A Watershed Approach to Mitigating Stormwater Impacts (2017)

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.

15 Defining Overall Stormwater Management Objectives Determining the regulatory requirements and identifying available BMP planning and design options is an essential step in developing criteria for selecting mitigation options. Each state or even watershed may have differ- ent requirements depending on the regulatory needs, capacity, or resources. Table 6 defines several common stormwater management objectives to which DOT projects may be subject. Table 7 lists more complex runoff management issues that DOTs may encounter in the mitigation process. Several techniques can be used to develop defensible strategies, metrics, and protocols for demonstrating and comparing the different types of mitigation approaches at the project and watershed level. This includes approaches for designing BMPs to address a specific stormwater pollutant associated with the project (in-kind mitigation) and for when offset programs or other non-conventional mechanisms are desired to provide an equivalent or greater benefit to the watershed than mitigating the direct project impact (out-of-kind mitigation). The first step in a watershed-based assessment is to define the overall stormwater mitigation regulatory requirements. This is the same whether in-kind or out-of-kind watershed approaches are desired. Second is to determine the primary issues of concern in the watershed, which predomi- nantly involve targeted pollutants in runoff and/or hydrologic conditions. A third step is quantifying the highway (project) impacts and evaluating their significance for the watershed. Table 8 presents some of the typical pollutants of concern associated with streets and highways. These are often the focus of regulatory requirements. Primary pollutants of concern are generally sediment, trash, hydrocarbons, and metals. Nutrients, pathogens, oxygen demanding substances, pesticides and herbicides, and chlorides may also be of concern depending on the local soils, pres- ence of wildlife, and road management activities. More complex impairments, such as hydromodification or alteration of the natural flow of water through a landscape, may also be a concern for the receiving water body, and flow dura- tion and/or volume control standards may be identified in the regulatory requirements. In areas prone to flooding, allowable peak flow release rates may be regulated, requiring runoff storage volumes and extended detention times. This can be an issue for road projects which can gen- erate a substantial increase in runoff volumes compared to pre-project conditions. Areas that require stream protection may also require flow controls to limit the frequency of geomorphi- cally important discharges to control erosion, prevent habitat degradation, and to protect public infrastructure. Volume control requirements may be required as part of low-impact develop- ment stormwater management in which volume control via infiltration and evapotranspira- tion (ET) contribute to a range of environmental improvements including pollutant reduction, Methods to Develop Mitigation Options C H A P T E R 3 To find the WBSMT, go to the TRB website and search for NCHRP Research Report 840.

16 A Watershed Approach to Mitigating Stormwater Impacts Category Typical Objectives of DOT Stormwater Management Projects Hydraulics Manage flow characteristics within and downstream of stormwater system Hydrology Mitigate floods; groundwater recharge; improve runoff characteristics (peak shaving and/or volume reduction); improve stream channel protection Water quality Reduce downstream pollutant loads and concentrations Avoid/minimize downstream temperature increases Achieve a target reduction of a not-to-exceed concentration or load in outflow Toxicity Reduce acute or chronic toxicity of runoff Regulatory Comply with NPDES permit, 401 certification, ESA, CERCLA, RCRA, UIC Program Meet local, state, or federal water quality, flow, and flood control criteria Mitigate development in other areas Create banking credits or advance mitigation for scheduled projects Address/avoid third party CWA and other lawsuits Implementation Function within management and oversight structure Cost Minimize capital, operation, and maintenance (life-cycle) costs Maintenance Operate within maintenance and repair schedule and requirements Design system to allow for retrofit, modification, or expansion Longevity Achieve long-term functionality (e.g., flow rates, capacity/volumetric storage, level of service) Resources Improve downstream aquatic environment/erosion control Improve wildlife habitat Replace aging infrastructure, bring infrastructure up to current standards, or modify/ replace infrastructure to handle current conditions Maintain and restore beneficial uses or achieve multiple-use functionality Safety, risk and liability Function with minimal environmental risk downstream Contingency for spill containment Address vector control, geotechnical stability, or other safety issues Public perception Clarify public understanding of runoff quality, quantity and impacts on receiving waters Source: Strecker et al. 2005 Table 6. Urban runoff management objectives. Objective Component Hydrology Hydrograph alteration (increased peak flow and volume) Floodplain presence and connectivity Flood storage Groundwater Habitat diversity and quality Floodplain quality and connectivity Hydraulic diversity Riparian condition Stream condition and morphology/type Stream connectivity Habitat types Bank and stream bed erosion Channel substrate (fine/coarse) Refugia (depth, boulders, woody debris) Water quality Water temperature Dissolved oxygen Nutrients and chlorophyll a Total suspended solids/turbidity Toxic contamination of water, sediments, and biota Groundwater quality Biological communities Biotic integrity Benthic communities Riparian wildlife Terrestrial wildlife Aquatic life Table 7. Complex urban runoff management topics common to DOT projects.

Methods to Develop Mitigation Options 17 groundwater recharge, flood reduction and stream protection, and reduced stormwater infra- structure capacity requirements. Potential runoff pollutant loads and pollutant concentrations, rates, or discharge volumes can be related to highway attributes such as the volume of traffic, roadway classification, or imper- vious surface area. This information allows DOTs to evaluate the relationships between the attributes and the pollutant loads and/or hydrologic changes. From this, an impact assessment can be prepared to help identify stormwater treatment options and help quantify and predict the long-term effects of highway construction on water quantity and quality. Characteristic highway runoff quality data for assessment purposes can be obtained from the Highway-Runoff Database (HRDB) (Granato and Cazenas 2009) and the National Stormwater Quality Database (NSQD) (Pitt 2008). Criteria for Selection of In-Kind BMPs A consistent evaluation process that uses the most frequently implemented structural BMPs by state DOTs is critical for selecting on-site and off-site BMPs. There are several key techni- cal elements of the BMP evaluation and selection process that are recommended for selecting, sizing, and evaluating on-site and off-site stormwater management mitigation options. While there are non-technical elements to BMP selection, such as public preference or stewardship requirements, these tend to be local or regional in nature. The evaluation and selection process (and potential work flow path) relies on providing guidance on the following, and is relevant at both the DOT project and watershed scale: • Pollutant type and source identification – The existing and potential pollutant contributions to the BMP. • Runoff and pollutant load estimation – Increased or decreased runoff volumes, peak flow rates, and pollutant loads from the post-construction condition of the project will be used to determine the treatment requirements or goals at the watershed scale. Many models are avail- able for estimating pollutant load including the Stochastic Empirical Loading and Dilution Model (SELDM), which was developed by the U.S. Geological Survey (USGS) and the Federal Highway Administration (FHWA) as a quantitative highway-runoff quality planning model that can be used to simulate storm flows, concentrations, and loads. Table 8. Typical pollutants of concern for highway runoff. Pollutant Potential Sources Gross solids, sediment and floatables Highways, roads, construction activities, atmospheric deposition, drainage channel erosion Pesticides and herbicides Applied to greenways and roadsides, utility ROWs, commercial and industrial landscaped areas, soil wash-off, drift from adjacent land applications Organic materials/oxygen demanding substances Landscaping, animal wastes, other organic substances such as detritus Metals Automobiles, bridges, atmospheric deposition, industrial areas, soil erosion, corroding metal surfaces, combustion processes Oil and grease/PAHs/ organics associated with petroleum Roads, parking lots, vehicle maintenance areas, illicit dumping, automobile emissions Total dissolved solids (salts) Deicing applications Nitrogen and phosphorus Fertilizers, atmospheric deposition, automobile exhaust, soil erosion, animal waste, detergents Source: Oregon State University et al. 2006

18 A Watershed Approach to Mitigating Stormwater Impacts • BMP selection, sizing, and performance – The appropriate type and size of BMP to mitigate the pollutants of concern will be based on the scale of implementation, site-specific conditions, recognized treatment processes, and calculation results that are acceptable to the regulatory agencies. This can include determination of the effect of the BMP systems(s) on receiving water loadings and the consistency, reliability, and performance of the system to evaluate the equivalency of an off-site alternative. • Implementation and management – Constructability and management of different systems must be considered to achieve efficient and effective system design. This can include on- and off-site determination or a combination of both. Methodology for Evaluating BMPs BMPs are designed with a set of metrics for stormwater mitigation that are based on the pollut- ants of concern and regulatory requirements (Note that the use of percent removal to characterize BMP performance can be problematic and should be avoided or used with caution. See discussion provided here: www.epa.gov/npdes/three-keys-bmp-performance-concentration-volume-and- total-load#why). Whichever mitigation strategy is selected will require the use of one or more metrics to evaluate the effectiveness of the on-site or alternative mitigation strategy at meeting the regulatory requirement. These metrics may include measurements of and/or benchmarks for the percentage removal of pollutants or average effluent quality, runoff per impervious acres captured and treated, runoff volume reduction, peak flow reduction, and/or maximum load or concentration discharged. These measurements are referred to as “currency” for the purposes of establishing the required alternative mitigation on a watershed basis. The term “currency” is often used in the mitigation banking context to describe the manner in which impacts and uplift are measured and used to define debits/credits. Other types of mitigation within the watershed, such as stream stabilization or wetland creation, may also be considered in lieu of, or in addition to, BMPs. The watershed-based approach may provide an opportunity for DOTs to demonstrate the effectiveness of alternative stormwater mitigation strategies based on appropriate currency. This mitigation can be considered an offset for on-site mitigation, or at a minimum, an equivalent mitigation measure to what would have been planned on-site. Preliminary BMP Sizing for Planning Level Analysis Once an appropriate BMP is selected, preliminary sizing of the BMP must be completed to estimate space requirements, cost, and performance. Calculating the size of the BMP also allows physical site constraints to be more fully evaluated. For most BMPs, a good selection of state-specific guidance and design manuals exist. Sizing information and project-specific BMP guidance may also be found from local regulatory agencies. Relevant literature is cited in the References section. Reduced Maintenance Costs with Implementation of a Watershed Approach to Stormwater Management The Moving Ahead for Progress in the 21st Century Act (MAP-21) requires asset and risk management for pavement and bridges and encourages DOTs to develop these approaches for other infrastructure as resources permit (www.fhwa.dot.gov/tpm/). Declining resources relative to the miles of infrastructure and number of facilities DOTs maintain has limited the ability of DOTs to perform maintenance on post-construction BMPs. DOTs are now engaged in a sub- stantial effort to improve and extend both the maintenance they perform and their systems for tracking such work as a means of achieving water quality goals with limited resources allocated for maintenance.

Methods to Develop Mitigation Options 19 One innovative technique DOTs have employed is to seek off-site partners and/or mitigation banks to care for off-site mitigation over the long-term. DOTs have also sought opportuni- ties to sell “excess parcels” of/in the ROW that may be maintained for natural resource or other non-transportation corridor needs (Venner Consulting and ICF International 2012). Such sites have often been developed for wetland creation and restoration for mitigation purposes; DOTs may take a similar approach to develop these sites for water quality miti- gation purposes. Overall, a watershed approach may be expected to alleviate demands on maintenance forces that would occur with increasing installation of post-construction BMPs in the ROW (as in a circumstance where all stormwater treatment needed to occur on-site), especially in ultra-urban areas. Key Issues for Evaluating the Use and Effectiveness of BMPs Any process for evaluating on-site, retrofit, and off-site stormwater runoff management options must recognize current DOT requirements (or preferences) for using on-site BMPs. It must also integrate this requirement into BMP selection and design approaches. This includes the following: • Identifying on-site options and frequency of BMP use among DOTs. A candidate set of BMPs and assessment tools (e.g., models, spreadsheets, nomographs) must be used to test the proto- cols for BMP selection. A list of potential BMPs and assessment tools for initial investigation should be identified. This list may expand or contract as the research progresses. • Incorporating a process and framework to assess the suitability of sites for on- or off-site stormwater management and mitigation. The process should provide a decision support framework that can be used by DOTs to evalu- ate whether on-site BMPs are adequate and preferable to what can be achieved off-site. Such a framework would enable DOTs to evaluate whether potential off-site measures can compensate for the potential impacts of highway runoff. Issues and Challenges There is a tremendous amount of variability in determining BMP effectiveness. The datasets and methods used range from simple equations that require minimal data and training to conduct and review projects to complex, long-term studies that require a high level of knowledge and experi- ence. The analysis of BMP effectiveness is typically based on empirical data related to pollutant loads and BMP pollutant removal performance. These can be used to develop a comprehensive tool to determine pollutant loads and the appropriate water quality mitigation BMP strategies and techniques. These provide a straightforward approach that can be easily understood, are universally recognized, and are easy to modify for DOT use. There is no comprehensive knowledge base regarding the effectiveness of BMPs to mitigate for all known pollutants of concern to DOTs, but this can be overcome by utilizing information on common pollutants for DOT projects, high performing BMPs, and climate conditions. This approach allows DOTs to determine the pollutant effectiveness and loading that can be used as currency for offsets and trading. On-Site BMP Assessment Process Figure 1 summarizes the recommended steps for evaluating the potential effectiveness of on-site or project related BMPs using an on-site stormwater mitigation process. These steps include selecting and sizing on-site BMPs and entering data requirements for the candidate processes. The five steps are further detailed here.

20 A Watershed Approach to Mitigating Stormwater Impacts Step 1: Identify Appropriate and Acceptable BMPs Regulatory requirements and guidance often dictate what design standard should be met (Caltrans 2010). The selection process involves identifying the reduction or mitigation of typi- cal pollutants of concern to a predetermined value or, if no value is set, to the maximum extent practicable. These are based on the scientific and technical attributes of the regulations. The location and type of setting for the BMPs have a significant influence on the selection pro- cess (Strecker et al. 2005). Two primary categories of BMP locations and settings are used: on-site and off-site. On-Site Treatment Options. There are a number of on-site (e.g., traditional in-kind) treat- ment BMPs available as these are well known practices used for stormwater runoff management within the ROW (e.g., constructed wetlands, detention ponds, retention ponds, bioretention, and proprietary device; see also Oregon State University et al. 2006). The location of on-site BMPs is typically constrained due to the physical restrictions of the drainage area or the limited amount of ROW at the construction location. The amount of available space for the BMPs, especially in urban areas, can severely restrict the options. BMP construction may also impact wetlands, waterways or other protected resources. Another on-site BMP option is installing new stormwater retrofits or upgrading existing stormwater controls for the existing roadway. In these instances, a DOT can enhance or modify existing stormwater management and conveyance systems within the project area to improve water quality mitigation. Examples include replacing concrete swales with vegetated treatment swales or modifying the outlet structure of an existing flood manage- ment pond to provide water quality treatment. Figure 1. On-site BMP assessment process. B M P Selection and Sizing B M P Perform anc e Cost Benefit STEP 1 Identify Appropriate and Acceptable On-Site BMPs STEP 2 Estimate Preliminary Size of BMP Using Accepted Sizing Methods STEP 3 Estimate Performance (Load Reduction) of Appropriately Sized On-Site BMP STEP 4 Estimate Load Reduction Cost Ratio of Appropriately Sized On-Site BMP Im plem entatio n and M anag em ent STEP 5 Identify Factors/Constraints that May Influence On-Site vs. Off-Site Implementation

Methods to Develop Mitigation Options 21 Off-Site Treatment Options. Off-site options to treat stormwater that originates outside the project footprint may be necessary to meet mitigation objectives where on-site BMP options are limited or where on-site designs achieve only part of the treatment goals. The kinds of BMPs used in off-site treatment are typically the same ones used in on-site treatment (e.g., in-kind), but applied where ROW or other opportunities are less constrained and there is a greater degree of flexibility in selecting the location and setting. For retrofit scenarios, stormwater treatment and management can result in partial or complete off-site BMP implementation. Examples include placing BMPs in other off-site ROW areas or identifying potential mitigation opportunities that are both off-site and outside of the ROW. Such options may require the purchase of additional ROW and easements and may involve additional environmental permits. The selection process for off-site treatment options should be based on the nature of the target pollutants and parameters (e.g., temperature and/or hydromodification) in relation to specific stormwater management goals. Most treatment facilities mitigate for more than one impact. For example, dry extended detention basins may reduce the total runoff volume due to infiltration and ET, as well as attenuate peak flows and facilitate particulate settlement. Some BMPs can be modi- fied to include unit processes that are typically not incorporated in their design, such as including amended soils to promote retention and infiltration/evapotranspiration in a vegetated swale. It is important to note that some stormwater BMPs such as vegetated swales may be used as either primary or secondary (e.g., pretreatment) components of a treatment train where the BMPs in the train may be focused on treating different pollutants. It may be more useful to categorize BMPs (and their components) according to the unit treatment processes that they provide (Urbonas 2002). Step 2: Estimate Preliminary Size of BMP Using Accepted Sizing Methods There are several quick and straightforward methods available to determine the type and size of the most commonly used DOT BMPs for the preliminary planning and selection process. A more comprehensive and in-depth analysis can be used to verify the selection and complete the conceptual design. The determination of BMP size is typically based on the flow rate and/or volume capacity of the BMP. Volume-based sizing is often used for BMPs that depend on a storage volume for pro- viding stormwater control. Flow-based sizing is typically used for BMPs that depend on flow rate (such as filters) or control of a peak rate for providing stormwater control. These two categories of BMPs can be used to accomplish the following stormwater mitigation objectives: • Flood management sizing. Flood management benefits from BMPs are attained by either vol- ume reduction or peak attenuation. The primary process for volume reduction is infiltration and, to a lesser extent, ET. Peak attenuation is achieved by capturing inflows at a high flow rate while releasing outflows at a lower flow rate and providing ample storage capacity to store the difference. This implies that BMPs with smaller storage volumes and no infiltration provide little or no flood management benefits. If flood management is a goal of the BMP implementa- tion, then selection and sizing should include volume-based BMP sizing and infiltration BMPs. Depending on the BMP, sizing for flood management will likely affect sizing for water quality and flow duration control. • Water quality sizing. Water quality treatment from BMPs is attained via the various unit operations or processes. A number of these processes depend on holding time and contact time to provide treatment. Treatment typically increases with increases in these two variables. Water quality sizing therefore seeks to maximize physical elements in the BMP that contribute to holding times and contact times. Many types of BMPs can be installed in offline configura- tions to prevent damage to the BMP from large runoff events. If flood management is a goal,

22 A Watershed Approach to Mitigating Stormwater Impacts using an offline configuration may not be an option and/or might require additional consid- erations for energy dissipation and overflow. • Flow duration sizing. Flow duration control is attained by a combination of volume reduction and peak attenuation. Flow duration sizing aims to match the distribution (or portion thereof) of post-development flow rates to the distribution of pre-development flow rates. Matching the distribution of flow rates is more challenging than matching just peak flows because post- development runoff volumes are typically higher than pre-development volumes due to the increased imperviousness. • Opportunistic BMP sizing. These are BMPs that are selected based on the location (space) or the project’s mitigation potential. Opportunistic BMP sizing is driven more by the on-site conditions rather than sizing requirements and is simply sized to fit within the available space. The first three of these objectives are largely driven by regulatory requirements and are designed to achieve protection of the receiving environment by reducing changes in stream morphology and water quality. The goal for opportunistic BMP sizing provides a link to establishing banking oppor- tunities for other ongoing or scheduled projects in the same watershed. The size of a BMP for any of these will be based on the area needing treatment and the design criteria (for opportunistic BMPs, sizing includes the BMP that will fit on-site plus banking or other mitigation equivalent required). Flow and volume-based BMP sizing can be determined using relatively simple calculations for planning level analysis. Commonly used national models include NRCS TR-55 or regression equations. The goal of any watershed-based decision-making framework is to ensure that the proposed assessment and calculation methods are consistent with common approaches and that the results can be easily “translated” to these methods. One of the challenges is to develop assessment methods with an appropriate length of analysis and time step. This includes synthetic design events, continuous simulation, and actual rainfall data over a select period of time (for continuous simulation). This stepped approach connects general engineering equations for sizing management sys- tems based on runoff linked to pollutant loads and BMP effectiveness and efficiency (as shown in Step 3). This helps to establish the straightforward method for determining sizing and per- formance expectations for a project and allows DOTs to adapt the approach to their specific mitigation needs. The first calculations determine the precipitation amounts that are initially needed for volumetric runoff determination. This data is readily available from NCEI. Using annual precipitation data with drainage area imperviousness will estimate annual runoff vol- ume. Combining this volume with average annual pollutant concentration based on each land use can estimate average annual loading as shown in Equation 1. ∑= = = L (1) 1 VCW i ii i n where L _ W is the average annual load from the tributary drainage area, V _ i is the average runoff volume for land use i, and C _ i is the average runoff concentration for land use i. Average annual runoff volume can be estimated by first estimating a runoff coefficient for each land use based on its average imperviousness. Equations 2 and 3 can be used to estimate these. This equation is com- monly used to make rough estimates of runoff (Schueler 1987 and Ohrel 2000). Other methods, such as USGS regression, are often used in larger watersheds (Granato and Cazenas 2009). ( )= +0.05 0.9 (2)R Impv ii where Rvi is the volumetric runoff coefficient and Impi is the impervious fraction for land use i. The runoff volume for each land use can then be estimated as: = (3)V R A Pi v ii

Methods to Develop Mitigation Options 23 where V _ i is the total runoff volume and Ai is the total area for land use i. P is the average annual rainfall depth. The average pollutant concentrations in runoff from each of the land uses may be estimated using data summarized from the HRDB (Granato and Cazenas 2009) and the NSQD (Pitt 2008). The application of Equations 2 and 3 is referred to as the “Simple Method” (Schueler 1987). Step 3: Estimate Performance of Appropriately Sized BMPs Pollutant load modeling can be used to determine the BMP performance after the candidate BMP or BMPs are determined. The projected performance can be used in the final design or to determine if off-site alternatives are needed. This requires an estimation of the quantity and quality of runoff from discrete drainage areas. These drainage areas are defined by topography and the drainage system. Stormwater runoff can be estimated using simple empirical formulas or physically based hydrologic models whereas runoff quality can be estimated empirically. A simple presentation of the process that is used to determine the performance of the BMP will be required to express the loads or currency that can be compared for on-site and off-site options. BMP load reduction from a project can be estimated with data regarding the influent volume, BMP treatment volume, event mean concentration, and estimated bypass volume. Data sources such as the NSQD can be used to estimate land use concentrations while basic runoff genera- tion using the simple method can calculate runoff loading (Pitt 2008). Continuous simulation is recommended for BMP design to estimate BMP bypass, to which a runoff concentration can be applied to determine bypass loads. Finally, event mean effluent concentrations from the International BMP Database along with continuous simulation flow data can be applied to BMP effluent to determine treated effluent loading. The NCHRP BMP Evaluation Tools (Taylor et al. 2014) utilize continuous simulation results along with the datasets identified earlier to allow users to quickly estimate potential load reductions from on-site BMPs. Step 4: Estimate Load Reduction Cost Ratio of Appropriately Sized On-Site BMP Load reductions are determined through the use of the performance determination in Step 3. Applying planning level costs based on size and type of application (new construction, retrofit, or redevelopment) for various BMP types is an efficient way of developing cost-benefit ratios. Typical area-weighted BMP costs are summarized in Table 9. This summary data is useful for cost com- parisons and can be customized based on location, site conditions, or other notable cost variables. It is recognized that this cost information is constantly in flux and is dependent on the scale, com- plexity, and location of the project. It is also recognized that cost information may be developed BMP Cost per drainage area (BMP serving less than 3 acres) ($/ac) Cost per drainage area (BMP serving more than 3 acres) ($/ac) New Construction ($) Retrofit ($) New Construction ($) Retrofit ($) Sand filter 87,953 113,835 48,136 62,301 Cartridge filter 163,884 201,521 153,039 188,186 Swale 19,499 37,460 2,287 4,394 Strip 11,147 30,940 1,890 5,247 Bioretention 24,458 35,380 13,961 20,196 EDB 29,184 58,843 9,662 19,482 Wet pond 32,631 52,051 13,109 20,911 Wetland 32,770 52,273 13,713 21,874 Table 9. Example area-weighted BMP costs.

24 A Watershed Approach to Mitigating Stormwater Impacts Category Constraint Ve ge ta te d s w al es R oc k sw al es Fi lte r s tri ps D ry d et en tio n ba si ns R et en tio n ba si n s In fil tra tio n ba si n /tr en ch Co m po st a m en de d sl op e W et la nd sw al e/ ch an ne l/b as in O il/ w at er /g rit s ep ar at or v au lt B io re te nt io n / ra in ga rd en Physical Limited available ROW Obstructions (existing structures) Utility conflicts On or near top of steep slopes Poor soils/rocks High groundwater High hydraulic loads High pollutant loads Lack of vertical relief Deficient storm drain infrastructure Limited access Drainage patterns (adverse slopes) H H M H H H H H M M L H H H M M L L L L M M L H H M M L H H H M M M L H H M M H H H L L H L M H H H M H H L M L L L M H L L L H H H H L L L L L H M M L H H H M M M M H H M M H H L H L L M L M L M H L L L L L L M M L H M M M H H H H L L M L Social High traffic volumes High public acceptance requirements Mosquito/vector concerns Odor concerns Poor perception of BMPs Safety concerns H L L L L L L M L L M L H M M L H L H M M L H M H H M L H H L L L L L L H M M L H L H M M L H M H L M M L L H M M L H L Economic High capital cost High maintenance cost High replacement cost High land cost L L L L L L L M H M M L H M M L H H M L M M H L H M M L H M M L L M L L H M M L Note: L – low, M – medium, H – high. L indicates low susceptibility for the constraint, which is better. Table 10. BMP practicability/constraint susceptibility risk screening matrix. through bond estimates and local contract line item negotiations. The cost data should be derived from RS Mean construction cost data or other nationally recognized sources to ensure consistency. Costs per unit volume or load removed can then be computed by dividing by the estimated load reduction results from Step 3. The NCHRP BMP Evaluation Tools (Taylor et al. 2014) can also be used to provide these estimates. Step 5: Identify Factors/Constraints That May Influence On-Site versus Off-Site Implementation A determination must be made as to whether it is inappropriate to construct one or more BMPs at the project location because of the constraints on-site or adjacent to the project. These constraints must be considered as a part of a stormwater mitigation and management effort. Table 10 is a BMP practicability screening risk matrix that can be used to assist in the selection of BMPs for a particular site. The table provides a concise list of the most common constraints and rates the sensitivity of each of the primary BMPs to each constraint. The ratings are subjective and vary by site, but are useful as a starting point of a site-specific analysis. Factors that contribute to a high susceptibility rating include how frequently a constraint is likely to pose a significant barrier to the implementation of a particular BMP and how easily that barrier can be overcome.

Methods to Develop Mitigation Options 25 It is recognized that there are additional local factors that may be applied and one factor may outweigh all of the others, such as ROW acquisition or the presence of other protected resources in the potential BMP location. Each DOT must customize this part of the toolbox. Once potential constraints, consideration of where the most benefit can be attained, and potential support for on- or off-site opportunities have been identified, a decision can be made whether one, the other, or a combination of these is the most cost-efficient approach to storm- water management. Off-Site BMP Assessment Process The goal of this process is to determine the effectiveness of potential off-site treatment loca- tions when compared to an on-site location. This process includes comparisons of amount of load reductions as well as suggested cost comparisons for the equivalent load reductions. Figure 2 summarizes the recommended steps of an off-site BMP assessment process needed to estimate the size of the BMP. Step 1: Identify Off-Site BMP Location and BMP Type The first step is to select one or more candidate off-site locations. In many circumstances, the candidate location for management controls will be determined previously based on the identi- fication of a strategically located parcel of land near an existing storm drain outfall or trunk line that is downstream of urban development. These locations may be identified through spatial analysis or by an existing watershed improvement plan or stormwater master plan and are gen- erally within the same watershed as the project. The scale at which the watershed is defined (e.g., HUC-4 versus HUC-12) may vary depending on the regulatory agency and receiving water con- ditions. If the watershed is located in an area with an existing water quality trading program, the trading ratio used to estimate the credits for an off-site BMP may be influenced by the distance of Cost Benefit B M P Locatio n and Selection B M P Perform anc e B M P D rainage Characteristics and Input D eterm inatio n STEP 1 Identify Off-Site BMP Location and BMP Type STEP 2 Summarize Drainage Area Land Use Characteristics STEP 3 Compute Average Annual Loads to Off-Site BMP STEP 4 Estimate Preliminary Size of Off-Site BMP Based on Load Reduction Target STEP 5 Estimate Load Reduction Cost Ratio Figure 2. Off-site BMP assessment process.

26 A Watershed Approach to Mitigating Stormwater Impacts the credit-generating BMP from the project site. For example, the Connecticut Nitrogen Credit Exchange (CNCE) uses varying trading ratios for Publicly Owned Treatment Works based on vicinity to the Long Island Sound (CT DEP 2010). Watershed scale and trading ratios are important aspects of BMP crediting that should be considered when selecting an off-site BMP location. The purpose of this step is to make the direct comparison between on-site and off-site BMPs. The types of BMPs vary. It may be a regional, volumetric BMP, such as an extended detention basin, wet pond, or wetland basin that can man- age runoff from a larger drainage area and treat the constituents of concern. Larger, centralized locations are often easier to acquire and construct. The selection criteria, as determined by the regulatory agency, may be to use a similar type of BMP that would be used in the on-site location. This and other selection criteria, such as land use, political issues, or other BMP benefits such as habitat enhancement, should be addressed at the local level. Step 2: Summarize Drainage Area Land Use Characteristics A spatial analysis is required to summarize the land use characteristics for the tributary drain- age area to the proposed off-site BMP. The spatial analysis begins with obtaining land use and impervious cover for the drainage area by overlaying impervious cover with the land use areas and tabulating the total area of each land use along with estimated average imperviousness. There are several geospatial sources of information and databases which assist in land use and watershed summaries. These include the NLCD and the USEPA EnviroAtlas Database, National Hydrography Data Set, Land Cover/Land Use, and local or regional datasets such as the Regional Land Information System in Portland (http://www.oregonmetro.gov/rlis-live). Step 3: Compute Average Annual Loads to Off-Site BMP Determining the average annual load to off-site BMPs follows the same process used for on- site BMPs. The same process as described in Step 3 of the on-site BMP Assessment Process can be applied. The average pollutant concentrations in runoff from each of the land uses may be estimated using data summarized from the HRDB (Granato and Cazenas 2009) and the NSQD (Pitt 2008). Models or other tools such as the Simple Method (Schueler 1994) that DOTs and regulatory agencies are already familiar with can also be used. Step 4: Estimate Preliminary Size of Off-Site BMP Based on Load Reduction Target The effluent volume, V _ Ef, from Step 5 of the On-Site BMP Assessment Process is used to deter- mine the volume of treatment that is required. The effectiveness of the BMP can be expressed in these volumes in terms of volumetric percent capture (%Cap) and percent influent volume loss (%VL). The volume reduction can be expressed as: [ ]( ) ( )∆ = − −%Cap L 1 % (4)L VL V CW W Ef where DL is load reduction, %Cap is the volumetric percent capture, L _ W is the average annual load from the tributary watershed, %VL is the percent volume loss within the BMP, V _ W is the average runoff volume from the tributary watershed, and C _ Ef is the average effluent concentra- tion from the BMP. Solving for percent capture is as follows: [ ]( ) ( )= − − ∆ %Cap L 1 % (5) VL V C L W W Ef Given a target load reduction, estimated loads and volumes from the drainage area, an esti- mated percent volume reduction (e.g., 20 percent), and characteristic effluent concentration from

Methods to Develop Mitigation Options 27 a BMP, the needed percent capture can be computed. This percent capture can be used to esti- mate the design volume through the use of a sizing nomograph, such as the example shown in Figure 3, for an extended dry detention basin with a 72-hour drain time (figure shows various precipitation patterns represented by differing U.S. locations). For example, if the calculated per- cent capture needed was found to be 60 percent, then the extended detention basin would need to be sized for approximately a 0.43 inch storm event over the tributary drainage area for Block Island, RI or Chicago, IL. The formulas used in this step are easy-to-use and can be included in a toolbox for simple application. Nomographs can be pre-processed using models such as stormwater management model (SWMM) through batch processing varying many input parameters. The nomographs provide an easy process for a tool to use similar to look-up tables as a part of design. Step 5: Estimate Load Reduction Cost Ratio Applying Step 4, it is possible to use this load reduction to calculate unit costs to assess the cost-benefit (based on load reduction, percent capture, or similar metric) of multiple projects. Simply dividing annual load reductions by area-weighted (per acre) costs can provide an easy- to-use cost-benefit ratio, which can be compared to the cost ratio estimated for on-site BMPs. See load reduction and cost information for Steps 3 and 4, respectively, of the On-Site BMP Assessment Process. Integrating On-Site and Off-Site BMP Assessment Processes into Watershed Framework There are many factors that are included in a selection matrix or scheme that are outside the realm of simply determining the BMP effectiveness at one location or another. The analysis, Source: Oregon State University et al. 2006 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 Unit Basin Size (in) Pe rc en t A nn ua l R un of f C ap tu re d 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Memphis, TN Lisbon, Fl Chicago, IL Block Island, RI Alturas, CA Figure 3. Example percent capture curve sizing nomograph.

28 A Watershed Approach to Mitigating Stormwater Impacts selection, cost, life-cycle, and sizing of BMPs, whether on-site or off-site, can be accomplished with a series of recognizable and readily understood procedures and formulas that can be adapted for use by any DOT. The issues of permit processing times and ROW acquisition for BMPs are more complex as are the considerations of institutional issues and stewardship. The comparison of the effectiveness to out-of-kind mitigation, provision of ecosystem services, and banking or offsets is also more complex due to technical and scientific considerations as well as political and watershed management influences. These issues should be considered on a project-by-project basis as they relate to each DOT. Chapter 6 provides a brief discussion on how DOTs might use the toolbox for on-site and off-site BMP assessment processes. Implementation Costs for Market Options Each of the market options such as banking and in lieu fee programs provide potential utility and value to highway agencies and each can help accelerate restoration efforts. As DOTs decide between on-site and off-site mitigation options, it is important to understand how restoration costs impact the effectiveness of each and to develop decision tools to help guide implementa- tion and financing activities. The first step in this direction is assessing the specific components of stormwater BMP costs. Each of the market-based mitigation options function in distinctive ways, but the common element among each option is that there are costs associated with mitigation activities. Under- standing how and to whom these costs accrue is essential for designing an effective mitigation financing system. Anticipated aggregate costs are important to understand but the most important cost issues associated with transportation stormwater banking and in lieu fee programs is the relative cost differences between on-site and off-site mitigation projects. Stormwater BMP costs are inher- ently variable. Cost estimates depend on location, time, scale, level of urbanization, and a host of other variables. These cost differences can occur essentially in two ways: • Out-of-kind differences. A mitigation site can be appropriate when out-of-kind mitigation options are more beneficial to the watershed and less expensive to construct and maintain than an on-site BMP based on various factors such as land costs, site constraints, access, and economies of scale. For example, one advantage of a stormwater bank is that multiple devel- opment projects which would otherwise result in multiple relatively small-scale restoration projects can be mitigated through the use of a large-scale banking project. These larger proj- ects are often less expensive to construct and maintain. • In-kind differences. Differences in site conditions between the development site and the bank- ing project can result in cost differences related to the same BMP. This is especially the case in urban settings where site preparation and demolition can be a significant cost contributor. Land values can differ significantly throughout an urban area and transportation envelope, even within the same watershed. Cost Categories The total cost of stormwater BMPs has been separated into the following categories: land costs, pre-construction, construction, capital costs, operation and maintenance, and program administration. These cost elements encompass the majority of costs associated with stormwater BMPs. Together these can be used to estimate life-cycle costs. Land. Managing stormwater in urban areas is complex and potentially expensive for a vari- ety of reasons, not the least of which is the cost and limited availability of land. In fact, the cost of land is often the most significant variable impacting capital stormwater BMP costs (USEPA

Methods to Develop Mitigation Options 29 1999a). Clearly, land costs can vary widely among communities (King and Hagen 2011) as well as within communities. As a result, land costs can significantly influence the potential impact of market tools such as stormwater banks and in lieu fees. In general, land valuation is based on an estimate of the highest and best use of the land, i.e., the use of the land that is reasonably probable, legally permitted, physically possible, economi- cally feasible, and results in the highest value for a property. The estimated market or appraised value of land can vary, significantly at times, from the value-in-use and the investment value of land. The investment value of land is the value of land to the owner or prospective owner for investment or operational objectives, and the value-in-use is the value to one particular user of the net present value of the cash flows that the land is expected to generate for a particular activity under a specific use. These differences between investment value, value-in-use, and market value of land provide motivation for buyers and sellers to trade in the market place (Schram 2006; International Valuation Standards Council 2011). While the relative value of land is often a key motivation for pursuing a market-based imple- mentation option such as stormwater banks or in lieu fees, there are other factors or components that influence land costs. These components of land costs include: • Easement costs. Those projects that are installed on private lands without fee simple pur- chase will require a property easement to ensure adequate operations and maintenance (O&M) over the life of the practice. This results in two corresponding cost issues. First, ease- ment property must always be restored to as-good or better condition after O&M activities. Second, an easement essentially results in loss of use or loss of development rights to the property owner. • Opportunity costs. An opportunity cost is the cost of an alternative that must be forgone in order to pursue a certain action. As it pertains to the value of land, the opportunity cost of land is the cost to the owner of giving up the utility generating uses of the property when the land is taken out of service. In a stormwater setting, opportunity costs are associated with the devaluing of land when it is taken out of service and is repurposed for stormwater mitigation with regards to previous or potential land use. The derivation of opportunity costs involves making an assumption that a property owner faces increasing opportunity costs for land that is taken out of service for other uses (Thurston 2006). The opportunity cost and associ- ated value of land is often not considered in many BMP cost assessments, and as a result, BMP cost estimates are often significantly undervalued. Though opportunity costs are very real to private landowners, they often have little impact on public land transactions. For example, a King and Hagan (2011) report correctly incorporates the value of develop- able land—either public or private—into BMP cost estimates. However, developable public land only becomes an accounting or realized cost if the forgone activity would have actually occurred and would have resulted in some sort of revenue or cash flow to the community. Many publicly financed BMPs are installed on lands that are technically developable but are not slated for development in the foreseeable future, if ever. Therefore, there is no revenue cost to the community. • Land acquisition and transaction costs. Acquisition costs are the most direct type of land costs. They are site-specific and depend on the type of BMP being installed. Components of the cost to acquire land include time to identify land, legal fees, commissions and broker- age fees, title search fees, appraisal fees, governmental fees, and settlement fees. One of the unique aspects of Baltimore, MD’s proposed banking and in lieu fee program, for example, is that it will include the conversion of vacant and abandoned properties to stormwater BMPs (Cappiella et al. 2014). This will presumably require condemning and/or purchasing proper- ties and/or easements for conversion to stormwater BMPs, which will result in potentially significant costs, including legal and administration costs.

30 A Watershed Approach to Mitigating Stormwater Impacts Pre-construction Costs. Before construction can begin, remediation sites have to be pre- pared. As a result, pre-construction costs are incurred before BMP installation can begin. These costs include surveying, design work, permitting, geotechnical testing, and transaction costs. Site conditions significantly influence pre-construction costs associated with urban BMPs. Mitigation projects in urban environments often require significant site preparation, including demolition activity, which must be considered in the overall project cost. Sediment and erosion control activities costs must also be recognized. Pre-construction costs average between 10 and 40 percent of overall construction costs (King and Hagan 2011). Construction. In addition to the costs of land, the primary cost of any BMP is the actual construction and installation. Construction costs consist of the cost of excavation, primary ero- sion and sediment control, control structure installation, appurtenance costs, landscaping, and BMP-specific installation costs. Expenditures for professional and technical services required for the construction of the stormwater BMP are also included in construction costs. Construction costs are dependent upon the type of BMP being installed, and can vary widely (King and Hagan 2011). As with pre-construction costs, site conditions have a significant impact on the variability of construction costs. For example, hydrology, soil type, and topography can result in significant variations in construction costs from site to site. The more restrictive the site conditions, the greater the possibility that less costly on-site BMPs can be utilized. This may impact banking and in lieu fee programs by requiring more off-site mitigation to meet regulatory requirements. Cost of Capital. Cost of capital must be considered for any stormwater management project. Cost of capital is defined as the opportunity cost of the funds employed as the result of an invest- ment decision; it is equivalent to the rate of return that a business or institution could earn if it chose another investment with equivalent risk. Included in the cost of capital calculation is the cost of debt. King and Hagan used a uniform rate of three percent over a 20 year borrow- ing period. It should be noted that the cost of capital can vary from site to site or institution to institution, depending on the party securing the credit and also depending on risk differences. Operations and Maintenance. O&M costs are post-construction activities that provide upkeep for stormwater BMPs. Re-occurring annual costs include site inspection during and after construction, labor, materials, energy, landscape maintenance equipment, structural main- tenance, dredging, disposal of sediments, and litter removal (Taylor et al. 2014). Additionally, determining O&M costs requires an estimate of the useful life of the BMP to be made as well as the estimation of a discount factor to be used in the derivation of an annualized BMP O&M cost. The level of O&M required will depend on the complexity of the BMP. Erickson et al. (2010) performed a survey of stormwater BMP maintenance practices and found that constructed wetlands and porous pavements required more informed maintenance than other BMPs because of the level of complexity of the technology. Typically, O&M costs are estimated as a percentage of base construction costs, ranging from less than one to 20 percent depending on the BMP and the level of maintenance adopted (USEPA 1999b). Over time, operations and maintenance costs can exceed the level of initial construction costs. Program Administration. Program administration entails the process or activity of run- ning a business or enterprise. The establishment of stormwater banks and in lieu fee programs presents unique administrative challenges for stormwater management programs. Administra- tive costs are primarily based on labor requirements. Out-of-Kind Watershed-Based Approaches Out-of-kind approaches mitigate for project stormwater impacts by reducing environmental impacts elsewhere in the watershed via implementation of alternative, non-standard storm- water management techniques. The key element of the systems which can be adapted and used

Methods to Develop Mitigation Options 31 in out-of-kind watershed-based approaches all center on the use and application of mitigation and restoration approaches that have been developed for other environmental impacts. Ecosystem Services The concept of ecosystem services has been around for many years, but the use of eco system services in designing environmental mitigation approaches is still evolving. The concept is deceptively simple. Nature has value and understanding that value is important in making deci- sions about whether to restore or preserve nature. The definition of ecosystem services is fairly straightforward. It refers to those ecosystem processes that benefit human society, such as the formation and maintenance of fertile soil, fresh water, flood-control functions, and mainte- nance of climatic conditions. Less understood are the complex relationships linking ecosystem biodiversity with the various processes that benefit society (PCAST 2011). Agencies at both the state and federal level are currently not required to manage or regulate resources in a holistic, ecosystem-based manner. Doing so generally requires additional effort and risk. There is also increasing recognition that managing resources within individual juris- dictional silos is ineffective, leads to unintended consequences, and has failed to provide the level of natural resource protection needed. In addition, more outcome-oriented agencies recognize that there is a difference between actively regulating outside of jurisdictional boundaries and allowing creative ecosystem-based solutions. The need for (or at least willingness to) move toward more effective, outcome-oriented solu- tions is not uniformly accepted across the state agencies that implement stormwater regulations. The trend is toward a focus on ecosystems, ecosystem services, and other holistic approaches. This trend is epitomized by a report produced by the President’s Council of Advisors on Science and Technology (PCAST) Report that was released in 2011. The report spoke to the need for federal agencies to figure out how to incorporate ecosystem services into their decision-making processes. The report made a number of specific recommendations for agencies with resource regulation or management responsibilities, including the following: Federal agencies with responsibilities relating to ecosystems and their services (e.g., USEPA, NOAA, DOI, USDA) should be tasked with improving their capabilities to develop valuations for the ecosys- tem services affected by their decision-making and factoring the results into analyses that inform their major planning and management decisions. (President’s Council of Advisors on Science and Technol- ogy, Report to the President—Sustaining Environmental Capital: Protecting Society and the Economy, July 2011) Federal agencies are actively working on how to respond to the PCAST Report, and the National Ecosystem Services Partnership released a publication on the best practices for integrating eco- system services into federal decision-making (Olander et al. 2015). Applying an ecosystem ser- vices framework can ensure that people adequately understand the value of nature so that it will be properly included in management decisions. In the context of justifying off-site mitigation, this can translate into providing an important basis for stating that an off-site location is the most appropriate. For a comparison to be developed between the provisions of ecosystem services and BMPs for mitigation, an ecosystem service approach must be linked to the current stormwater manage- ment approaches addressed above. A common currency or recognition of the direct benefits for each activity must be established. A simplified example of this comparison is shown in Table 11. It is an example of how this approach may be applied for a few simple pollutant mitigation issues. The table lists typical pollutants of concern that are associated with highway runoff and the potential effect on key ecosystem services. A check indicates that the pollutant of concern can adversely affect the ecosystem service. For example, untreated stormwater discharges have the potential to reduce populations of harvestable fish (provisioning services), reduce aquatic primary productivity (regulating services), and change the patterns of soil formation through changing

32 A Watershed Approach to Mitigating Stormwater Impacts patterns of aggradation and sedimentation (supporting services). Projects in the same watershed that enhance these same services, such as aquatic habitat restoration that increases spawning and rearing habitat for harvestable fish populations, can be connected to project level effects such as out-of-kind mitigation. The challenge for implementing this approach is the development of the “currency” or “equivalency” within a crediting program or framework that can be used to provide reasonable assurance that the out-of-kind mitigation of stormwater will result in the protection, restoration, or enhancement of the service. Ecosystem Services Tools and Approaches Some resources that can help DOTs identify and track the ecosystem services–based tools and approaches include the following: • The World Business Council for Sustainable Development has published a guide for organi- zations looking for ecosystem services–based decision support tools (Eco4Biz—Ecosystem services and biodiversity tools to support business decision-making, Version 1, April 2013). This report can be found at www.wbcsd.org/eco4biz2013.aspx. • Business for Social Responsibility (BSR) publishes an annual summary of the ecosystem services–based decision support tools that are being used and tested by member corpo- rations. The report can be found at www.bsr.org/en/our-insights/report-view/measuring- managing-corporate-performance-in-an-era-of-expanded-disclosure. Typical Pollutants of Concern Ecosystem Service G ro ss s ol id s, s ed im en t a nd flo at ab le s Pe st ic id es a nd h er bi ci de s O rg an ic m at er ia ls /o xy ge n de m an di ng s ub st an ce s M et al s O il an d gr ea se /o rg an ic s as so ci at ed w ith pe tro le u m To ta l d is so lv ed s ol id s (S alt s) Example Link Provisioning Services Provide food Increased toxics can reduce harvestable fish populations Provide fresh water Toxics can reduce water potability Regulating Services Carbon sequestration Increased toxics, and lower DO levels can reduce aquatic primary productivity Prevent disease transmission Increased oxygen demanding substances can lead to increased disease vector production, and increases in disease organisms in aquatic systems Pollutant reduction and detoxification Toxics can lead to higher pollutant levels and reduced algae and vegetation that can contribute to detoxification Supporting Services Nutrient cycling Increased pollutants can reduce primary productivity and, when coupled with higher nutrient loads, lead to greater amounts of available nutrients in aquatic systems by enabling soil development to proceed at a faster pace Soil formation Increased sediment loads can increase aggradation and sedimentation rates in aquatic systems by enabling soil development to proceed at a faster pace Primary productivity Increased toxics can reduce aquatic primary productivity Biodiversity Increased toxics can reduce algal, macroinvertebrate, and fish populationbiodiversity Waste decomposition Toxics can lead to higher pollutant levels and reduced bacteria, algae, vegetation, and invertebrates that can contribute to waste detoxification Table 11. Typical pollutants of concern for highway runoff that can adversely affect ecosystem services.

Methods to Develop Mitigation Options 33 • The Ecosystem Based Management (EBM) Tool Network is an organization coordinated by NatureServe that provides information on ecosystem measurement tools. It is oriented toward coastal and marine focused tools, but the organization tracks many ecosystem services–based tools that have a broader application. The EBM network can be accessed at www.ebmtools.org/. Site Level Some of the most prominent and accepted tools currently available for evaluating ecosystem services that could be adapted for use in out-of-kind stormwater mitigation programs are dis- cussed in the following paragraphs. Currently, the primary site level tool available is EcoMetrix. This tool has been widely tested for measuring site-level changes in a comprehensive suite of ecosystem services. In addition to EcoMetrix, an Ecosystem Services Identification & Inventory (ESII) tool was developed by Dow Chemical and The Nature Conservancy (TNC). Officially released in February of 2016, the tool is freely available to the public and provides utility for a variety of skill levels. • EcoMetrix was developed in Oregon, but has been applied and tested throughout the United States, as well as being tested internationally in the Caucasus region (a region at the border of Europe and Asia). EcoMetrix can provide spatial or tabular outputs that show the production of ecosystem services at a site, as well as the change in production based on proposed scenarios. EcoMetrix relies on collected attribute data as the primary input into the system. The tool is currently maintained by the EcoMetrix Solutions Group (http://ecometrixsolutions.com/). For reviews and more information on EcoMetrix and other tools: – Business for Social Responsibility, Measuring and Managing Corporate Performance in an Era of Expanded Disclosure—A Review of the Emerging Domain of Ecosystem Services Tools, January 2013 (www.bsr.org/reports/BSR_Ecosystem_Services_Tools.pdf). – President’s Council of Advisors on Science and Technology, Report to the President— Sustaining Environmental Capital: Protecting Society and the Economy, July 2011 (https://www.whitehouse.gov/sites/default/files/microsites/ostp/pcast_sustaining_ environmental_capital_report.pdf). – United States Geological Survey, Ecosystem Services Valuation to Support Decisionmaking on Public Lands—A Case Study of the San Pedro River Watershed, Arizona, Scientific Investi- gations Report 2012-5251 (http://pubs.usgs.gov/sir/2012/5251/sir2012-5251.pdf). • ESII is the result of a three-year collaboration between Dow Chemical and TNC. The tool includes eight ecosystem services, including water quality and water quantity, and provides outputs in percent performance, functional acres, service acres, and engineering units (e.g., gallons/min, mg/l, etc.). ESII utilizes field collected data but includes a data collection appli- cation to make the data collection user friendly. The ESII Tool is available for free and can be accessed at www.esiitool.com. Watershed Scale There are a number of ecosystem services–based tools that measure services at a watershed scale, such as EnviroAtlas, Artificial Intelligence for Ecosystem Services (ARIES), and Integrated Valuation of Ecosystem Services and Trade-offs (InVEST). • USEPA’s EnviroAtlas is a web-based, open-access tool for exploring the benefits people receive from nature or ecosystem goods and services. Reporting units include watershed basins across the United States and Census block groups within specific urban areas (Pickard et al. 2015). EnviroAtlas evaluates the following ecosystem services: clean air; clean and plentiful water; natural hazard mitigation; biodiversity conservation; food, fuel, and materials; recreational opportunities; and cultural and aesthetic value. EnviroAtlas contains information on the

34 A Watershed Approach to Mitigating Stormwater Impacts status of these benefits, the ecosystems that provide and protect them, and related health and economic impacts. The EnviroAtlas is available at www.epa.gov/enviroatlas. • ARIES is a tool that was originally developed at the University of Vermont in partnership with Conservation International. ARIES focuses on measuring ecosystem services production and movement across the landscape. It uses available GIS data as the primary data input into the system. ARIES has been tested in various parts of the United States and has also been used in international applications. • InVEST is a tool that was developed at the Natural Capital Institute at Stanford in partnership with TNC. InVEST provides spatial mapping of ecosystem services production, as well as net present value information about those services. InVEST is designed to be user friendly and is available through a free download. InVEST relies on available GIS data as the primary input into the system, although it is recommended that InVEST be integrated into decision-making processes accompanied by stakeholder engagement. Methodology for Evaluating Out-of-Kind A few DOTs have begun to establish protocols and criteria for ranking, prioritizing, and selecting areas for out-of-kind mitigation construction. The protocols often begin with outreach efforts within transportation agencies, with other local, state, and federal agencies, and also using stakeholder involvement in the identification and selection process. The example programs also include other strategies and techniques to reduce stormwater impacts of non-DOT land uses and properties within the watersheds. Examples of these additional techniques and strategies include acceptance of land conservation, improved stream and land buffers, development of nutrient management programs, and other non-structural approaches in which regulatory agencies recognize the watershed benefit to these out-of-kind mitigation projects. The follow- ing are some key elements from off-site mitigation programs utilized in Maryland and Delaware: • Maryland State Highway Administration (MDSHA). MDSHA’s stormwater management pro- gram considers stormwater mitigation at the watershed scale. The agency recognizes that permit requirements and stakeholders view stormwater management from the project or site perspec- tive. MDSHA has initiated a significant investment in outreach programs to educate regulators and stakeholders about the value of partnerships and the consideration of watershed-based mitigation. Some of the key elements in place or under consideration include: – Incorporating local level priorities for restoration/mitigation. MDSHA is establishing partnerships with local counties to help meet local and MDOT NPDES requirements. This includes establishment of off-site mitigation projects that meet both local and MDSHA requirements. – Department of Natural Resources (DNR) partnership for reforestation. MDSHA has been examining and undertaking reforestation as an alternative mechanism for pollution con- trol and runoff reduction through land cover change. Approximately 800 acres have been identified as having potential for the establishment of forest areas within the MDSHA ROW. The MDSHA provides funding to buy the trees, the DNR provides the land, and the Department of Corrections provides the labor for planting under this agreement. This will also contribute to the Governor’s commitment to plant a million trees over the next five years and is independent of the state’s TMDL effort. – Agricultural Controls. MDSHA is investigating opportunities beyond the ROW and beyond traditional stormwater management structural controls in order to provide the state with opportunities to trade nutrients or reduce nutrient loads under the TMDL program. For example, MDSHA may be able to provide or pay for shelters for poultry to reduce nutri- ent runoff. MDSHA has also initiated discussion with the Department of Agriculture on providing buffer planting along stream banks on farmland.

Methods to Develop Mitigation Options 35 • Delaware DOT. DelDOT participates in stakeholders groups called Tributary Action Teams that have been convened by the state Department of Natural Resources and Environmental Control (DNREC) to help recommend strategies and actions to reduce nonpoint source pol- lution in several watersheds that are under TMDL load reduction requirements. – Implementation of Water Quality Improvement Plans. Implementation of Water Quality Improvement Plans on a watershed basis is a required element of DelDOT’s proposed new permit. DelDOT, in coordination with seven co-permittees, must plan and implement proj- ects that aim toward meeting TMDL allocations and applicable water quality standards in two priority watersheds during the five year permit term. – Other Watershed Plans. DelDOT has also partnered with various agencies and organiza- tions in the state on two smaller subwatershed assessment and improvement projects. Other watershed plans and strategies developed by DNREC include a Watershed Implementation Plan for the Chesapeake Bay TMDLs. DelDOT is participating in this effort by helping to write the stormwater section. These are available online (www.dnrec.delaware.gov/swc/wa/ Pages/WatershedManagementPlans.aspx).