-developed imperviousness of the ROW is approximately 60%, and site soils are assumed to be predominantly NRCS Hydrologic Soil Group C soils. Seven candidate BMP types were evaluated in this hypo- thetical example: swales, bioretention, a wet pond, filter strips, a dry detention basin, permeable friction course over- lay, and a sand filter. Cost and performance results are sum- marized and compared for each BMP type, as described in the following sections. 9.2.1 BMP Evaluation The corresponding spreadsheet tool for each BMP type was used, and the following BMP evaluation steps are repeated for each BMP type: 1. Project Location Selection The first step of the evaluation consists of filling out project description fields in the tool for the BMP under evaluation Variables Assumed Values Units Location Daytona Beach Intl Airport, FL – Candidate treatment alternatives under evaluation Swale, bioretention, dry detention basin, wet pond, filter strip, permeable friction course, and sand filter – Rain gauge ID 307167 – Climate division and name [3] NORTH CENTRAL - DAYTONA BEACH INTL AP – Elevation, feet 31 ft 85th-percentile, 24-hour storm depth 1.20 in. 95th-percentile, 24-hour storm depth 2.03 in. Average annual precipitation 49.1 in. Average slope 0.03 ft/ft Soil type Hydrologic Soil Group C – Zoning Commercial/industrial – Local design standard (90th-percentile rainfall) 0.85 in. Imperviousness 60 % Drainage area length 1,150 ft Average drainage area width (8 travel lanes, 4 shoulders, 1 median) 100 ft Drainage area size 2.6 acre Table 9-1. Worked example site properties.

134 and then selecting the project location by clicking on the region where the site under consideration is located. 2. State and Rain Gauge Selection The next step is to select the state where the site is located from a filtered list of states in the region that was selected in the previous step. For this example Florida was selected, and the available rain gauges in Florida became available in the adjacent rain gauge drop-down menu. Daytona Beach International Airport gauge was then selected. Section 5.3.2 of the 2014 Florida DOT Drainage Manual (Florida DOT, 2014) specifies either the modified rational method (for facilities with time of concentration of 15 min or less) or the Soil Conservation Service (SCS) Unit Hydro- graph Method for the design of stormwater management facilities. Florida DOT also provides intensity–duration– frequency (IDF) curves and custom precipitation distribu- tions for multiple durations. To simplify the calculations for the purposes of this example, the default NRCS Type III distribution for Volusia County, Florida, (where Daytona Beach is located) from the watershed hydrology modeling program WinTR-55 (http://www.nrcs.usda.gov/wps/portal/ nrcs/detailfull/national/water/?cid=stelprdb1042901) was used to determine the design peak flow and volume for the pre- and post-development conditions. The key inputs and outputs from TR-55 are shown in Table 9-2. 3. Optional Design Storm Override The next step in the process gives the user the ability to override the 85th-percentile rainfall depth for cases where the nearest gauge is still not representative of the site con- ditions. The site used for this example is right next to the airport where the rain gauge is located and is, therefore, a suitable location to apply values from the Daytona Beach International Airport gauge. Note that the 85th-percentile rainfall depth reported in the tool is not used in the actual evaluation of the BMP within the tool. It is included for information purposes and to allow users to apply an adjustment ratio based on the 85th-percentile depth of their specific rain gauge to adjust the volumetric percent capture results generated from the tool. 4. Project Options Beyond Step 3, the individual BMP evaluation tools con- tain sensible defaults that need to be checked for the site under evaluation. For this example, all the values in the Project Options worksheet were left at their default entries except for sales tax under the Cost Inputs section, which was changed to reflect the 6.5% sales tax rate for Florida. 5. Project Design The Project Design worksheet contains input param- eters for individual BMPs. For each of the seven candi- date BMPs under consideration for this site, BMP-specific input parameters were developed and entered into the appropriate BMP evaluation tool for that BMP. The input values used for each BMP are discussed in the subsections that follow. 9.2.2 Swales For swales, the available project area for the installation of swales was assumed to be the highway median, which was estimated to be approximately 1,150-ft long and 40-ft wide. It was assumed that one long swale could be installed in the Inputs* Rainfall distribution Type III** Land use category Urban Subarea 1 – median (open space, fair condition grass cover 50%–70%) (acres) 1.1 Subarea 2 – paved (acres) 1.5 Time of concentration (hours) 0.25 Total area (acres) 2.6 Weighted curve number 90 Outputs* 2-year pre-developed peak flow (cfs) 5.75 2-year pre-developed volume (watershed in.) 2.79 2-year pre-developed volume (ft3) 314,769 2-year post-developed peak flow (cfs) 8.4 2-year post-developed volume (watershed in.) 3.9 2-year post-developed volume (ft3) 441,698 *Available online: http://www.dot.state.fl.us/rddesign/Hydraulics/files/IDFCurves.pdf. **The Type III rainfall distribution is used here to demonstrate that the default runoff coefficients in the tool can be overridden. Table 9-2. WinTR-55 inputs and outputs.

135 median parallel to the travel lanes with outlet structures installed in the swale every 150 ft to collect treated runoff. This layout is effectively equivalent to having 8 individual swale segments (total length of 1,150/150 = 8 swales). The water quality design flow rate was assumed to be the 2-year post-development peak flow rate (8.4 cfs). The Swale Evalu- ation Tool does not directly support the evaluation of mul- tiple swales, so the total bottom width computed by the tool should be assumed to equal the combined bottom width of the 8 swales. The input assumptions for this example are shown in Table 9-3. Recall that the total water quality design flow (2-year peak flow) for the site was estimated to be 8.4 cfs. Therefore, the effective design flow for each individual segment of the swale is 1.05 cfs. Based on a design flow of 1.05 cfs, the dimensions of each segment needed to meet the design criteria (flow depth less than 4 in.) are shown in Table 9-4. 9.2.3 Bioretention For the candidate bioretention area, the post-development design volume computed for the site was 441,698 ft3, and the pre-development design volume was 314,769 ft3. The dif- ference of 126,929 ft3 was therefore entered into the Project Design worksheet of the Bioretention Evaluation Tool to obtain the results for the bioretention alternative. An addi- tional design requirement constrained the total design depth of the BMP (combined ponding depth, planning media thickness, and stone storage layer thickness) to 3 ft. A pond- ing depth of 0.5 ft, a planting media depth thickness of 2 ft, and a stone reservoir thickness of 0.5 ft were used. In reality, this bioretention area represents the aggregate total of mul- tiple smaller bioretention areas that would be installed in the available area of the median. However, note that this is a very large volume for complete infiltration and would require an excessively large footprint in Type C soils (~46% of the drainage area with an assumed 1-ft ponding depth, 2 ft of media, and 1-ft gravel storage layer). Underdrains would likely be appropriate for this site. However, unless routing-based sizing is used, the footprint would not change. The effects of routing-based sizing are pre- sented and discussed in Section 9.2.10. 9.2.4 Wet Pond or Retention Pond For the candidate wet pond BMP, the BMP is sized for the attenuation volume, which is the difference between the pre- and post-developed volumes. The difference of 126,929 ft3 was therefore entered into the Project Design worksheet of the Wet Pond Evaluation Tool to obtain the evaluation results for the wet pond alternative. The wet pond was assumed to have a 3-ft permanent pool and a 1-ft water quality surcharge depth. Other values were left at their defaults. 9.2.5 Vegetated Filter Strips Filter strip sizing is often dependent on the available area adjacent to the project right-of-way. For the purpose of this example, the Filter Strip BMP was sized to be comparable to the area used for swales. Thus, the length of the filter strip extends the length of the highway section of 1,150 ft. The width of the filter strip was calculated to achieve a hydraulic residence time greater than 10 min using Excel’s Goal Seek function. This width was calculated as 36 ft, and for compari- son, is of a similar size to the highway median. The Filter Strip Tool assumes 100% capture of the tributary runoff. 9.2.6 Dry Detention Basins For the candidate dry detention BMP, the BMP is sized for the attenuation volume, which is the difference between the pre- and post-developed volumes. The difference of 126,929 ft3 was therefore entered into the Project Design worksheet of the Dry Detention Evaluation Tool to obtain the evaluation results for the dry detention pond alternative. A maximum water quality design depth of 3 ft was assumed for the dry detention basin. 9.2.7 PFC Overlay For this site, PFC is considered an opportunistic BMP and is controlled by the available space as opposed to a required Table 9-4. Individual swale segment properties. Bottom width (ft) 8.9 Bottom length (ft) 150 Calculated flow depth (in.) 4 Calculated velocity (ft/s) 0.35 Table 9-3. Vegetated swale design inputs. Water quality design flow rate (cfs) 8.4 Bottom length (ft) 150 Effective amended soil depth (in.) 6 Underlying soil design infiltration rate (in./hr) 0.2 Fraction of runoff as lateral inflow (%) 0 Longitudinal slope (ft/ft) 0.03 Online versus offline Offline Time of concentration (min) 15 Manning's friction coefficient (n) 0.35 Horizontal/vertical side slope ratio (H:1V) 3 Water quality flow depth (in.) 4 Maximum depth (ft) 1 Freeboard depth (ft) 1

136 size. The PFC was assumed to be installed over the four 12-ft travel lanes for an approximate total installed footprint area of 1.26 acres (55,200 ft2). In this instance, the tributary area is assumed as 1.26 acres, or the total area of the highway, as no runoff is expected to enter the BMP from the median and shoulders. For this reason, it would be inappropriate to use the total tributary area of 2.6 acres for this BMP. An assumed over- lay depth of 3 in. and a footprint area of 55,200 ft2 were entered into the Project Design worksheet of the PFC Evaluation Tool to obtain the evaluation results for the PFC alternative. 9.2.8 Sand Filter The 2004 Florida DOT Drainage Handbook: Stormwater Management Facility (Florida DOT, 2004) specifies a design hydraulic conductivity (K) of 6 in./hr for sand filters (Sec- tion 4.4.3.1 of the handbook). Assuming the design criterion is to treat the change in runoff from pre- to post-developed conditions, the design peak flow (Q) for the sand filter is 2.6 cfs (post-development flow of 8.4 cfs minus pre- development flow 5.8 cfs). Assuming a media bed thick- ness (L) of 2 ft and an allowable ponding depth (d) of 2 ft above the media, the surface area was then calculated using Darcy’s equation. The footprint area of the sand filter based on Darcy’s law is approximately 12,960 ft2. The design hydraulic conductivity (K) of 6 in./hr, the maximum ponding depth (d) of 2 ft, and the media thickness (L) of 2 ft were entered into the Project Design worksheet of the Sand Filter Evaluation Tool. Then, the footprint area of 12,960 was achieved by using the Goal Seek function to calculate the appropriate storage volume to obtain the evaluation results for the sand filter alternative. 9.2.9 Comparison of Results The volume and load reduction results for the seven BMP evaluation scenarios described previously are provided in Table 9-5 through Table 9-15. Table 9-12 compares the average annual percent capture volumes as well as volume and load reductions for each. As shown, the percent capture volumes for all BMPs are above 95%, indicating that these are oversized for water quality when using the design assumptions described previously. The bioretention alternative provides the highest volume reduction and highest associated load reductions for all modeled pollutants. In this worked example, swales and dry detention basins have the lowest estimated load reduction performance. Bioretention provides the most volume reduc- tion (96.9%), followed by filter strips (72.3%) and swales (37.5%). Wet ponds, PFC overlay, and sand filters would not be expected to provide any significant volume reduction. Table 9-13 summarizes the annualized life-cycle costs per unit of volume and load reduction performance. Filter strips are estimated to have the lowest cost per cubic foot of volume reduction and are estimated to have the lowest cost per pound of pollutant removed for all modeled pollutants, followed by swales. Bioretention has the highest unit annual- ized cost per pollutant load reduced for most pollutants, but as indicated in Table 9-12, bioretention provides the most Table 9-5. Swale volume and pollutant load performance. Average Annual Volume (f3/year) Percent of Baseline Runoff Volume Average Annual Pollutant Loads Pathogens (colonies /year) Metals (lb/year) Nutrients (lb/year) Sediment (lb/year) E. Coli FC TCu TPb TZn NO3 TKN TN DP TP TSS Baseline average annual runoff 144,970 – 2.474E+15 3.572E+15 0.3780 0.3989 1.7190 9.59 21.00 30.59 2.26 4.00 1256.6 Runoff bypassed 2,160 1.5% 3.690E+13 5.330E+13 0.0056 0.0060 0.0257 0.14 0.31 0.46 0.03 0.06 18.8 BMP captured 142,800 98.5% 2.437E+15 3.518E+15 0.3723 0.3930 1.6934 9.45 20.68 30.14 2.23 3.94 1237.9 Total volume reduction 54,330 37.5% 9.271E+14 1.339E+15 0.1417 0.1495 0.6443 3.60 7.87 11.47 0.85 1.50 471.0 ET reduction 2,510 1.7% – – – – – – – – – – – Infiltration reduction 51,820 35.7% – – – – – – – – – – – Treatment reduction – – 0.000E+00 0.000E+00 0.0987 0.1185 0.7331 0.00 0.01 0.00 0.00 0.54 469.9 BMP effluent 88,470 61.0% 1.510E+15 2.180E+15 0.132 0.125 0.316 5.85 12.80 18.70 1.38 1.90 297.0 Total discharge 90,630 62.5% 1.547E+15 2.233E+15 0.1376 0.1310 0.3417 5.99 13.11 19.16 1.41 1.96 315.8 BMP load reduction – – 9.266E+14 1.338E+15 0.2403 0.2680 1.3774 3.60 7.88 11.44 0.85 2.04 940.9 Percent annual BMP load reduction1 – – 37% 37% 64% 67% 80% 38% 38% 37% 38% 51% 75% 1 Computed as the total volume reduction loads plus the treatment reduction loads divided by the baseline average runoff loads.

137 Average Annual Volume (f3/year) Percent of Baseline Runoff Volume Average Annual Pollutant Loads Pathogens (colonies /year) Metals (lb/year) Nutrients (lb/year) Sediment (lb/year) E. Coli FC TCu TPb TZn NO3 TKN TN DP TP TSS Baseline average annual runoff 144,980 – 2.474E+15 3.571E+15 0.378 0.399 1.719 9.59 21.00 30.59 2.26 4.00 1256.6 Runoff bypassed 4,540 3.1% 7.750E+13 1.120E+14 0.012 0.013 0.054 0.30 0.66 0.96 0.07 0.13 39.4 BMP captured 140,440 96.9% 2.396E+15 3.459E+15 0.366 0.386 1.665 9.29 20.34 29.63 2.19 3.87 1217.2 Total volume reduction 140,440 96.9% 2.396E+15 3.459E+15 0.366 0.386 1.665 9.29 20.34 29.63 2.19 3.87 1217.2 ET reduction 90,340 62.3% – – – – – – – – – – – Infiltration reduction 50,100 34.6% – – – – – – – – – – – Treatment reduction – – 0.000E+00 0.000E+00 0.000 0.000 0.000 0.00 0.00 0.00 0.00 0.00 0.0 BMP effluent 0 0.0% 0.000E+00 0.000E+00 0.000 0.000 0.000 0.00 0.00 0.00 0.00 0.00 0.0 Total discharge 4,540 3.1% 7.750E+13 1.120E+14 0.012 0.013 0.054 0.30 0.66 0.96 0.07 0.13 39.4 BMP load reduction – – 2.396E+15 3.459E+15 0.366 0.386 1.665 9.29 20.34 29.63 2.19 3.87 1217.2 Percent annual BMP load reduction1 – – 97% 97% 97% 97% 97% 97% 97% 97% 97% 97% 97% 1 Computed as the total volume reduction loads plus the treatment reduction loads divided by the baseline average runoff loads. Table 9-6. Bioretention area volume and pollutant load performance. Average Annual Volume (f3/year) Percent of Baseline Runoff Volume Average Annual Pollutant Loads Pathogens (colonies /year) Metals (lb/year) Nutrients (lb/year) Sediment (lb/year) E. Coli FC TCu TPb TZn NO3 TKN TN DP TP TSS Baseline average annual runoff 145,000 – 2.474E+15 3.572E+15 0.378 0.398 1.720 9.60 21.00 32.50 2.26 3.98 1258.2 Runoff bypassed 1,660 1.1% 2.833E+13 4.090E+13 0.004 0.005 0.020 0.11 0.24 0.37 0.03 0.05 14.4 BMP captured 143,340 98.9% 2.446E+15 3.531E+15 0.374 0.394 1.700 9.49 20.76 32.12 2.24 3.94 1243.8 Total volume reduction 0 0.0% 0.000E+00 0.000E+00 0.000 0.000 0.000 0.00 0.00 0.00 0.00 0.00 0.0 ET reduction 0 0.0% – – – – – – – – – – – Infiltration reduction 0 0.0% – – – – – – – – – – – Treatment reduction 140,360 96.8% 2.736E+14 1.416E+15 0.107 0.029 0.407 4.47 11.47 15.94 1.07 1.75 196.2 BMP effluent 2,979 2.1% 7.160E+12 3.730E+13 0.004 0.003 0.015 0.17 0.35 0.52 0.05 0.05 7.5 Total discharge 144,999 100.0% 3.091E+14 1.494E+15 0.116 0.037 0.442 4.75 12.07 16.83 1.15 1.85 218.2 BMP load reduction – – 2.165E+15 2.078E+15 0.263 0.361 1.278 4.85 8.94 15.66 1.12 2.13 1040.1 Percent annual BMP load reduction1 – – 88% 58% 69% 91% 74% 51% 43% 48% 49% 54% 83% 1 Computed as the total volume reduction loads plus the treatment reduction loads divided by the baseline average runoff loads. Table 9-7. Wet pond volume and pollutant load performance. significant volume reductions and treatment for all evalu- ated pollutants. The final selection of a BMP depends on the pollutants and hydrologic conditions of concern for the project along with physical and financial constraints. If bacteria and dissolved nutrients are not a concern at the site, PFC overlay could be a very cost-effective option if implemented as part of a planned pavement rehabilitation project. PFC combined with swales would be expected to have a lower combined annualized cost per volume and load reduced than bioretention alone, while still reducing 40% or more of all pollutants. However, this example only evaluates volume and load reductions. If there are effluent concentration targets for some of the pollutants, then the BMP that achieves the target while also providing the

138 lowest unit cost may be the preferred solution. Also, the sizing and related costs of the BMPs are not completely equitable because hydrologic routing was not performed. Hydrologic routing allows for BMPs to be sized to treat a target volume of runoff rather than simply requiring complete storage of a water quality event. For example, the effects of routing can be evaluated by using continuous simulation to size BMPs based on a minimum percent capture (e.g., 80%) of the average annual runoff volume. As described in the following, BMPs sized in this way result in more comparable performance and life-cycle costs. 9.2.10 Effects of Routing Most BMPs capture and treat a significant quantity of flows during a storm event. Static sizing approaches that do Table 9-8. Filter strip volume and pollutant load performance. Average Annual Volume (f3/year) Percent of Baseline Runoff Volume Average Annual Pollutant Loads Pathogens (colonies /year) Metals (lb/year) Nutrients (lb/year) Sediment (lb/year) E. Coli FC TCu TPb TZn NO3 TKN TN DP TP TSS Baseline average annual runoff 144,974 – 2.473E+15 3.571E+15 0.378 0.399 1.719 9.59 21.00 29.94 2.26 4.00 1256.6 Runoff bypassed 0 0.0% 0.000E+00 0.000E+00 0.000 0.000 0.000 0.00 0.00 0.00 0.00 0.00 0.0 BMP captured 144,974 >99.9% 2.473E+15 3.571E+15 0.378 0.399 1.719 9.59 21.00 29.94 2.26 4.00 1256.6 Total volume reduction 104,840 72.3% 1.788E+15 2.582E+15 0.273 0.288 1.243 6.94 15.19 21.65 1.64 2.89 908.7 ET reduction 6,450 4.4% – – – – – – – – – – – Infiltration reduction 98,390 67.9% – – – – – – – – – – – Treatment reduction – – 0.000E+00 0.000E+00 0.065 0.077 0.352 0.77 0.00 0.59 0.00 0.48 256.7 BMP effluent 40,134 27.7% 6.850E+14 9.890E+14 0.040 0.033 0.124 1.89 5.81 7.70 0.63 0.63 91.2 Total discharge 40,134 27.7% 6.850E+14 9.890E+14 0.040 0.033 0.124 1.89 5.81 7.70 0.63 0.63 91.2 BMP load reduction – – 1.788E+15 2.582E+15 0.338 0.366 1.595 7.70 15.19 22.24 1.64 3.37 1165.4 Percent annual BMP load reduction1 – – 72% 72% 90% 92% 93% 80% 72% 74% 72% 84% 93% 1 Computed as the total volume reduction loads plus the treatment reduction loads divided by the baseline average runoff loads. Table 9-9. Dry detention volume and pollutant load performance. Average Annual Volume (f3/year) Percent of Baseline Runoff Volume Average Annual Pollutant Loads Pathogens (colonies /year) Metals (lb/year) Nutrients (lb/year) Sediment (lb/year) E. Coli FC TCu TPb TZn NO3 TKN TN DP TP TSS Baseline average annual runoff 144,972 – 2.474E+15 3.572E+15 0.378 0.399 1.719 9.59 21.00 30.59 2.26 4.00 1256.6 Runoff bypassed 1,660 1.1% 2.830E+13 4.090E+13 0.004 0.005 0.020 0.11 0.24 0.35 0.03 0.05 14.4 BMP captured 143,312 98.9% 2.445E+15 3.531E+15 0.374 0.394 1.699 9.48 20.76 30.24 2.24 3.95 1242.2 Total volume reduction 17,810 12.3% 3.039E+14 4.388E+14 0.046 0.049 0.211 1.18 2.58 3.76 0.28 0.49 154.4 ET reduction 0 0.0% – – – – – – – – – – – Infiltration reduction 17,810 12.3% – – – – – – – – – – – Treatment reduction – – 1.839E+15 1.522E+15 0.149 0.217 0.865 1.21 3.28 4.58 0.00 1.20 771.8 BMP effluent 125,502 86.6% 3.020E+14 1.570E+15 0.178 0.128 0.623 7.09 14.90 21.90 1.96 2.26 316.0 Total discharge 127,162 87.7% 3.300E+14 1.610E+15 0.183 0.132 0.642 7.20 15.10 22.30 1.98 2.31 331.0 BMP load reduction – – 2.143E+15 1.961E+15 0.196 0.266 1.076 2.39 5.86 8.34 0.28 1.69 926.2 Percent annual BMP load reduction1 – – 87% 55% 52% 67% 63% 25% 28% 27% 12% 42% 74% 1 Computed as the total volume reduction loads plus the treatment reduction loads divided by the baseline average runoff loads.

139 not account for hydrologic routing, therefore, tend to be con- servative and result in BMPs with larger footprints or stor- age volumes than needed to provide a cost-effective level of treatment. This section briefly evaluates the effect of routing on the performance and cost of the BMPs. To size based on long-term hydrologic routing, a volumetric percent capture target of 80% was assumed for all BMPs, and a drawdown time of 48 hours was added as an additional requirement for dry detention basins and wet ponds. (Lower retention times would not be expected to provide adequate time for sedimentation-based treatment.) The BMP Evaluation Tool for each of the BMPs was used with the Goal Seek function in Excel to iteratively determine the volume or peak flow capac- ity needed to meet the volume percent capture target. Goal Seek is part of the what-if analysis tools and is used to seek a desirable value in a formula cell by changing a value of one Average Annual Volume (f3/year) Percent of Baseline Runoff Volume Average Annual Pollutant Loads Pathogens (colonies /year) Metals (lb/year) Nutrients (lb/year) Sediment (lb/year) E. Coli FC TCu TPb TZn NO3 TKN TN DP TP TSS Baseline average annual runoff 144,975 – 2.473E+15 3.571E+15 0.3779 0.3989 1.719 9.59 21.00 32.49 2.26 4.00 1256.6 Runoff bypassed 3 0.0% 4.460E+10 6.440E+10 0.0000 0.0000 0.000 0.00 0.00 0.00 0.00 0.00 0.0 BMP captured 144,972 100.0% 2.473E+15 3.571E+15 0.3779 0.3989 1.719 9.59 21.00 32.49 2.26 4.00 1256.5 Total volume reduction 0 0.0% 0.000E+00 0.000E+00 0.0000 0.0000 0.000 0.00 0.00 0.00 0.00 0.00 0.0 ET reduction 0 0.0% – – – – – – – – – – – Infiltration reduction 0 0.0% – – – – – – – – – – – Treatment reduction – – 1.732E+15 2.551E+15 0.2099 0.3528 1.481 0.00 11.92 10.79 0.49 2.23 1118.5 BMP effluent 144,972 100.0% 7.410E+14 1.020E+15 0.1680 0.0461 0.238 9.59 9.08 21.70 1.77 1.77 138.0 Total discharge 144,975 100.0% 7.410E+14 1.020E+15 0.1680 0.0461 0.238 9.59 9.08 21.70 1.77 1.77 138.0 BMP load reduction – – 1.732E+15 2.551E+15 0.2099 0.3528 1.481 0.00 11.92 10.79 0.49 2.23 1118.5 Percent annual BMP load reduction1 – – 70% 71% 56% 88% 86% 0% 57% 33% 22% 56% 89% 1 Computed as the total volume reduction loads plus the treatment reduction loads divided by the baseline average runoff loads. Table 9-11. Sand filter volume and pollutant load performance. Average Annual Volume (f3/year) Percent of Baseline Runoff Volume Average Annual Pollutant Loads Pathogens (colonies /year) Metals (lb/year) Nutrients (lb/year) Sediment (lb/year) E. Coli FC TCu TPb TZn NO3 TKN TN DP TP TSS Baseline average annual runoff 70,257 – 1.200E+15 1.730E+15 0.183 0.193 0.833 4.65 10.18 14.82 1.10 1.94 609.0 Runoff bypassed 0 0.00% 0.000E+00 0.000E+00 0.000 0.000 0.000 0.00 0.00 0.00 0.00 0.00 0.0 BMP captured 70,257 >99.9% 1.200E+15 1.730E+15 0.183 0.193 0.833 4.65 10.18 14.82 1.10 1.94 609.0 Total volume reduction 0 0.00% 0.000E+00 0.000E+00 0.000 0.000 0.000 0.00 0.00 0.00 0.00 0.00 0.0 ET reduction 0 0.00% – – – – – – – – – – – Infiltration reduction 0 0.00% – – – – – – – – – – – Treatment reduction – – 0.000E+00 0.000E+00 0.126 0.190 0.720 0.00 5.31 5.30 0.00 0.84 548.9 BMP effluent 70,257 >99.9% 1.200E+15 1.730E+15 0.057 0.004 0.113 4.65 4.87 9.52 1.10 1.10 60.1 Total discharge 70,257 >99.9% 1.200E+15 1.730E+15 0.057 0.004 0.113 4.65 4.87 9.52 1.10 1.10 60.1 BMP load reduction – – 0.000E+00 0.000E+00 0.126 0.190 0.720 0.00 5.31 5.30 0.00 0.84 548.9 Percent annual BMP load reduction1 – – 0% 0% 69% 98% 86% 0% 52% 36% 0% 43% 90% 1 Computed as the total volume reduction loads plus the treatment reduction loads divided by the baseline average runoff loads. Table 9-10. PFC volume and pollutant load performance.

140 Volume Pathogens Loads Metals Loads Nutrients Loads Sediment Loads Captured1 Reduced2 E. Coli FC TCu TPb TZn NO3 TKN TN DP TP TSS Swales 98.5% 37.5% 37% 37% 64% 67% 80% 38% 38% 37% 38% 51% 75% Bioretention 96.9% 96.9% 97% 97% 97% 97% 97% 97% 97% 97% 97% 97% 97% Wet pond 98.9% 0.0% 88% 58% 69% 91% 74% 51% 43% 48% 49% 54% 83% Filter strip >99.9% 72.3% 72% 72% 90% 92% 93% 80% 72% 74% 72% 84% 93% Dry detention 98.9% 12.3% 87% 55% 52% 67% 63% 25% 28% 27% 12% 42% 74% PFC >99.9% 0.0% 0% 0% 69% 98% 86% 0% 52% 36% 0% 43% 90% Sand filter >99.9% 0.0% 70% 71% 56% 88% 86% 0% 57% 33% 22% 56% 89% 1 The captured volume is the percent of the runoff that enters the BMP and either receives treatment and is released or is infiltrated. 2 The reduced volume is the percent of the runoff that enters the BMP and is infiltrated. Therefore, the percent of the volume treated and discharged can be computed as the difference between captured volume and the reduced volume. Table 9-12. Percent average annual volume and load reductions. Hydrologic Performance Pathogens ($/10 12 colonies) Metals ($/lb) Nutrients ($/lb) Sed. ($/lb) Volume Reduction ($/ft3 removed) Volume Capture ($/ft3 captured) E. Coli FC TCu TPb TZn NO3 TKN TN DP TP TSS Swales $0.14 $0.05 $8.10 $5.61 $31,217 $27,993 $5,447 $2,084 $952 $656 $8,838 $3,676 $7.97 Bioretention $0.67 $0.67 $39.32 $27.24 $257,366 $243,817 $56,579 $10,138 $4,632 $3,179 $42,985 $24,315 $77.40 Wet pond N/A $0.29 $19.26 $20.06 $158,649 $115,434 $32,619 $8,603 $4,667 $2,662 $37,326 $19,537 $40.09 Filter strip $0.04 $0.03 $2.44 $1.69 $12,880 $11,921 $2,732 $566 $287 $196 $2,662 $1,291 $3.74 Dry detention $1.35 $0.17 $11.18 $12.22 $122,507 $89,968 $22,266 $10,014 $4,092 $2,873 $86,613 $14,141 $25.87 PFC N/A $0.13 N/A N/A $69,966 $46,558 $12,264 N/A $1,663 $1,666 N/A $10,524 $16.09 Sand filter N/A $0.14 $11.30 $7.68 $93,282 $55,503 $13,224 N/A $1,643 $1,815 $39,758 $8,780 $17.51 Table 9-13. Annualized cost per unit of performance [whole life-cycle cost per unit, annualized (2013 dollars)]. of the reference cells. Bioretention with underdrains was also added since this may be more appropriate and cost-effective for Type C soils. Volume and load reductions for each of the BMPs sized for 80% capture (except for PFC and filter strips, which are assumed to be 100% self-treating) are shown in Table 9-14. Based on this comparison, bioretention is still a top per- former for volume and load reduction for most pollutants, but filter strips show better load reduction for several pol- lutants. As shown in Table 9-15, routing-based sizing results in huge price differences for most BMPs. Bioretention with- out underdrain changes from $0.67 to $0.11 per cubic foot captured, dry detention changes from $0.17 to $0.03 per cubic foot captured, wet ponds change from $0.29 to $0.12, and sand filters change from $0.14 to $0.05 per cubic foot captured. Unit costs for swales increase slightly for pollut- ants that are only removed by volume reduction and decrease for the other pollutants. Unit costs for PFC and filter strips

141 Volume Pathogens Metals Nutrients Sediment Captured Reduced E. Coli FC TCu TPb TZn NO3 TKN TN DP TP TSS Swales 80.0% 12.9% 13% 13% 42% 46% 70% 13% 13% 13% 13% 28% 54% Bioretention (without underdrain) 80.0% 80.0% 80% 80% 80% 80% 80% 80% 80% 80% 80% 80% 80% Bioretention (with underdrain) 80.0% 25.8% 78% 79% 60% 26% 69% 26% 34% 31% 26% 26% 73% Wet pond 80.0% 0.0% 71% 46% 53% 70% 58% 36% 31% 36% 34% 41% 65% Filter strip* >99.9% 72.3% 72% 72% 90% 92% 93% 80% 72% 74% 72% 84% 93% Dry detention 80.0% 17.1% 71% 48% 46% 57% 54% 26% 29% 28% 17% 39% 62% PFC* >99.9% 0.0% 0% 0% 69% 98% 86% 0% 52% 36% 0% 43% 90% Sand filter 80.0% 0.0% 56% 57% 44% 71% 69% 0% 45% 27% 18% 45% 71% *Note that routing has no effect on PFC and filter strips since PFC was applied to the entire available area as an opportunistic BMP and filter strips assume 100% capture. Table 9-14. Percent average annual volume and load reductions (sized with routing). Hydrologic Performance Pathogens ($/1012 colonies) Metals ($/lb) Nutrients ($/lb) Sed. ($/lb) Volume Reduction ($/ft3 removed) Volume Capture ($/ft3 captured) E. Coli FC TCu TPb TZn NO3 TKN TN DP TP TSS Swales $0.23 $0.04 $13.29 $9.26 $26,883 $23,221 $4,117 $3,408 $1,567 $1,065 $14,574 $3,811 $6.22 Bioretention (no underdrain) $0.11 $0.11 $6.33 $4.39 $41,451 $39,265 $9,113 $1,633 $746 $512 $6,923 $3,916 $12.47 Bioretention (with underdrain) $0.21 $0.07 $3.97 $2.73 $33,979 $74,866 $6,490 $3,105 $1,070 $799 $13,173 $7,423 $8.30 Wet pond N/A $0.12 $8.00 $8.51 $69,305 $50,134 $13,971 $3,991 $2,119 $1,201 $18,193 $8,563 $16.97 Filter strip* $0.04 $0.03 $2.44 $1.69 $12,880 $11,921 $2,732 $566 $287 $196 $2,662 $1,291 $3.74 Dry detention $0.15 $0.03 $2.08 $2.13 $21,234 $16,168 $3,964 $1,450 $610 $427 $9,373 $2,346 $4.72 PFC* N/A $0.13 N/A N/A $69,966 $46,558 $12,264 N/A $1,663 $1,666 N/A $10,524 $16.09 Sand filter N/A $0.05 $3.89 $2.64 $32,204 $19,102 $4,549 N/A $565 $627 $13,431 $3,010 $6.03 *Note that routing has no effect on PFC and filter strips since PFC was applied to the entire available area as an opportunistic BMP and filter strips assume 100% capture. Table 9-15. Annualized cost per unit of performance [whole life-cycle cost per unit, annualized (2013 dollars)] (sized with routing).

142 did not change because they are still based on 100% capture (an underlying assumption for these BMPs). Based on this new comparison of unit annualized costs, which is considered more equitable than the results based on static sizing assumptions, filter strips still have the lowest cost per load removed for all pol- lutants. However, the PFC and wet pond alternatives are now the highest-cost alternatives rather than bioretention. Bioreten- tion without underdrains is estimated to cost more than bio- retention with underdrains for several pollutants (E. coli, fecal coliform, total copper, total zinc, and TSS) because the costs associated with a footprint required to infiltrate 80% of the average annual volume in Type C soils (~13% of the drainage area) outweigh the costs associated with the extra infrastructure for a system with an underdrain (~3% of the drainage area). 9.3 Tool Customization The tool has been designed to be customizable to allow for overwriting of much of the default data so that users can use the best available project information for their sites. It is rec- ognized that customization will allow for each DOT to input information based on localized rainfall statistics and water quality data, as well as BMP construction and maintenance specifications, practices, and costs. Default data that are editable include precipitation infor- mation (85th-percentile storm event depth and annual aver- age rainfall depth), pollutant concentrations, BMP design parameters, and cost inputs. It is suggested that for design purposes, a local precipitation gauge and site-specific infor- mation be used to increase the accuracy of volume and pol- lutant loading results. Editable cost inputs include: • Location adjustment factor for unit costs; • Expected level of maintenance; • Discount rate; • Inflation rate; • Percent local sales tax; • Capital cost quantities and unit costs; and • Maintenance frequency, hours, labor crew size, labor rates, machinery rates, and incidental costs. 9.4 Tool Intended Uses The tool treatment performance results together with the whole life cost estimates are intended to provide DOTs with planning-level information useful for evaluating receiving water protection benefits and the magnitude of costs asso- ciated with BMP installation efforts. This type of feedback can have a number of potential applications in BMP selec- tion and design for various direct and indirect uses that are described in the following. 9.4.1 Direct Tool Uses Evaluate Volume and Pollutant Load Reduction in Comparison to Baseline Conditions and/or Performance Targets/Standards. The tool can be used to estimate the volume and pollutant load reduction (i.e., percent reduction of runoff volume and loads compared to the baseline condi- tion without controls) for a wide range of potential BMP con- figurations. The results from the tool can also be compared directly to project goals or regulatory requirements such as TMDL implementation plans or volume reduction goals. Design parameters can be adjusted in the tool to improve BMP performance and meet project goals. Quickly Compare Several BMPs for a Given Drainage Area. Once project location and tributary area have been established, the BMP workbooks can be used to evaluate different BMP types, configurations, performance, and costs to provide an understanding of the varying sizing and pollutant removal capabilities of the BMP types and to aid in choosing the most appropriate, cost-effective BMP for a given site. Evaluate Performance Relationships and Sensitivities of Design Parameters. The tool provides the ability to adjust design parameters and obtain near-immediate estimates of long-term performance (i.e., without requiring delay required to set up and run a continuous simulation model). This func- tionality can be used to evaluate performance relationships and sensitivities as well as to understand how changing design parameters affect project costs. For example, the water qual- ity benefits of increasing BMP sizing to provide 90% average annual runoff capture instead of 80% can be compared along- side the BMP costs to assess if there is a proportional benefit to increasing the average annual runoff capture. Additionally, BMP sizing can be adjusted to assess the volume and pollut- ants being captured and treated by the BMP versus the volume and pollutants that bypass or overflow the BMP. 9.4.2 Indirect Tool Uses Aid in Development of Stormwater Programs. The tool can be used to identify and establish needs and resources as part of DOT stormwater program development, including, for example, BMP land requirements, BMP costs per drain- age area to meet local regulatory requirements, and mainte- nance requirements and costs. The ability to customize input in the tool allows for easy year-to-year changes such as for inflation and tax increases. Quantify Local Precipitation Statistics. The tool con- tains the results of an analysis of 343 precipitation gauges across the conterminous United States. Key precipitation

143 statistics include the 85th percentile and 95th percentile, 24-hour precipitation depths, and average annual precipita- tion depths. These statistics are provided after the user selects the gauge that best represents the project. Establish Planning-Level Sizing Targets. At the start of the planning process, it may be useful to hold certain param- eters fixed and simply vary storage volume or footprint over a representative range to develop general relationships between BMP size and the expected performance. This can help iden- tify how much space may be needed within a site to achieve a certain goal and provide early feedback on what goals are reasonable. The percent capture nomographs can be used to evaluate the BMP sizing impacts of a higher annualized cap- ture volume. Evaluate Potential Regional Variability in Performance Associated with a Given Design Standard. By holding all other parameters fixed and changing the project loca- tion attributes, the user can quickly determine how much variability would be expected in performance as a function of project location if a uniform design standard were to be adopted across an entire jurisdiction (for example, a single design storm depth across a state).