Pollutant Load Reductions for Total Maximum Daily Loads for Highways (2013)

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.

43 Based on the information presented in the previous chapters, the following conclusions were drawn. They are grouped into four general topic areas: (1) best management practice (BMP) performance and cost, (2) effective total maximum daily load (TMDL) implementation strategies, (3) main challenges, and (4) further research. BEST MANAGEMENT PRACTICES PERFORMANCE AND COST Our BMP matrix/toolbox identified relative performance rank- ings (high, medium, low) based on percent removal efficiency for a wide variety of structural highway BMPs. Many of the BMPs had a high ranking for total suspended solids (TSS) indicating >65% removal; for example, infiltration basins, sand filters, bioretention, permeable friction course, and others (see Table 20). Nutrients, on the other hand, were more difficult to treat, especially total nitrogen for which none of the structural BMPs had a high performance rank- ing. Austin sand filters and wet basins appear to be the most promising for removing total nitrogen; both had a medium ranking (31%–65% removal). The apparent inability of most BMPs to treat nitrogen is most likely the result of the com- plex nature of the nitrogen cycle and the many site-specific factors that affect the transformation of nitrogen species, such as microbes, and the extent of aeration. Total phospho- rus results were highly variable depending on the BMP, rang- ing from negative (net export) to high. However, infiltration basins, infiltration trenches, and bioretention each had a high ranking for total phosphorus removal. Fecal coliform performance data were limited; however, several of the BMPs were identified as having a high ranking, including infiltration basins, infiltration trenches, Delaware sand filters, and wet basins. Finally, for metals, performance is difficult to evaluate as it tends to vary depending on the analyte and in many cases data were not available for total copper and total lead. However, the Delaware sand filter and the wet basin stood out as having a high performance ranking for total zinc, total copper, and total lead. Regarding costs, several sources of life-cycle cost data are presented in the matrix/toolbox for structural BMPs. How- ever, a true cost–benefit analysis was not possible owing to differences in cost estimating approaches and reporting units, variability in costs by region, and inconsistencies in BMP naming conventions in the source reports. The reader is encouraged to access the complete reports (provided as hyperlinks in Table 20) to obtain cost data that are most relevant for their state/region. More detailed quantitative performance data are also pre- sented from a large number of studies from the International Stormwater BMP Database. The data are reported as influent/ effluent concentrations with the 95% confidence interval and statistical significance. Based on these data, TSS and metals (total zinc, total copper, and total lead) appear to be relatively easy to treat with many types of BMPs (e.g., grass strips, wetland basins, etc.; see Table 22), whereas nutrients and fecal coliform are relatively difficult to remove. This is gen- erally consistent with the findings of the literature review. Media filters and retention ponds stood out as being effective at treating all of the TMDL pollutants of concern examined (TSS, nutrients, fecal coliform, and metals), where effective- ness is defined as having a statistically significant reduction in pollutant concentrations. As for nonstructural BMPs, quantitative performance data tend to be sparse and performance metrics vary widely and/or may not be transferable nationwide. Therefore, deriv- ing numerical load reductions for TMDL purposes continues to be a challenge. However, some studies were identified with performance data that are summarized in Table 23. Based on these findings, street sweeping and catch basin cleaning have the potential to be moderately effective at removing TSS, nutrients, and metals provided they are per- formed frequently (weekly to biweekly for sweeping, every three months for catch basin cleaning) to prevent build-up of pollutants; also, in the case of sweeping, the technology must be suitable to maximize pollutant removal. Other non- structural practices are not as well quantified for the range of TMDL pollutants examined in this report; however, tree planting and stream restoration have been documented as providing water quality benefits for nutrients. The Mary- land State Highway Administration has also developed a method to convert acres of tree planting and linear feet of stream restoration to equivalent acres of impervious area treated. Other practices such as anti-icing management are difficult to quantify. However, one very successful exam- ple was noted in New Hampshire where the department of transportation (DOT) has demonstrated a 20% reduction in chloride loads by upgrading the technology of their salt application fleet. chapter five CONCLUSIONS AND FURTHER RESEARCH

44 EFFECTIVE TOTAL MAXIMUM DAILY LOAD IMPLEMENTATION STRATEGIES Developing an effective TMDL strategy begins with aware- ness and education within the DOT on TMDL issues, which may be challenging in cases where the different DOT divisions (design, maintenance, environmental, etc.) are not integrated. Based on our findings, we found that awareness of TMDLs ranged from basically no awareness at all in states where the DOT was not named in any TMDLs, to full awareness and very active participation in implementing strategies in states where the DOT was named a stakeholder in a large number of TMDLs. Awareness and training will become especially important as TMDLs continue to emerge nationwide; in some states, hundreds more are expected to be implemented where the DOT may be named a contributor. Collaboration with other stakeholders and jurisdictions is key to developing an effective TMDL strategy. A prime example is Delaware, where the DOT shares a joint National Pollutant Discharge Elimination System MS4 permit with other stakeholders, which facilitates collaboration among the co-permittees to implement BMPs on a watershed-wide basis rather than just within the DOT right-of-way. Another exam- ple of collaboration is in New Hampshire where the DOT reduced the application of road salts by 20% to address chlo- ride TMDLs by upgrading their fleet of plows and applicators with the latest technologies for salt application. These tech- nologies were shared with private and municipal operators through a collaborative approach known as the Technology Transfer program through the University of New Hampshire. Another key element of an effective TMDL strategy is early and active participation in the TMDL development pro- cess. North Carolina is an ideal DOT in this regard. It has a strong working relationship with its regulatory agency and in most cases contributes data and scientific expertise to help define its own contribution to the waste load allocation (WLA) and ensure a realistic TMDL strategy. The Washington State DOT also actively participates in TMDL development and it works with the regulatory agency to write specific action items into their National Pollutant Discharge Elimination Sys- tem permits. Proper estimation of loads is also important to developing an effective TMDL strategy. Calculating baseline pollutant loads and predicting potential load reductions from various BMP implementation scenarios is critical to a success- ful TMDL program. This procedure has two main advantages. First, it provides a decision support system to assist highway managers in developing the most cost-effective TMDL strat- egy. Second, the data generated may be provided to the state regulatory agency during the TMDL development process and may help define the DOT’s contribution to the WLA, poten- tially resulting in a more targeted and effective TMDL strategy. In North Carolina, this process has helped inform the develop- ment of a unique TS4 permit, which represents recognition by the regulatory agency that permits need to address specific DOT concerns given the unique nature of their linear highway assets. Transportation Separate Storm Sewer System (TS4) permits may be a useful model for other states with traditional Municipal Separate Storm Sewer System (MS4) permits. Based on the state DOT interview responses, five of the 12 states (Colorado, Delaware, New Hampshire, New York, and Virginia) use some type of modeling tool(s) to estimate loads. The Simple Method was the most common model cited (used by New Hampshire, New York, Delaware, and Virginia); others included SELDM (Stochastic Empirical Loading and Dilution Model, used by Colorado and New York), WinSLAMM (Source Loading and Management Model for Windows, by New York), PLOAD (Pollutant Loading, by Delaware), and the Watershed Treatment Model (by Virginia). The California DOT (Caltrans) uses a water quality planning tool with embedded calculations to estimate loads; however, it is intended primarily for designers. North Carolina DOT uses the Simple Method to develop their nutri- ent management strategies, but do not use any models for TMDL purposes. However, the state regulatory agency in North Carolina (the North Carolina Department of Environ- ment and Natural Resources) uses several modeling tools such as HSPF (Hydrological Simulation Program FOR- TRAN), LSPC (Loading Simulation Program in C++), load duration curves, and others. MAIN CHALLENGES Although some states had relatively successful TMDL pro- grams, many states noted significant challenges to develop- ing an effective implementation strategy. A common theme across most DOTs was a lack of manpower and financial resources. In addition, several DOTs cited the lack of effective BMP technologies for linear highway applications, and some expressed difficulties in navigating the complex regulatory environments within their state. For example, several states with multiple regulatory agencies and other government entities noted a number of challenges including (1) incon- sistent enforcement and interpretation of TMDL require- ments among the state regulatory agencies that prevented the DOT from developing comprehensive TMDL strategies, (2) requirements placed on DOTs by regulatory agencies on which BMPs the DOT can use, and (3) communication chal- lenges between the DOT and state regulatory and federal agencies. FURTHER RESEARCH More research is needed on long-term adverse environmen- tal and cultural impacts related to BMP implementation. For example, in arid climates where BMP vegetation is difficult to grow, soil degradation (wind and erosion loss, decreased organic addition to soils), human health effects from dust, and adverse impacts of decreased grass surface on water infiltration and stormwater movement may be an issue. Some DOTs also noted that fire hazards were a concern for some

45 BMPs in arid regions; others cited mosquitoes and West Nile Virus in wet pool BMPs as potential threats to human health and safety. A few DOTs were also concerned about ground- water pollution impacts for infiltration BMPs. Many DOTs cited the need for new and innovative BMP technologies designed for linear highway applications, espe- cially in ultra-urban corridors. Although the list of BMP types is continually expanding, there have been relatively few studies on which BMPs are specifically effective for TMDL implementation for highways. Traditional DOT prac- tices are typically ineffective for TMDL pollutants such as nitrogen, bacteria, and pathogens. There need to be more studies conducted on BMP lon- gevity, life-cycle costs, and maintenance costs and standards. Some life-cycle cost data are available (see Life-Cycle Costs in chapter three and Table 20); however, in general the cost estimating approaches, reporting units, and BMP naming conventions are not consistent, which prevents adequate syn- thesis of the information and negates the ability to conduct a cost–benefit analysis. Greater standardization of mainte- nance practices would benefit DOTs by ensuring continued performance of BMPs. Entirely new TMDL strategies may be needed to address some of the less common pollutants of concern (e.g., biological integrity, sediment toxicity, organic compounds, or “surrogate” pollutants such as flow). To ensure permanent reduction of these pollutants from the right-of-way, more research is needed on alternative strategies (e.g., source control, institutional controls, and water quality trading). Finally, there needs to be more standardization of BMP nam- ing conventions in the literature. For example, what is called a “bioswale” in some states is called a “bioinfiltration swale” in others. Some states distinguish between Austin and Delaware sand filters (typically those in the western United States), while others lump them together as “sand filters” or “media filters” (which may or may not include sand). Similarly, bioretention practices are sometimes called “rain gardens” even though the design is essentially the same. Glossaries do exist with stan- dardized BMP types and descriptions [e.g., see MDE (n.d.)], but tend to be specific to that state or region.