The Role Street Sweeping Must Play in Achieving Numeric Pollutant Limits
Recently, there have been some significant changes in the regulation of pollutants found in urban stormwater. And it appears that more changes are still on the horizon. These changes have been categorized as either the establishment of numeric pollutant standards where narrative ones existed previously or numeric effluent pollutant limits in stormwater discharges.
In December 2009, EPA issued a final rule that established numeric limits in stormwater discharges from construction sites that was to be implemented over a four-year period. However, lawsuits from the National Association of Home Builders and others have forced EPA into rewriting the new rules, which are now expected in 2012.
As the result of a 2008 lawsuit from a coalition of environmental groups, EPA finalized numeric nutrient criteria (NNC) for Florida’s flowing waters and lakes in November 2010. Additional NNCs are expected in November 2011 for Florida’s estuaries and coastal waters. The NNCs that have been proposed will become the new water-quality standards for those parameters, essentially replacing the narrative standards that were previously used. The Conservancy of Southwest Florida noted in its NNC fact sheet that as of 2008, seven other states have already adopted NNCs for estuarine water bodies, 13 have adopted some NNCs for lakes, and nine have adopted some NNCs for rivers.
In November 2010, EPA’s Office of Water issued a memorandum entitled “Revisions to the November 22, 2002, Memorandum Establishing Total Maximum Daily Loads (TMDLs) Waste Load Allocations (WLAs) for Stormwater Sources and NPDES Permit Requirements Based on those WLAs”. The 2010 memorandum addresses the use of water-quality-based effluent limits in stormwater permits, including the use of surrogates such as flow or impervious cover. Many stakeholders including the American Public Works Association (APWA) were concerned that the 2010 memorandum seemed to imply that the National Pollutant Discharge Elimination System (NPDES) permitting authorities should impose end-of-pipe limitations on each individual outfall in every municipal separate storm sewer system (MS4).
In response to the concerns raised, EPA issued further clarifications that the 2010 memorandum was not intended to result in the widespread use of numeric end-of-pipe limits in stormwater. In an effort to calm these types of fears, EPA further stated that the term “numeric effluent limitation” should be read in a broader context than just end-of-pipe limitations and could instead relate to limitations expressed as pollutant reduction levels for parameters that are applied systemwide rather than to individual discharge locations. Because the courts are likely to be the ultimate interpreter of the meaning of the 2010 memorandum, these assurances from EPA of how the written words should be interpreted are of little comfort to the MS4s and other NPDES permit holders. In fact, as a result of continued concerns raised by APWA and others, EPA published the 2010 memorandum in the Federal Register as a proposal and accepted written comments on it up to May 16, 2011. EPA anticipates making a decision by December 2011 as to whether to retain the memorandum as originally written without change, to reissue it with revisions, or to withdraw it altogether.
Clearly, all of these actions and others that will follow in the future will have a financial impact on municipal and industrial NPDES permit holders. But perhaps the greatest financial impact will ultimately result from a March 10, 2011, federal court ruling. This very recent opinion has broad implications for the capital and operational budgets of all MS4s. The Federal Court of Appeals for the Ninth Circuit has reaffirmed that the Clean Water Act (CWA) requires that stormwater discharge points (e.g., pipes releasing stormwater into rivers, wetlands, and lakes) must have an NPDES permit because such discharge points are “point sources.” More importantly, the court ruled that it does not matter who is responsible for the contaminants that the stormwater picks up; the CWA puts the onus of responsibility for remediating the stormwater before it is discharged on the entity operating the stormwater system, which in many cases is a MS4 (NRDC v. County of Los Angeles, 2011 U.S. App. LEXIS 4647 3/10/11).
The court went on to state that the pre-discharge remedial standards for stormwater will need to be laid out in the NPDES permit. These treatment requirements must be met before the stormwater can be released or discharged. The court also pointed out that states operating these types of programs under the CWA have the authority to impose more stringent requirements that those imposed by EPA.
The generally applicable CWA standard requires a reduction of the contaminants in the stormwater to the maximum extent practicable (MEP). Unfortunately, EPA has never defined the term. But for many of us common-sense folks that practice in the field of stormwater, the costs associated with implementing the required remediation to achieve compliance should be considered in defining MEP.
In fact, most of the outcry against the previously described EPA actions have centered on what the remediation is expected to cost. In a letter to US Senators Bill Nelson and Marco Rubio of Florida regarding that state’s final NNC rule, a coalition of more than 50 representatives from agriculture, water utilities, the business community, local governments, and labor stated “Studies produced by the Florida Department of Environmental Protection and the Florida Department of Agriculture and Consumer Services and two independent studies by Cardno ENTRIX and Carollo Engineers all show the impact of the EPA’s mandates to Florida’s economy will be in the billions, including household water utility bill increases of approximately $700 per year.”
Many communities such as Portland, OR, have already experienced sewer rate increases that have been greater, primarily for the remediation of combined sewer overflows (CSOs), which was also mandated by the CWA. So although these increases are troublesome, especially in our current economy, who is to say whether a $58 monthly increase (assuming it does occur) is not “practicable” when others in MS4s like Portland have experienced greater increases to achieve CWA objectives?
If the needed capital expenditures and operational costs are to be kept within reason, the treatment mandate that arises from these and future EPA actions and this recent Ninth Circuit opinion will require some creative thinking about the means and methods needed to reduce pollutants in stormwater. With that said, it must be emphasized that street sweeping and other public works–related cleaning programs like catch basin cleaning and snow removal are in fact some of the most cost-effective practices when it comes to removing pollutants that would otherwise have been transported by urban stormwater and discharged from the built environment. Cost effectiveness as described herein should be measured as the cost to remove a given mass of a specific pollutant from the stormwater of a given community. This unit cost is usually expressed in the states as dollars per pound of pollutant removed from stormwater.
In October 2007, I published an article in Stormwater Solutions entitled “Sweep Before You Treat” (Sutherland 2007). This brief article lays out the important facts that support the assertion that street sweeping practices should be the first line of defense against urban stormwater pollution. In September 2009, I published in the APWA Reporter an article entitled “Recent Street Sweeping Pilot Studies are Flawed” (Sutherland 2009). This article presents evidence as to why these pilot studies were flawed. It also documents that in a 2001 study of the potential pollutant reduction benefits associated with the implementation of effective street sweeping programs in both Livonia and Jackson, MI, concluded that a 60% annual reduction in total suspended solids (TSS) loads is very likely. These benefits could be obtained through switching to more efficient particulate pickup machinery, slowing down sweeping speeds, ensuring the sweeper gets to the curb through effective parking restrictions, and cleaning approximately every two weeks during the non-frozen sweeping season, or approximately 17 times per year on the average (Sutherland and Jelen 2002).
The unit cost of sweeping pollutant reductions, estimated between $1 to $2 per pound of TSS removed from stormwater in the 2001 Michigan study, are several orders of magnitude lower than the cost of using low-impact-development (LID) retrofit techniques, estimated by others to be from $50 to $500 per pound of TSS removed. However, for newly developed areas, the use of LID provides great value, because the cost of design and construction is essentially identical to that of traditional stormwater systems, and those costs are always borne by the developer and not the public. Unfortunately, many a stormwater consultant has recommended the LID retrofit option as the “selected” stormwater pollutant reduction alternative, even though the communities may be essentially 100% urbanized, while turning a blind eye to the lower-cost and much more cost-effective pollutant removal benefits of cleaning practices. And there is reason to believe that EPA’s new stormwater regulations will include some provision for mandatory annual LID retrofit requirements. If this does occur, it will be a huge mistake; LID retrofit should be mandatory only when economic conditions dictate that existing urbanized areas redevelop, because those costs are best borne by the developer who stands to profit from that significant monetary investment.
Those that have been critical of street sweeping as an effective stormwater pollutant reduction practice often lack a complete understanding of the complicated interactions of the processes that relate directly to the accumulation and transport by stormwater of the contaminated particulate material that is found on urban streets, often referred to as street dirt. Street dirt data are rarely collected, even though they remain a much better indicator of best management practice (BMP) effectiveness (Minton and Sutherland 2010). These critics often don’t understand how the effective removal of this accumulated material by cleaning practices can manage to deplete the available supply of these containments normally transported by runoff. Thus, the pollutant mass loads and concentrations being transported are essentially reduced. This is especially true for higher concentrations and loads associated with the larger storms. Many of these higher values would appear as outliers in statistical analyses of the stormwater quality data. The street sweeping pilot studies conducted by others (that the APWA Reporter article criticized) all concluded that street cleaning was an ineffective BMP. The measurement that these studies typically used was the percent charge in the observed average or medium concentration of various pollutants while the streets in the watershed were being swept compared to when they were not being swept at some specified frequency. Unfortunately, these pilot studies all relied heavily on the use of either very simple models or models whose washoff components did not include sediment transport equations. As a result, I believe they all have reached erroneous conclusions regarding street sweeping’s pollutant reduction effectiveness (Selbig and Bannerman 2007, Center for Watershed Protection 2008).
I have developed and verified a physically based explicit stormwater quality model called the Simplified Particulate Transport Model (SIMPTM) that includes sediment transport–based washoff algorithms (Sutherland and Jelen 1998). SIMPTM can accurately simulate these accumulation and washoff behaviors, including the complex interaction of removal by cleaning practices (Sutherland and Jelen 1996a, 1997, 2003, 2011).
One important interaction included in SIMPTM is called wet-weather washon. Wet-weather washon is the transport of both particulate and associated pollutant loadings including soluble forms of pollutants from adjacent paved and unpaved areas to the street and parking lot surfaces where they can be deposited and later transported in subsequent runoff events (Sutherland and Jelen 1996b). Critics that have stated that SIMPTM does not account for any “off-street” loadings sources are obviously unaware of how the model actually operates (Geosyntec Consultants 2008).
Another plus for street sweeping, since it picks up polluted material prior to its dilution into the runoff stream, is that it can remove soluble pollutants such as metals and nutrients that generally thwart all types of structural BMPs except media filters. In 1998, as part of a study of stormwater pollution from a container handling yard for the Port of Seattle, it was concluded that weekly cleaning with a high-efficiency sweeper would likely reduce annual total copper and zinc washoff by 25 to 55% and annual soluble copper and zinc by 12 to 27% (Sutherland, Minton, and Jelen 1998). Wet vaults, which were required for stormwater treatment by the regulating agency at the time, were estimated to have a lifecycle cost nine times greater than sweeping and were not expected to provide any reduction in soluble metals.
Because of sweeping’s demonstrated lower cost per pound of pollutant removal, MS4s under NPDES Phase I and Phase II mandates should now start the development of optimal sweeping programs designed to reduce solids, metals, nutrients, hydrocarbons, and other pollutants to the maximum extent practicable. Only by using appropriate physically based calibrated models and comparing the pollutant reduction benefits of optimal sweeping to both a range of end-of-pipe solutions and LID-based retrofit practices can the most cost-effective mix of sweeping and/or other BMPs be attained. This exercise, if done properly, will minimize the overall cost of meeting stormwater pollutant reduction goals throughout the communities’ built environments. Well-informed NPDES managers will then be able to determine what meaningful role cleaning programs can play in achieving numeric pollutant limits in stormwater. Numeric limits have arrived, and more are just around the corner.
Author's Bio: Roger C. Sutherland, P.E., is a senior water resources engineer with AMEC Earth & Environmental in Tigard, OR.
HTMLOutput: Object reference not set to an instance of an object.
HTMLOutput: Object reference not set to an instance of an object.HTMLOutput: Object reference not set to an instance of an object.