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.
1 Introduction URBANIZATION AND ITS IMPACTS The influence of humans on the physical and biological systems of the Earthâs surface is not a recent manifestation of modern societies; instead, it is ubiquitous throughout our history. As human populations have grown, so has their footprint, such that between 30 and 50 percent of the Earthâs surface has now been transformed (Vitousek et al., 1997). Most of this land area is not cov- ered with pavement; indeed, less than 10 percent of this transformed surface is truly âurbanâ (GrÃ¼bler, 1994). However, urbanization causes extensive changes to the land surface beyond its immediate borders, particularly in ostensibly rural regions, through alterations by agriculture and forestry that support the urban population (Lambin et al., 2001). Within the immediate boundaries of cities and suburbs, the changes to natural conditions and processes wrought by urbaniza- tion are among the most radical of any human activity. In the United States, population is growing at an annual rate of 0.9 percent (U.S. Census Bureau, http://www.census.gov/compendia/statab/2007edition.html); the majority of the population of the United States now lives in suburban and urban areas (Figure 1-1). Because the area appropriated for urban land uses is growing even faster, these patterns of growth all but guarantee that the influ- ences of urban land uses will continue to expand over time. Cities and suburbia obviously provide the homes and livelihood for most of the nationâs population. But, as this report makes clear, these benefits have been accompanied by signifi- cant environmental change. Urbanization of the landscape profoundly affects how water moves both above and below ground during and following storm events; the quality of that stormwater (defined in Box 1-1); and the ultimate condition of nearby rivers, lakes, and estuaries. Unlike agriculture, which can display significant interchange with forest cover over time scales of a century (e.g., Hart, 1968), there is no indication that once-urbanized land ever returns to a less intensive state. Urban land, however, does continue to change over time; by one estimate, 42 percent of land currently considered âurbanâ in the United States will be redeveloped by 2030 (Brookings Institute, 2004). In their words, ânearly half of what will be the built environment in 2030 doesnât even exist yetâ (p. vi). This truth belies the common belief that efforts to improve man- agement of stormwater are doomed to irrelevancy because so much of the land- scape is already built. Opportunities for improvement have indeed been lost, but many more still await an improved management approach. Measures of urbanization are varied, and the disparate methods of quantify- ing the presence and influence of human activity tend to confound analyses of environmental effects. Population density is a direct metric of human presence, but it is not the most relevant measure of the influence of those people on their surrounding landscape. Expressions of the built environment, most commonly 13
14 URBAN STORMWATER MANAGEMENT IN THE UNITED STATES FIGURE 1-1 Histogram of population for the United States, based on 2000 census data. The median population density is about 1,000 people/km2. SOURCE: Modified from Pozzi 2 and Small (2005), who place the ruralâsuburban boundary at 100 people/km . Reprinted, with permission, from ASPRS (2005). Copyright 2005 by the American Society for Photo- grammetry and Remote Sensing. BOX 1-1 What Is âStormwaterâ? âStormwaterâ is a term that is used widely in both scientific literature and regulatory documents. It is also used frequently throughout this report. Although all of these usages share much in common, there are important differences that benefit from an explicit discus- sion. Most broadly, stormwater runoff is the water associated with a rain or snow storm that can be measured in a downstream river, stream, ditch, gutter, or pipe shortly after the pre- cipitation has reached the ground. What constitutes âshortlyâ depends on the size of the watershed and the efficiency of the drainage system, and a number of techniques exist to precisely separate stormwater runoff from its more languid counterpart, âbaseflow.â For small and highly urban watersheds, the interval between rainfall and measured stormwater discharges may be only a few minutes. For watersheds of many tens or hundreds of square miles, the lag between these two components of storm response may be hours or even a day. From a regulatory perspective, stormwater must pass through some sort of engi- neered conveyance, be it a gutter, a pipe, or a concrete canal. If it simply runs over the ground surface, or soaks into the soil and soon reemerges as seeps into a nearby stream, it may be water generated by the storm but it is not regulated stormwater. This report emphasizes the first, more hydrologically oriented definition. However, at- tention is focused mainly on that component of stormwater that emanates from those parts of a landscape that have been affected in some fashion by human activities (âurban storm- waterâ). Mostly this includes water that flows over the ground surface and is subsequently collected by natural channels or artificial conveyance systems, but it can also include water that has infiltrated into the ground but nonetheless reaches a stream channel relatively rapidly and that contributes to the increased stream discharge that commonly accompanies almost any rainfall event in a human-disturbed watershed.
INTRODUCTION 15 road density or pavement coverage as a percentage of gross land area, are more likely to determine stormwater runoff-related consequences. An inverse metric, the percentage of mature vegetation or forest across a landscape, expresses the magnitude of related, but not identical, impacts to downstream systems. Alter- natively, these measures of land cover can be replaced by measures of land use, wherein the types of human activity (e.g., residential, industrial, commercial) are used as proxies for the suite of hydrologic, chemical, and biological changes imposed on the surrounding landscape. All of these metrics of urbanization are strongly correlated, although none can directly substitute for another. They also are measured differently, which renders one or another more suitable for a given application. Land use is a common measure in the realm of urban planning, wherein current and future conditions for a city or an entire region are characterized using equivalent cate- gories across parcels, blocks, or broad regions. Road density can be reliably and rapidly measured, either manually or in a Geographic Information System envi- ronment, and it commonly displays a very good correlation with other measures of human activity. âLand cover,â however, and particularly the percentage of impervious cover, is the metric most commonly used in studying the effects of urban development on stormwater, because it clearly expresses the hydrologic influence and watershed scale of urbanization. Box 1-2 describes the ways in which the percent of impervious cover in a watershed is measured. There is no universally accepted terminology to describe land-cover or land- use conditions along the rural-to-urban gradient. Pozzi and Small (2005), for example, identified ârural,â âsuburban,â and âurbanâ land uses on the basis of population density and vegetation cover, but they did not observe abrupt transi- tions that suggested natural boundaries (see Figure 1-1). In contrast, the Center for Watershed Protection (2005) defined the same terms but used impervious area percentage as the criterion, with such labels as âruralâ (0 to 10 percent im- perviousness), âsuburbanâ (10 to 25 percent imperviousness), âurbanâ (25 to 60 percent imperviousness) and âultra-urbanâ (greater than 60 percent impervious- ness). Beyond the problems posed by precise yet inconsistent definitions for commonly used words, none of the boundaries specified by these definitions are reflected in either hydrologic or ecosystem responses. Hydrologic response is strongly dependent on both land cover and drainage connectivity (e.g., Leopold, 1968); ecological responses in urbanizing watersheds do not show marked thresholds along an urban gradient (e.g., Figure 1-2) and they are dependent on not only the sheer magnitude of urban development but also the spatial configu- ration of that development across the watershed (Alberti et al., 2006). This re- port, therefore, uses such terms as âurbanâ and âsuburbanâ under their common usage, without implying or advocating for a more precise (but ultimately limited and discipline-specific) definition. Changing land cover and land use influence the physical, chemical, and bio- logical conditions of downstream waterways. The specific mechanisms by which this influence occurs vary from place to place, and even a cursory review
16 URBAN STORMWATER MANAGEMENT IN THE UNITED STATES BOX 1-2 Measures of Impervious Cover The percentage of impervious surface or cover in a landscape is the most frequently used measure of urbanization. Yet this parameter has its limitations, in part because it has not been consistently used or defined. Most significant is the distinction between total imper- vious area (TIA) and effective impervious area (EIA). TIA is the âintuitiveâ definition of impervi- ousness: that fraction of the watershed covered by constructed, non-infiltrating surfaces such as concrete, asphalt, and buildings. Hydrologically, however, this definition is incomplete for two reasons. First, it ignores nominally âperviousâ surfaces that are sufficiently compacted or otherwise so low in permeability that the rate of runoff from them is similar or indistinguishable from pavement. For example, Burges and others (1998) found that the impervious unit-area runoff was only 20 percent greater than that from pervious areasâprimarily thin sodded lawns over glacial tillâin a western Washington residential subdivision. Clearly, this hydrologic con- tribution cannot be ignored entirely. The second limitation of TIA is that it includes some paved surfaces that may contribute nothing to the stormwater-runoff response of the downstream channel. A gazebo in the middle of parkland, for example, probably will impose no hydrologic changes into the catchment except for a very localized elevation of soil moisture at the edge of its roof. Less obvious, but still rele- vant, would be the different downstream consequences of rooftops that drain alternatively into a piped storm-drain system with direct discharge into a natural stream or onto splash blocks that disperse the runoff onto the garden or lawn at each corner of the building. This metric therefore cannot recognize any stormwater mitigation that may result from alternative runoff- management strategies, for example, pervious pavements or rainwater harvesting. The first of these TIA limitations, the production of significant runoff from nominally pervi- ous surfaces, is typically ignored in the characterization of urban development. The reason for such an approach lies in the difficulty in identifying such areas and estimating their contribution, and because of the credible belief that the degree to which pervious areas shed water as over- land flow should be related, albeit imperfectly, with the amount of impervious area: where con- struction and development are more intense and cover progressively greater fractions of the of the literature demonstrates that many different factors can be important, such as changes to flow regime, physical and chemical constituents in the water col- umn, or the physical form of the stream channel itself (Paul and Meyer, 2001). Not all of these changes are present in any given systemâlakes, wetlands, and streams can be altered by human activity in many different ways, each unique to the activity and the setting in which it occurs. Nonetheless, direct influences of land-use change on freshwater systems commonly include the following (Nai- man and Turner, 2000): Altering the composition and structure of the natural flora and fauna, Changing disturbance regimes, Fragmenting the land into smaller and more diverse parcels, and Changing the juxtaposition between parcel types. Historically, human-induced alteration was not universally seen as a prob- lem. In particular, dams and other stream-channel âimprovementsâ were a
INTRODUCTION 17 watershed, it is more likely that the intervening green spaces have been stripped and com- pacted during construction and only imperfectly rehabilitated for their hydrologic functions dur- ing subsequent âlandscaping.â The second of these TIA limitations, inclusion of non-contributing impervious areas, is formally addressed through the concept of EIA, defined as the impervious surfaces with direct hydraulic connection to the downstream drainage (or stream) system. Thus, any part of the TIA that drains onto pervious (i.e., âgreenâ) ground is excluded from the measurement of EIA. This parameter, at least conceptually, captures the hydrologic significance of imperviousness. EIA is the parameter normally used to characterize urban development in hydrologic models. The direct measurement of EIA is complicated. Studies designed specifically to quantify this parameter must make direct, independent measurements of both TIA and EIA (Alley and Veenhuis, 1983; Laenen, 1983; Prysch and Ebbert, 1986). The results can then be general- ized either as a correlation between the two parameters or as a âtypicalâ value for a given land use. Sutherland (1995) developed an equation that describes the relationship between EIA and TIA. Its general form is: EIA = A (TIA)B where A and B are a unique combination of numbers that satisfy the following criteria: TIA = 1 then EIA = 0% TIA = 100 then EIA = 100% 1.41 A commonly used version of this equation (EIA = 0.15 TIA ) was based on samples from highly urbanized land uses in Denver, Colorado (Alley and Veenhuis, 1983; Gregory et al., 2005). These results, however, are almost certainly region- and even neighborhood- specific, and, although highly relevant to watershed studies, they can be quite laborious to develop. common activity of municipal and federal engineering works of the mid-20th century (Williams and Wolman, 1984). âFlood controlâ implied a betterment of conditions, at least for streamside residents (Chang, 1992). And fisheries âen- hancements,â commonly reflected by massive infrastructure for hatcheries or artificial spawning channels, were once seen as unequivocal benefits for fish populations (White, 1996; Levin et al., 2001). By almost any currently applied metric, however, the net result of human al- teration of the landscape to date has resulted in a degradation of the conditions in downstream watercourses. Many prior researchers, particularly when consid- ering ecological conditions and metrics, have recognized a crude but monotoni- cally declining relationship between human-induced landscape alteration and downstream conditions (e.g., Figure 1-2; Horner et al., 1997; Davies and Jack- son, 2006). These include metrics of physical stream-channel conditions (e.g., Bledsoe and Watson, 2001), chemical constituents (e.g., Figure 1-3; House et al., 1993), and biological communities (e.g., Figure 1-4; Steedman, 1988; Wang et al., 1997).
18 URBAN STORMWATER MANAGEMENT IN THE UNITED STATES FIGURE 1-2 Conceptual model (top) and actual response (bottom) of a biological systemâs response to stress. The âUrban Gradient of Stressorsâ might be a single metric of urbani- zation, such as percent watershed impervious or road density; the âBiological Indicatorâ may be single-metric or multi-metric measures of the level of disturbance in an aquatic community. The right-declining line traces the limits of a âfactor-ceiling distributionâ (Thom- son et al., 1986), wherein individual sites (i.e., data points) have a wide range of potential values for a given position along the urban gradient but are not observed above a maxi- mum possible limit of the biological index. The bottom graph illustrates actual biological responses, using a biotic index developed to show responses to urban impacts plotted against a standardized urban gradient comprising urban land use, road density, and popu- lation. SOURCE: Top figure reprinted, with permission, from Davies and Jackson (2006). Copyright by the Ecological Society of America. Bottom figure reprinted, with permission, from Barbour et al. (2006). Copyright by the Water Environment Research Foundation.
INTRODUCTION 19 FIGURE 1-3 Example relationships between road density (a surrogate measure of urban development) and common water quality constituents. Direct causality is not necessarily implied by such relationships, but the monotonic increase in concentrations with increasing âurbanization,â however measured, is near-universal. SOURCE: Reprinted, with permis- sion, from Chang and Carlson (2005). Copyright 2005 by Springer.
20 URBAN STORMWATER MANAGEMENT IN THE UNITED STATES FIGURE 1-4 Plots of Effective Impervious Area (EIA, or âconnected imperviousnessâ) against metrics of biologic response in fish populations. SOURCE: Reprinted, with permis- sion, from Wang et al. (2001). Copyright 2001 by Springer. The association between watercourse degradation and landscape alteration in general, and urban development in particular, seems inexorable. The scien- tific and regulatory challenge of the last three decades has been to decouple this relationship, in some cases to reverse its trend and in others to manage where these impacts are to occur. WHATâS WRONG WITH THE NATIONâS WATERS? Since passage of the Water Quality Act of 1948 and the Clean Water Act (CWA) of 1972, 1977, and 1987, water quality in the United States has meas- urably improved in the major streams and rivers and in the Great Lakes. How- ever, substantial challenges and problems remain. Major reporting efforts that have examined state and national indicators of condition, such as CWA 305(b) reports (EPA, 2002) and the Heinz State of the Nationâs Ecosystem report (Heinz Center, 2002), or environmental monitoring that was designed to provide statistically valid estimates of condition (e.g., National Wadeable Stream As- sessment; EPA, 2006), have confirmed widespread impairments related to dif- fuse sources of pollution and stressors. The National Water Quality Inventory (derived from Section 305b of the CWA) compiles data in relation to use designations and water quality standards.
INTRODUCTION 21 As discussed in greater detail in Chapter 2, such standards include both (1) a description of the use that a waterbody is supposed to achieve (such as a source of drinking water or a cold water fishery) and (2) narrative or numeric criteria for physical, chemical, and biological parameters that allow the designated use to be achieved. As of 2002, 45 percent of assessed streams and rivers, 47 per- cent of assessed lakes, 32 percent of assessed estuarine areas, 17 percent of as- sessed shoreline miles, 87 percent of near-coastal ocean areas, 51 percent of assessed wetlands, 91 percent of assessed Great Lakes shoreline miles, and 99 percent of assessed Great Lakes open water areas were not meeting water qual- ity standards set by the states (2002 EPA Report to Congress).1 The U.S. Environmental Protection Agency (EPA) has also embarked on a five-year statistically valid survey of the nationâs waters (http://www.epa.gov/ owow/monitoring/guide.pdf). To date, two waterbody typesâcoastal areas and wadeable streamsâhave been assessed. The most recent data indicate that 42 percent of wadeable streams are in poor biological condition and 25 percent are in fair condition (EPA, 2006). The overall condition of the nationâs estuaries is generally fair, with Puerto Rico and Northeast Coast regions rated poor, the Gulf Coast and West Coast regions rated fair, and the Southeast Coast region rated good to fair (EPA, 2007). These condition ratings for the National Estuary Pro- gram are based on a water quality index, a sediment quality index, a benthic index, and a fish tissue contaminants index. The impairment of waterbodies is manifested in a multitude of ways. In- deed, EPAâs primary process for reporting waterbody condition (Section 303(d) of the CWAâsee Chapter 2) identifies over 200 distinct types of impairments. As shown in Table 1-1, these have been categorized into 15 broad categories, encompassing about 94 percent of all impairments. 59,515 waterbodies fall into one of the top 15 categories, while the total reported number of waterbodies impaired from all causes is 63,599 (which is an underestimate of the actual total because not all waterbodies are assessed). Mercury, microbial pathogens, sedi- ments, other metals, and nutrients are the major pollutants associated with im- paired waterbodies nationwide. These constituents have direct impacts on aquatic ecosystems and public health, which form the basis of the water quality standards set for these compounds. Sediments can harm fish and macroinverte- brate communities by introducing sorbed contaminants, decreasing available light in streams, and smothering fish eggs. Microbial pathogens can cause dis- ease to humans via both ingestion and dermal contact and are frequently cited as the cause of beach closures and other recreational water hazards in lakes and estuaries. Nutrient over-enrichment can promote a cascade of events in water- bodies from algal blooms to decreases in dissolved oxygen and associated fish kills. Metals like mercury, pesticides, and other organic compounds that enter 1 EPA does not yet have the 2004 assessment findings compiled in a consistent format from all the states. EPA is also working on processing the states 2006 Integrated Reports as the 303(d) portions are approved and the states submit their final assessment findings. Susan Holdsworth, EPA, personal communication, September 2007.
22 URBAN STORMWATER MANAGEMENT IN THE UNITED STATES waterways can be taken up by fish species, accumulating in their tissues and presenting a health risk to organisms (including humans) that consume the fish. However, Table 1-1 can be misleading if it implies that degraded water quality is the primary metric of impairment. In fact, many of the nationâs streams, lakes, and estuaries also suffer from fundamental changes in their flow regime and energy inputs, alteration of aquatic habitats, and resulting disruption of biotic interactions that are not easily measured via pollutant concentrations. Such waters may not be listed on State 303(d) lists because of the absence of a corresponding water quality standard that would directly indicate such condi- tions (like a biocriterion). Figure 1-5A, B, and C show examples of such im- pacted waterbodies. TABLE 1-1 Top 15 Categories of Impairment Requiring CWA Section 303(d) Action Cause of Impairment Number of Waterbodies Percent of the Total Mercury 8,555 14% Pathogens 8,526 14% Sediment 6,689 11% Metals (other than mercury) 6,389 11% Nutrients 5,654 10% Oxygen depletion 4,568 8% pH 3,389 6% Cause unknown - biological 2,866 5% integrity Temperature 2,854 5% Habitat alteration 2,220 4% PCBs 2,081 3% Turbidity 2,050 3% Cause unknown 1,356 2% Pesticides 1,322 2% Salinity/TDS/chlorides 996 2% Note: âWaterbodiesâ refers to individual river segments, lakes, and reservoirs. A single waterbody can have multiple impairments. Because most waters are not assessed, how- ever, there is no estimate of the number of unimpaired waters in the United States. SOURCE: EPA, National Section 303(d) List Fact Sheet (http://iaspub.epa.gov/waters/ national_rept.control). The data are based on three-fourths of states reporting from 2004 lists, with the remaining from earlier lists and one state from a 2006 list.
INTRODUCTION 23 FIGURE 1-5A Headwater tributary in Philadelphia suffering from Urban Stream Syndrome. SOURCE: Courtesy of Chris Crockett, Philadelphia Water Department. Center for Watershed Protection FIGURE 1-5B A destabilized stream in Vermont. SOURCE: Courtesy of Pete LaFlamme, Vermont Department of Environmental Conservation.
24 URBAN STORMWATER MANAGEMENT IN THE UNITED STATES FIGURE 1-5C An urban stream, the Lower Oso Creek in Orange County, California, fol- lowing a storm event. Oso Creek was formerly an ephemeral stream, but heavy develop- ment in the contributing watershed has created perennial flowâstormwater flow during wet weather and minor wastewater discharges and authorized non-stormwater discharges such as landscape irrigation runoff during dry weather. Courtesy of Eric Stein, Southern Califor- nia Coastal Research Water Project. Over the years, the greatest successes in improving the nationâs waters have been in abating the often severe impairments caused by municipal and industrial point source discharges. The pollutant load reductions required of these facili- ties have been driven by the National Pollutant Discharge Elimination System (NPDES) permit requirements of the CWA (see Chapter 2). Although the major- ity of these sources are now controlled, further declines in water quality remain likely if the land-use changes that typify more diffuse sources of pollution are not addressed (Palmer and Allan, 2006). These include land-disturbing agricul- tural, silvicultural, urban, industrial, and construction activities from which hard-to-monitor pollutants emerge during wet-weather events. Pollution from these landscapes has been almost universally acknowledged as the most pressing challenge to the restoration of waterbodies and aquatic ecosystems nationwide. All population and development forecasts indicate a continued worsening of the environmental conditions caused by diffuse sources of pollution under the na- tionâs current growth and land-use trajectories. Recognition of urban stormwaterâs role in the degradation of the nationâs waters is but the latest stage in the history of this byproduct of the human envi-
INTRODUCTION 25 ronment. Runoff conveyance systems have been part of cities for centuries, but they reflected only the desire to remove water from roads and walkways as rap- idly and efficiently as possible. In some arid environments, rainwater has al- ways been collected for irrigation or drinking; elsewhere it has been treated as an unmetered, and largely benign, waste product of cities. Minimal (unengi- neered) ditches or pipes drained developed areas to the nearest natural water- course. Where more convenient, stormwater shared conveyance with wastewa- ter, eliminating the cost of a separate pipe system but commonly resulting in sewage overflows during rainstorms. Recognition of downstream flooding that commonly resulted from upstream development led to construction of stormwa- ter storage ponds or vaults in many municipalities in the 1960s, but their per- formance has typically fallen far short of design objectives (Booth and Jackson, 1997; Maxted and Shaver, 1999; Nehrke and Roesner, 2004). Water-quality treatment has been a relatively recent addition to the management of stormwater, and although a significant fraction of pollutants can be removed through such efforts (e.g., Strecker et al., 2004; see http://www.bmpdatabase.org), the con- stituents remaining even in âtreatedâ stormwater represent a substantial, but largely unappreciated, impact to downstream watercourses. Of the waterbodies that have been assessed in the United States, impair- ments from urban runoff are responsible for about 38,114 miles of impaired riv- ers and streams, 948,420 acres of impaired lakes, 2,742 square miles of impaired bays and estuaries, and 79,582 acres of impaired wetlands (2002 305(b) report). These numbers must be considered an underestimate, since the urban runoff category does not include stormwater discharges from municipal separate storm sewer systems (MS4s) and permitted industries, including construction. Urban stormwater is listed as the âprimaryâ source of impairment for 13 percent of all rivers, 18 percent of all lakes, and 32 percent of all estuaries (2000 305(b) re- port). Although these numbers may seem low, urban areas cover just 3 percent of the land mass of the United States (Loveland and Auch, 2004), and so their influence is disproportionately large. Indeed, developed and developing areas that are a primary focus of stormwater regulations contain some of the most de- graded waters in the country. For example, in Ohio few sites with greater than 27 percent imperviousness can meet interim CWA goals in nearby waterbodies, and biological degradation is observed with much less urban development (Miltner et al., 2004). Numerous authors have found similar patterns (see Meyer et al., 2005). Although no water quality inventory data have been made available from the EPA since 2002, the dimensions of the stormwater problem can be further gleaned from several past regional and national water quality inventories. Many of these assessments are somewhat dated and are subject to the normal data and assessment limitations of national assessment methods, but they indicate that stormwater runoff has a deleterious impact on nearly all of the nationâs waters. For example:
26 URBAN STORMWATER MANAGEMENT IN THE UNITED STATES Harvesting of shellfish is prohibited, restricted, or conditional in nearly 40 percent of all shellfish beds nationally due to high bacterial levels, and urban runoff and failing septic systems are cited as the prime causes. Reopening of shellfish beds due to improved wastewater treatment has been more than offset by bed closures due to rapid coastal development (NOAA, 1992; EPA, 1998). In 2006 there were over 15,000 beach closings or swimming advisories due to bacterial levels exceeding health and safety standards, with polluted run- off and stormwater cited as the cause of the impairment 40 percent of the time (NRDC, 2007). Pesticides were detected in 97 percent of urban stream water samples across the United States, and exceeded human health and aquatic life bench- marks 6.7 and 83 percent of the time, respectively (USGS, 2006). In 94 percent of fish tissues sampled in urban areas nationwide, organochlorine compounds were detected. Urban development was responsible for almost 39 percent of freshwa- ter wetland loss (88,960 acres) nationally between 1998 and 2004 (Dahl, 2006), and the direct impact of stormwater runoff in degrading wetland quality is pre- dicted to affect an even greater acreage (Wright et al., 2006). Eastern brook trout are present in intact populations in only 5 percent of more than 12,000 subwatersheds in their historical range in eastern North America, and urbanization is cited as a primary threat in 25 percent of the re- maining subwatersheds with reduced populations (Trout Unlimited, 2006). Increased flooding is common throughout urban and suburban areas, sometimes as a consequence of improperly sited development (Figure 1-6A) but more commonly as a result of increasing discharges over time resulting from progressive urbanization farther upstream (Figure 1-6B). According to FEMA (undated), property damage from all types of flooding, from flash floods to large river floods, averages $2 billion a year. The chemical effects of stormwater runoff are pervasive and severe throughout the nationâs urban waterways, and they can extend far downstream of the urban source. Stormwater discharges from urban areas to marine and estua- rine waters cause greater water column toxicity than similar discharges from less urban areas (Bay et al., 2003). A variety of studies have shown that stormwater runoff is a vector of pathogens with potential human health implications in both freshwater (Calderon et al., 1991) and marine waters (Dwight et al., 2004; Colford et al., 2007).
INTRODUCTION 27 A A B B FIGURE 1-6 (A) New residential construction in the path of episodic stream discharge th (Issaquah, Washington); (B) recent flooding of an 18 -century tavern in Collegeville, Penn- sylvania following a storm event in an upstream developing watershed. SOURCES: Top, Derek Booth, Stillwater Sciences, Inc., and bottom, Robert Traver, Villanova University. WHY IS IT SO HARD TO REDUCE THE IMPACTS OF STORMWATER? âUrban stormwaterâ is the runoff from a landscape that has been affected in some fashion by human activities, during and immediately after rain. Most visi- bly, it is the water flow over the ground surface, which is collected by natural channels and artificial conveyance systems (pipes, gutters, and ditches) and ul- timately routed to a stream, river, lake, wetland, or ocean. It also includes water that has percolated into the ground but nonetheless reaches a stream channel relatively rapidly (typically within a day or so of the rainfall), contributing to the high discharge in a stream that commonly accompanies rainfall. The subsurface
28 URBAN STORMWATER MANAGEMENT IN THE UNITED STATES flow paths that contribute to this stormflow response are typically quite shallow, in the upper layers of the soil, and are sometimes termed âinterflow.â They stand in contrast to deeper groundwater paths, where water moves at much lower velocities by longer paths and so reaches the stream slowly, over periods of days, weeks, or months. This deeper flow sustains streamflow during rainless periods and is usually called baseflow, as distinct from âstormwater.â A formal distinction between these types of runoff is sometimes needed for certain com- putational procedures, but for most purposes a qualitative understanding is suffi- cient. These runoff paths can be identified in virtually all modified landscapes, such as agriculture, forestry, and mining. However, this report focuses on those settings with the particular combination of activities that constitute âurbaniza- tion,â by which we mean to include the commonly understood conversion (whether incremental or total) of a vegetated landscape to one with roads, houses, and other structures. Although the role of urban stormwater in degrading the nationâs waters has been recognized for decades (e.g., Klein, 1979), reducing that role has been no- toriously difficult. This difficulty arises from three basic attributes of what is commonly termed âstormwaterâ: 1. It is produced from literally everywhere in a developed landscape; 2. Its production and delivery are episodic, and these fluctuations are dif- ficult to attenuate; and 3. It accumulates and transports much of the collective waste of the urban environment. Wherever grasslands and forest are replaced by urban development in gen- eral, and impervious surfaces in particular, the movement of water across the landscape is radically altered (see Figure 1-7). Nearly all of the associated prob- lems result from one underlying cause: loss of the water-retaining function of the soil and vegetation in the urban landscape. In an undeveloped, vegetated land- scape, soil structure and hydrologic behavior are strongly influenced by biologi- cal activities that increase soil porosity (the ratio of void space to total soil vol- ume) and the number and size of macropores, and thus the storage and conduc- tivity of water as it moves through the soil. Leaf litter on the soil surface dissi- pates raindrop energy; the soilâs organic content reduces detachment of small soil particles and maintains high surface infiltration rates. As a consequence, rainfall typically infiltrates into the ground surface or is evapotranspired by vegetation, except during particularly intense rainfall events (Dunne and Leo- pold, 1978). In the urban landscape, these processes of evapotranspiration and water reten- tion in the soil may be lost for the simple reason that the loose upper layers of the soil and vegetation are goneâstripped away to provide a better foundation for roads and buildings. Even if the soil still exists, it no longer functions if precipita- tion is denied access because of paving or rooftops. In either case, a stormwater
INTRODUCTION 29 FIGURE 1-7 Schematic of the hydrologic pathways in humid-region watersheds, before and after urban development. The sizes of the arrows suggest relative magnitudes of the different elements of the hydrologic cycle, but conditions can vary greatly between individ- ual catchments and only the increase in surface runoff in the post-development condition is ubiquitous. SOURCE: Adapted from Schueler (1987) and Maryland Department of the Environment; http://www.mde.state.md.us/Programs/WaterPrograms. runoff reservoir of tremendous volume is removed from the stormwater runoff system; water that may have lingered in this reservoir for a few days or many weeks, or been returned directly to the atmosphere by evaporation or transpiration by plants, now flows rapidly across the land surface and arrives at the stream channel in short, concentrated bursts of high discharge. This transformation of the hydrologic regime from one where subsurface flow once dominated to one where overland flow now dominates is not simply a read- justment of runoff flow paths, and it does not just result in a modest increase in flow volumes. It is a wholesale reorganization of the processes of runoff genera- tion, and it occurs throughout the developed landscape. As such, it can affect every aspect of that runoff (Leopold, 1968)ânot only its rate of production, its volume, and its chemistry, but also what it indirectly affects farther downstream (Walsh et al., 2005a). This includes erosion of mobile channel boundaries, mobili- zation of once-static channel elements (e.g., large logs), scavenging of contami- nants from the surface of the urban landscape, and efficient transfer of heat from warmed surfaces to receiving waterbodies. These changes have commonly in- spired human reactionsâtypically with narrow objectives but carrying additional, far-ranging consequencesâsuch as the piping of once-exposed channels, bank armoring, and construction of large open-water detention ponds (e.g., Lieb and Carline, 2000).
30 URBAN STORMWATER MANAGEMENT IN THE UNITED STATES This change in runoff regime is also commonly accompanied by certain land- use activities that have the potential to generate particularly harmful or toxic dis- charges, notably those commercial activities that are the particular focus of the industrial NPDES permits. These include manufacturing facilities, transport of freight or passengers, salvage yards, and a more generally defined category of âsites where industrial materials, equipment, or activities are exposed to stormwa- terâ (e.g., EPA, 1992). Other human actions are associated with urban landscapes that do not affect stormwater directly, but which can further amplify the negative consequences of altered flow. These actions include clearing of riparian vegetation around streams and wetlands, introduction of atmospheric pollutants that are subsequently depos- ited, inadvertent release of exotic chemicals into the environment, and channel crossings by roads and utilities. Each of these additional actions further degrades downstream waterbodies and increases the challenge of finding effective meth- ods to reverse these changes (Boulton, 1999). There is little doubt as to why the problem of urban stormwater has not yet been âsolvedââbecause every func- tional element of an aquatic ecosystem is affected. Urban stormwater has re- sulted in such widespread impacts, both physical and biological, in aquatic sys- tems across the world that this phenomenon has been termed the âUrban Stream Syndromeâ (see Figure 1-5; Walsh et al., 2005b). Of the many possible ways to consider these conditions, Karr (1991) has recommended a simple yet comprehensive grouping of the major stressors aris- ing from urbanization that influence aquatic assemblages (Figure 1-8). These include chemical pollutants (water quality and toxicity); changes to flow magni- tude, frequency, and seasonality of various discharges; the physical aspects of stream, lake, or wetland habitats; the energy dynamics of food webs, sunlight, and temperature; and biotic interactions between native and exotic species. Stormwater and stormwater-related impacts encompass all of these categories, some directly (e.g., water chemistry) and some indirectly (e.g., habitat, energy dynamics). Because of the wide-ranging effects of stormwater, programs to abate stormwater impacts on aquatic systems must deal with a broad range of impairments far beyond any single altered feature, whether traditional water- chemistry parameters or flow rates and volumes. The broad spatial scale of where and how these impacts are generated sug- gests that solutions, if effective, should be executed at an equivalent scale. Al- though the âproblemâ of stormwater runoff is manifested most directly as an altered hydrograph or elevated concentrations of pollutants, it is ultimately an expression of land-use change at a landscape scale. Symptomatic solutions, applied only at the end of a stormwater collection pipe, are not likely to prove fully effective because they are not functioning at the scale of the original dis- turbance (Kloss and Calarusse, 2006). The landscape-scale generation of stormwater has a number of conse- quences for any attempt to reduce its effects on receiving waters, as described below.
INTRODUCTION 31 Urbanization Urbanization drivers effects Human population Impervious area Vegetation loss Road density FIGURE 1-8 Five features that are affected by urban development and, in turn, affect bio- logical conditions in urban streams. SOURCES: Modified from Karr (1991), Karr and Yoder (2004), and Booth (2005). Reprinted, with permission, from Karr (1991). Copyright 2001 by Ecological Society of America. Reprinted, with permission, from Karr and Yoder (2004). Copyright 2004 by American Society of Civil Engineers. Reprinted, with permission, from Booth (2005). Copyright 2005 by the North American Benthological Society. Sources and Volumes The âsourceâ of stormwater runoff is dispersed, making collection and cen- tralized treatment challenging. To the extent that collection is successful, how- ever, the flip side of this conditionâvery large volumesâbecomes manifest. Either an extensive infrastructure brings stormwater to centralized facilities, whose operation and maintenance may be relatively straightforward (e.g., Anderson et al., 2002) but of modest effectiveness, or stormwater remains dis- persed for management, treatment, or both across the landscape (e.g., Konrad and Burges, 2001; Holman-Dodds et al., 2003; Puget Sound Action Team, 2005; Walsh et al., 2005a; Bloom, 2006; van Roon, 2007), better mimicking the natu- ral processes of runoff generation but requiring a potentially unlimited number of âfacilitiesâ that may have their own particular needs for space, cost, and maintenance.
32 URBAN STORMWATER MANAGEMENT IN THE UNITED STATES Treatment Challenges Regardless of the scale at which treatment is attempted, technological diffi- culties are significant because of the variety of âpollutantsâ that must be ad- dressed. These include physical objects, from large debris to microscopic parti- cles; chemical constituents, both dissolved and immiscible; and less easily cate- gorized properties such as temperature. Wastewater treatment plants manage a similarly broad range of pollutants, but stormwater flows have highly unsteady inflows and, when present, typically much greater volumes to treat. Industrial sources of stormwater pose a particularly challenging problem because potential generators of polluted or toxic runoff are widespread and are regulated under NPDES permitting by their activities, not by the specific cate- gory of industrial activity under which they fall. This complicates any system- atic effort to identify those entities that should be regulated (Duke et al., 1999). Even for the limited number of regulated generators, pollution prevention meas- ures are of uncertain effectiveness. Soil erosion from construction sites is another pollution source that has proven difficult to effectively control. Although most bare sites are relatively small and only short-lived, at any given time there can be many sites under con- struction, each of which can deliver sediment loads to downstream waterbodies at rates that exceed background levels by many orders of magnitude (e.g., Wol- man and Schick, 1967). Relatively effective approaches and technologies exist to dramatically reduce the magnitude of these sediment discharges (e.g., Raskin et al., 2005), but they depend on conscientious installation and regular mainte- nance. Enforcement of such requirements, normally a low-priority activity of local departments of building or public works, is commonly lacking. Another difference between the stormwater and wastewater streams is that stormwater treatment must address not only âpollutantsâ but also physically and ecologically deleterious changes in flow rate and total runoff volume. Treating these changes constitutes a particularly difficult task for two reasons. First, there is simply more runoff, as a rule, and so replicating the predevelopment hydrograph is not an optionâthe increased volume of runoff guarantees that some discharges, some of the time, must be allowed to increase. Second, there is little agreement on what constitutes âadequateâ or âeffectiveâ treatment for the various attributes of flow. Even the most basic metrics, such as the magni- tude of peak flow, can require extensive infrastructure to achieve (e.g., Booth and Jackson, 1997); other flow metrics that correlate more directly with unde- sired effects on physical and biological systems can require even greater efforts to match. In many cases, the urban-induced transformation of the flow regime makes true âmitigationâ virtually impossible.
INTRODUCTION 33 Widespread Cause and Effects The spatial scale of stormwater generation and its impacts is wide-ranging. âGeneratorsâ are literally landscape-wide, and impacts can occur at every loca- tion in the path followed by urban runoff, from source to receiving waterbody (Hamilton et al., 2004). There are few ways to demonstrate causal connections between distributed landscape sources and cumulative downstream effects (Allan, 2004), and so site-specific mitigation typically provides little lasting improvement in the watershed as a whole (Maxted and Shaver, 1997). Stormwater Measurements The desired attributes of stormwater runoff are normally expressed through a combination of physical and chemical parameters. These parameters are commonly presumed to have direct correlation to attributes of human or eco- logical concern, such as the condition of human or fish communities, or the sta- bility of a stream channel, even though these parameters do not directly measure those effects. The most commonly measured physical parameters are hydrologic and simply measure the rate of flow past a specified location. Both the absolute, instantaneous magnitude of that flow rate (i.e., the discharge) and the variations in that rate over multiple time scales (i.e., how rapidly the discharge varies over an hour, a day, a season, etc.) can be captured by analysis of a continuous time series of a flow. Obviously, however, a nearly unlimited number of possible metrics, capturing a multitude of temporal scales, could be defined (Poff et al., 1997, 2006; Cassin et al., 2004; Konrad et al., 2005; Roy et al., 2005; Chang, 2007). Commonly only a single parameterâthe peak storm discharge for a given return period (Hollis, 1975)âhas been emphasized in the past. Mitigation of urban-induced flow increases have followed this narrow approach, typically by endeavoring to reduce peak discharge by use of detention ponds but leaving the underlying increase in runoff volumesâand the associated augmentation of both frequency and duration of high dischargesâuntouched. This partly ex- plains why evaluation of downstream conditions commonly document little im- provement resulting from traditional flow-mitigation measures (e.g., Maxted and Shaver, 1997; Roesner et al., 2001; May and Horner, 2002). Other physical parameters, less commonly measured or articulated, can also express the conditions of downstream watercourses. Measures of size or com- plexity, particularly for stream channels, are particularly responsive to the changes in flow regime and discharge. Booth (1990) suggested that discriminat- ing between channel expansion, the proportional increase in channel cross- sectional area with increasing discharge, and channel incision, the catastrophic vertical downcutting that sometimes accompanies urban-induced flow increases, captures important end-members of the physical response to hydrologic change. The former (proportional expansion) is more thoroughly documented (Hammer, 1972; Hollis and Luckett, 1976; Morisawa and LaFlure, 1982; Neller, 1988;
34 URBAN STORMWATER MANAGEMENT IN THE UNITED STATES Whitlow and Gregory, 1989; Booth and Jackson, 1997; Moscrip and Montgom- ery, 1997; Booth and Henshaw, 2001); the latter (catastrophic incision) is more difficult to quantify but has been recognized in both urban and agricultural set- tings (e.g., Simon, 1989). Both types of changes result not only in a larger channel but also in substantial simplification and loss of features normally asso- ciated with high-quality habitat for fish and other in-stream biota. The sediment released by these âgrowing channelsâ also can be the largest component of the overall sediment load delivered to downstream waterbodies (Trimble, 1997; Nelson and Booth, 2002). Chemical parameters (or, historically, âwater-quality parametersâ; see Din- ius, 1987; Gergel et al., 2002) cover a host of naturally and anthropogenically occurring constituents in water. In flowing water these are normally expressed as instantaneous measurements of concentration. In waterbodies with long resi- dence times, such as lakes, these may be expressed as either concentrations or as loads (total accumulated amounts, or total amounts integrated over an extended time interval). The CWA defined a list of priority pollutants, of which a subset is regularly measured in many urban streams (e.g., Field and Pitt, 1990). Pa- rameters that are not measured may or may not be present, but without assess- ment they are rarely recognized for their potential (or actual) contribution to waterbody impairment. Other attributes of stormwater do not fit as neatly into the categories of wa- ter quantity or water quality. Temperature is commonly measured and is nor- mally treated as a water quality parameter, although it is obviously not a chemi- cal property of the water (LeBlanc et al., 1997; Wang et al., 2003). Similarly, direct or indirect measures of suspended matter in the water column (e.g., con- centration of total suspended solids, or secchi disk depths in a lake) are primarily physical parameters but are normally included in water quality metrics. Flow velocity is rarely measured in either context, even though it too correlates di- rectly to stream-channel conditions. Even more direct expressions of a flowâs ability to transport sediment or other debris, such as shear stress or unit stream power, are rarely reported and virtually never regulated. *** Urban runoff degrades aquatic systems in multiple ways, which confounds our attempts to define causality or to demonstrate clear linkages between mitiga- tion and ecosystem improvement. It is generally recognized from the conceptual models that seek to describe this system that no single element holds the key to ecosystem condition. All elements must be functional, and yet every element can be affected by urban runoff in different ways. These impacts occur at virtu- ally all spatial scales, from the site-specific to the landscape; this breadth and diversity challenges our efforts to find effective solutions. This complexity and the continued growth of the built environment also present fundamental social choices and management challenges. Stormwater control measures entail substantial costs for their long-term maintenance, moni-
INTRODUCTION 35 toring to determine their performance, and enforcement of their useâall of which must be weighed against their (sometimes unproven) benefits. Further- more, the overarching importance of impervious surfaces inextricably links stormwater management to land-use decisions and policy. For example, where a reversal of the effects of urbanization cannot be realized, more intensive land- use development in certain areas may be a paradoxically appropriate response to reduce the overall impacts of stormwater. That is, increasing population density and impervious cover in designated urban areas may reduce the creation of im- pervious surface and the associated ecological impacts in areas that will remain undeveloped as a result. In these highly urban areas (with very high percentages of impervious surface), aquatic conditions in local streams will be irreversibly changed and the Urban Stream Syndrome may be unavoidable to some extent. Where these impacts occur and what effort and cost will be used to avoid these impacts are both fundamental issues confronting the nation as it attempts to ad- dress stormwater. IMPETUS FOR THE STUDY AND REPORT ROADMAP In 1972 Congress amended the Federal Water Pollution Control Act (subse- quently referred to as the Clean Water Act) to require control of discharges of pollutants to waters of the United States from point sources. Initial efforts to improve water quality using NPDES permits focused primarily on reducing pol- lutants from industrial process wastewater and municipal sewage discharges. These point source discharges were clearly and easily shown to be responsible for poor, often drastically degraded conditions in receiving waterbodies because they tended to emanate from identifiable and easily monitored locations, such as pipe outfalls. As pollution control measures for industrial process wastewater and mu- nicipal sewage were implemented and refined during the 1970s and 1980s, more diffuse sources of water pollution have become the predominant causes of water quality impairment, including stormwater runoff. To address the role of storm- water in causing water quality impairments, Congress included Section 402(p) in the CWA; this section established a comprehensive, two-phase approach to stormwater control using the NPDES program. In 1990 EPA issued the Phase I Stormwater Rule (55 Fed. Reg. 47990; November 16, 1990) requiring NPDES permits for operators of municipal separate storm sewer systems (MS4s) serving populations over 100,000 and for runoff associated with industrial activity, in- cluding runoff from construction sites five acres and larger. In 1999 EPA issued the Phase II Stormwater Rule (64 Fed. Reg. 68722; December 8, 1999), which expanded the requirements to small MS4s in urban areas and to construction sites between one and five acres in size. Since EPAâs stormwater program came into being, several problems inher- ent in its design and implementation have become apparent. As discussed in more detail in Chapter 2, problems stem to a large extent from the diffuse nature
36 URBAN STORMWATER MANAGEMENT IN THE UNITED STATES of stormwater discharges combined with a regulatory process that was created for point sources (the NPDES permitting approach). These problems are com- pounded by the shear number of entities requiring oversight. Although exact numbers are not available, EPA estimates that the number of regulated MS4s is about 7,000, including 1,000 Phase I municipalities and 6,000 from Phase II. The number of industrial permittees is thought to be around 100,000. Each year, the construction permit covers around 200,000 permittees each for both Phase I (five acres or greater) and Phase II (one to five acres) projects. Thus, the total number of permittees under the stormwater program at any time numbers greater than half a million. There are fewer than 100,000 non-stormwater (meaning wastewater) permittees covered by the NPDES program, such that stormwater permittees account for approximately 80 percent of NPDES-regulated entities. To manage this large number of permittees, the stormwater program relies heav- ily on the use of general permits to control industrial, construction, and Phase II MS4 discharges, which are usually statewide, one-size-fits-all permits in which general provisions are stipulated. An example of the burden felt by a single state is provided by Michigan (David Drullinger, Michigan Department of Environmental Quality Water Bu- reau, personal communication, September 2007). The Phase I Stormwater regu- lations that became effective in 1990 regulate 3,400 industrial sites, 765 con- struction sites per year, and five large cities in Michigan. The Phase II regula- tions, effective since 1999, have extended the requirements to 7,000 construction sites per year and 550 new jurisdictions, which are comprised of about 350 âprimary jurisdictionsâ (cities, villages, and townships) and 200 ânested juris- dictionsâ (county drains, road agencies, and public schools). Often, only a hand- ful of state employees are allocated to administer the entire program (see the survey in Appendix C). In order to comply with the CWA regulations, permittees must fulfill a number of requirements, including the creation and implementation of a storm- water pollution prevention plan, and in some cases, monitoring of stormwater discharges. Stormwater pollution prevention plans document the stormwater control measures (SCMs; sometimes known as best management practices or BMPs) that will be used to prevent or slow stormwater from quickly reaching nearby waterbodies and degrading their quality. These include structural meth- ods such as detention ponds and nonstructural methods such as designing new development to reduce the percentage of impervious surfaces. Unfortunately, data on the degree of pollutant reduction that can be assigned to a particular SCM are only now becoming available (see Chapter 5). Other sources of variability in EPAâs stormwater program are that (1) there are three permit types (municipal, industrial, and construction), (2) some states and local governments have assumed primacy for the program from EPA while others have not, and state effluent limits or benchmarks for stormwater dis- charges may differ from the federal requirements, and (3) whether there are monitoring requirements varies depending on the regulating entity and the type of activity. For industrial stormwater there are 29 sectors of industrial activity
INTRODUCTION 37 covered by the general permit, each of which is characterized by a different suite of possible contaminants and SCMs. Because of the industry-, site-, and community-specific nature of stormwa- ter pollution prevention plans, and because of the lack of resources of most NPDES permitting authorities to review these plans and conduct regular compli- ance inspections, water quality-related accountability in the stormwater program is poor. Monitoring data are minimal for most permittees, despite the fact that they are often the only indicators of whether an adequate stormwater program is being implemented. At the present time, available monitoring data indicate that many industrial facilities routinely exceed âbenchmark valuesâ established by EPA or the states, although it is not clear whether these exceedances provide useful indicators of stormwater pollution prevention plan inadequacies or poten- tial water quality problems. These uncertainties have led to mounting and con- tradictory pressure from permittees to eliminate monitoring requirements en- tirely as well as from those hoping for greater monitoring requirements to better understand the true nature of stormwater discharges and their impact. To improve the accountability of it Stormwater Program, EPA requested ad- vice on stormwater issues from the National Research Councilâs (NRCâs) Water Science and Technology Board as the next round of general permits is being prepared. Although the drivers for this study have been in the industrial storm- water arena, this study considered all entities regulated under the NPDES pro- gram (municipal, industrial, and construction). The following statement of task guided the work of the committee: (1) Clarify the mechanisms by which pollutants in stormwater discharges affect ambient water quality criteria and define the elements of a âprotocolâ to link pollutants in stormwater discharges to ambient water quality criteria. (2) Consider how useful monitoring is for both determining the potential of a discharge to contribute to a water quality standards violation and for determin- ing the adequacy of stormwater pollution prevention plans. What specific pa- rameters should be monitored and when and where? What effluent limits and benchmarks are needed to ensure that the discharge does not cause or contribute to a water quality standards violation? (3) Assess and evaluate the relationship between different levels of storm- water pollution prevention plan implementation and in-stream water quality, considering a broad suite of SCMs. (4) Make recommendations for how to best stipulate provisions in storm- water permits to ensure that discharges will not cause or contribute to ex- ceedances of water quality standards. This should be done in the context of gen- eral permits. As a part of this task, the committee will consider currently avail- able information on permit and program compliance.
38 URBAN STORMWATER MANAGEMENT IN THE UNITED STATES (5) Assess the design of the stormwater permitting program implemented under the CWA. The report is intended to inform decision makers within EPA, affected indus- tries, public stormwater utilities, other government agencies and the private sec- tor about potential options for managing stormwater. EPA requested that the study be limited to those issues that fall under the agencyâs current regulatory scheme for stormwater, which excludes nonpoint sources of pollution such as agricultural runoff and septic systems. Thus, these sources are not extensively covered in this report. The reader is referred to NRC (2000, 2005) for more detailed information on the contribution of agricultural runoff and septic systems to waterbody impairment and on innovative technolo- gies for treating these sources. Also at the request of EPA, concentrated animal feeding operations and combined sewer overflows were not a primary focus. However, the committee felt that in order to be most useful it should opine on certain critical effects of regulated stormwater beyond the delivery of traditional pollutants. Thus, changes in stream flow, streambank erosion, and habitat altera- tions caused by stormwater are considered, despite the relative inattention given to them in current regulations. Chapter 2 presents the regulatory history of stormwater control in the United States, focusing on relevant portions of the CWA and the regulations that have been created to implement the Act. Federal, state, and local programs for or affecting stormwater management are described and critiqued. Chapter 3 deals with the first item in the statement of task. It reviews the scientific aspects of stormwater, including sources of pollutants in stormwater, how stormwater moves across the land surface, and its impacts on receiving waters. It reflects the best of currently available science, and addresses biological endpoints that go far beyond ambient water quality criteria. Methods for monitoring and mod- eling stormwater (the subject of the second item in the statement of task) are described in Chapter 4. The material evaluates the usefulness of current bench- mark and MS4 monitoring requirements, and suggestions for improvement are made. The latter half of the chapter considers the multitude of models available for linking stormwater discharges to ambient water quality. This analysis makes it clear that stormwater pollution cannot yet be treated as a deterministic system (in which the contribution of individual dischargers to a waterbody impairment can be identified) without significantly greater investment in model develop- ment. Addressing primarily the third item in the statement of task, Chapter 5 considers the vast suite of both structural and nonstructural measures designed to control stormwater and reduce its pollutant loading to waterbodies. It also takes on relevant larger-scale concepts, such as the benefit of stormwater man- agement within a watershed framework. In Chapter 6, the limitations and possi- bilities associated with a new regulatory approach are explored, as are those of an enhanced but more traditional scheme. Numerous suggestions for improving the stormwater permitting process for municipalities, industrial sites, and con-
INTRODUCTION 39 struction are made. Along with Chapter 2, this chapter addresses the final two items in the committeeâs statement of task. REFERENCES Alberti, M., D. B. Booth, K. Hill, B. Coburn, C. Avolio, S. Coe, and D. Spiran- delli. 2006. The impact of urban patterns on aquatic ecosystems: an em- pirical analysis in Puget lowland sub-basins. Landscape Urban Planning, doi:10.1016/j.landurbplan.2006.08.001. Allan, J. D. 2004. Landscapes and riverscapes: the influence of land use on stream ecosystems. Annual Review of Ecology, Evolution, and Systematics 35:257â284. Alley, W. A., and J. E. Veenhuis. 1983. Effective impervious area in urban run- off modeling. Journal of Hydrological Engineering ASCE 109(2):313â319. Anderson, B .C., W. E. Watt, and J. Marsalek. 2002. Critical issues for storm- water ponds: learning from a decade of research. Water Science and Tech- nology 45(9):277â283. Barbour, M. T., M. J. Paul, D. W. Bressler, A. H. Purcell, V. H. Resh, and E. T. Rankin. 2006. Bioassessment: a tool for managing aquatic life uses for ur- ban streams. Water Environment Research Foundation Research Digest 01- WSM-3. Bay, S., B. H. Jones, K. Schiff, and L. Washburn. 2003. Water quality impacts of stormwater discharges to Santa Monica Bay . Marine Environmental Re- search 56:205â223. Bledsoe, B. P., and C. C. Watson. 2001. Effects of urbanization on channel in- stability. Journal of the American Water Resources Association 37(2):255â 270. Bloom, M. F. 2006. Low Impact Development approach slows down drainage, reduces pollution. Water and Wastewater International 21(4):59. Booth, D. B. 1990. Stream channel incision in response following drainage basin urbanization. Water Resources Bulletin 26:407â417. Booth, D. B. 2005. Challenges and prospects for restoring urban streams: a perspective from the Pacific Northwest of North America. Journal of the North American Benthological Society 24(3):724â737. Booth, D. B., and C. R. Jackson. 1997. Urbanization of aquatic systemsâ degradation thresholds, stormwater detention, and the limits of mitigation. Water Resources Bulletin 33:1077 1090. Booth, D. B., and P. C. Henshaw. 2001. Rates of channel erosion in small urban streams. Pp. 17â38 In: Land Use and Watersheds: Human Influence on Hy- drology and Geomorphology in Urban and Forest Areas. M. Wigmosta and S. Burges (eds.). AGU Monograph Series, Water Science and Application, Volume 2. Boulton, A. J. 1999. An overview of river health assessment: philosophies, practice, problems and prognosis. Freshwater Biology 41(2):469â479.
40 URBAN STORMWATER MANAGEMENT IN THE UNITED STATES Brookings Institute. 2004. Toward a new metropolis: the opportunity to rebuild America. Arthur C. Nelson, Virginia Polytechnic Institute and State Uni- versity. Discussion paper prepared for The Brookings Institution Metropoli- tan Policy Program. Burges, S. J., M. S. Wigmosta, and J. M. Meena. 1998. Hydrological effects of land-use change in a zero-order catchment. Journal of Hydrological Engi- neering 3:86â97. Calderon, R., E. Mood, and A. Dufour. 1991. Health effects of swimmers and nonpoint sources of contaminated water. International Journal of Environ- mental Health Research 1:21â31. Cassin, J., R. Fuerstenberg, F. Kristanovich, L. Tear, and K. Whiting. 2004. Application of normative flow on small streams in Washington Stateâ hydrologic perspective. Pp. 4281â4299 In: Proceedings of the 2004 World Water and Environmental Resources Congress: Critical Transitions in Water and Environmental Resources Management. Center for Watershed Protection (CWP). 2005. An integrated framework to restore small urban watersheds. Ellicott City, MD: CWP. Available at http://www.cwp.org/Store/usrm.htm. Last accessed September 23, 2008. Chang, H. 2007. Comparative streamflow characteristics in urbanizing basins in the Portland Metropolitan Area, Oregon, USA. Hydrological Processes 21(2):211â222. Chang, H., and T. N. Carlson. 2005. Water quality during winter storm events in Spring Creek, Pennsylvania USA. Hydrobiologia 544(1):321â332. Chang, H. H. 1992. Fluvial Processes in River Engineering. Malabar, FL: Krieger Publishing. Colford, J. M., Jr, T. J. Wade, K. C. Schiff, C. C. Wright, J. F. Griffith, S. K. Sandhu, S. Burns, J. Hayes, M. Sobsey, G. Lovelace, and S. Weisberg. 2007. Water quality indicators and the risk of illness at non-point source beaches in Mission Bay, California. Epidemiology 18(1):27â35. Dahl, T. 2006. Status and trends of wetlands in the conterminous United States: 1998â2004. Washington, DC: U.S. Department of the Interior Fish and Wildlife Service. Davies, S. P., and S. K. Jackson. 2006. The biological condition gradient: a descriptive model for interpreting change in aquatic ecosystems. Ecological Applications 16(4):1251â1266. Dinius, S. H. 1987. Design of an index of water quality. Water Resources Bul- letin 23(5):833â843. Duke, L. D., K. P. Coleman, and B. Masek. 1999. Widespread failure to comply with U.S. storm water regulations for industryâPart I: Publicly available data to estimate number of potentially regulated facilities. Environmental Engi- neering Science 16(4):229â247. Dunne, T., and L. B. Leopold. 1978. Water in Environmental Planning. New York: W. H. Freeman.
INTRODUCTION 41 Dwight, R. H., D. B. Baker, J. C. Semenza, and B. H. Olson. 2004. Health ef- fects associated with recreational coastal water use: urban vs. rural Califor- nia. American Journal of Public Health 94(4):565â567. EPA (U.S. Environmental Protection Agency). 1992. Storm Water Management for Industrial Activities, Developing Pollution Prevention Plans and Best Management Practices. Available at http://www.ntis.gov. EPA. 1998. EPA Project Beach. Washington, DC: EPA Office of Water. EPA. 2000. National Water Quality Inventory. 305(b) List. Washington, DC: EPA Office of Water. EPA. 2002. 2000 National Water Quality Inventory. EPA-841-R-02-001. Washington, DC: EPA Office of Water. EPA. 2006. Wadeable Streams Assessment: A Collaborative Survey of the Na- tionâs Streams. EPA 841-B-06-002. Washington, DC: EPA Office of Water. EPA. 2007. National Estuary Program Coastal Condition Report. EPA-842-B- 06-001. Washington, DC: EPA Office of Water and Office of Research and Development. FEMA (Federal Emergency Management Agency). No date. Flood. A report of the Subcommittee on Disaster Reduction. Available at http://www.sdr.gov. Last accessed September 23, 3008. Field, R., and R. E. Pitt. 1990. Urban storm-induced discharge impacts: U.S. Envi- ronmental Protection Agency research program review. Water Science and Technology 22(10â11):1â7. Gergel, S. E., M. G. Turner, J. R. Miller, J. M. Melack, and E. H. Stanley. 2002. Landscape indicators of human impacts to riverine systems. Aquatic Sciences 64(2):118â128. Gregory, M., J. Aldrich, A. Holtshouse, and K. Dreyfuss-Wells. 2005. Evalua- tion of imperviousness impacts in large, developing watersheds. Pp. 115â 150 In: Efficient Modeling for Urban Water Systems, Monograph 14. W. James, E. A. McBean, R. E. Pitt, and S. J. Wright (eds.). Guelph, Ontario, Canada: CHI. GrÃ¼bler, A. 1994. Technology. Pp. 287â328 In: Changes in Land Use and Land Cover: A Global Perspective. W. B. Meyer and B. L. Turner II (eds.). Cambridge: Cambridge University Press. Hamilton, P. A., T. L. Miller, and D. N. Myers. 2004. Water Quality in the Na- tionâs Streams and AquifersâOverview of Selected Findings, 1991â2001. U.S. Geological Survey Circular 1265. Available at http://pubs.usgs.gov/ circ/2004/1265/pdf/circular1265.pdf. Last accessed September 23, 2008. Hammer, T. R. 1972. Stream and channel enlargement due to urbanization. Water Resources Research 8:1530â1540. Hart, J. F. 1968. Loss and abandonment of cleared farm land in the Eastern United States. Annals of the Association of American Geographers 58(5):417â440. Heinz Center. 2002. The State of the Nationâs Ecosystems. Measuring the Lands, Waters, and Living Resources of the United States. Cambridge: Cambridge University Press.
42 URBAN STORMWATER MANAGEMENT IN THE UNITED STATES Hollis, G. E. 1975. The effect of urbanization on floods of different recurrence interval. Water Resources Research 11:431â435. Hollis, G. E., and J. K. Luckett. 1976. The response of natural river channels to urbanization: two case studies from southeast England. Journal of Hydrology 30:351â363. Holman-Dodds, J. K., A. A. Bradley, and K. W. Potter. 2003. Evaluation of hydrologic benefits of infiltration based urban storm water management. Journal of the American Water Resources Association 39(1):205â215. Horner, R. R., D. B. Booth, A. A. Azous, and C. W. May. 1997. Watershed de- terminants of ecosystem functioning. Pp. 251-274 In: Effects of Watershed Development and Management on Aquatic Ecosystems. L. A. Roesner (ed.). Proceedings of the Engineering Foundation Conference, Snowbird, UT, Au- gust 4â9, 1996. House, M. A., J. B. Ellis, E. E. Herricks, T. Hvitved-Jacobsen, J. Seager, L. Lijklema, H. Aalderink, and I. T. Clifforde. 1993. Urban drainage: Impacts on receiving water quality. Water Science and Technology 27(12):117â158. Karr, J. R. 1991. Biological integrity: a long-neglected aspect of water resource management. Ecological Applications 1:66â84. Karr, J. R., and C. O. Yoder. 2004. Biological assessment and criteria improve TMDL planning and decision making. Journal of Environmental Engineer- ing 130:594â604. Klein, R. D. 1979. Urbanization and stream quality impairment. Water Re- sources Bulletin 15:948â969. Kloss, C., and C. Calarusse. 2006. Rooftops to riversâgreen strategies for con- trolling stormwater and combined sewer overflows. New York: National Resources Defense Council. Available at http://www.nrdc.org/water/pollu- tion/rooftops/rooftops.pdf. Last accessed September 23, 2008. Konrad, C. P., and S. J. Burges. 2001. Hydrologic mitigation using on-site resi- dential storm-water detention. Journal of Water Resources Planning and Management 127:99 107. Konrad, C. P., D. B. Booth, and S. J. Burges. 2005. Effects of urban develop- ment in the Puget Lowland, Washington, on interannual streamflow pat- terns: consequences for channel form and streambed disturbance. Water Resources Research 41(7):1â15. Laenen, A. 1983. Storm runoff as related to urbanization based on data col- lected in Salem and Portland, and generalized for the Willamette Valley, Oregon. U.S. Geological Survey Water-Resources Investigations Report 83-4238, 9 pp. Lambin, E. F., B. L. Turner, H. J. Geist, S. B. Agbola, A. Angelsen, J. W. Bruce, O. T. Coomes, R. Dirzo, G. Fischer, C. Folke, P. S. George, K. Homewood, J. Imbernon, R. Leemans, X. Li, E. F. Moran, M. Mortimore, P. S. Rama- krishnan, J. F. Richards, H. Skanes, W. Steffen, G. D. Stone, U. Svedin, T. A. Veldkamp, C. Vogel, and J. Xu. 2001. The causes of land-use and land- cover change: moving beyond the myths. Global Environmental Change 11(4):261â269.
INTRODUCTION 43 LeBlanc, R. T., R. D. Brown, and J. E. FitzGibbon. 1997. Modeling the effects of land use change on the water temperature in unregulated urban streams. Journal of Environmental Management 49(4):445â469. Leopold, L. B. 1968. Hydrology for urban land planning: a guidebook on the hydrologic effects of urban land use. U.S. Geological Survey Circular 554. Washington, DC: USGS. Levin, P. S., R. W. Zabel, and J. G. Williams. 2001. The road to extinction is paved with good intentions: negative association of fish hatcheries with threatened salmon. Proceedings of the Royal SocietyâBiological Sciences (Series B) 268(1472):1153â1158. Lieb, D. A., and R. F. Carline. 2000. Effects of urban runoff from a detention pond on water quality, temperature and caged gammarus minus (say) (am- phipoda) in a headwater stream. Hydrobiologia 441:107â116. Loveland, T., and R. Auch. 2004. The changing landscape of the eastern United States. Washington, DC: U.S. Geological Survey. Available at http://www.usgs.gov/125/articles/eastern_us.html. Last accessed November 25, 2007. Maxted, J. R., and E. Shaver. 1997. The use of retention basins to mitigate stormwater impacts on aquatic life. Pp. 494-512 In: Effects of Watershed Development and Management on Aquatic Ecosystems. L. A. Roesner (Ed.). New York: American Society of Civil Engineers. Maxted, J. R., and E. Shaver. 1999. The use of detention basins to mitigate stormwater impacts to aquatic life. Pp. 6â15 In: National Conference on Retrofit Opportunities for Water Resource Protection in Urban Environ- ments, Chicago, February 9â12, 1998. EPA/625/R-99/002. Washington, DC: EPA Office of Research and Development. May, C. W., and R. R. Horner. 2002. The limitations of mitigation-based stormwater management in the pacific northwest and the potential of a con- servation strategy based on low-impact development principles. Pp. 1-16 In: Global Solutions for Urban Drainage. Proceedings of the Ninth Inter- national Conference on Urban Drainage. Meyer, J. L., M. J. Paul, and W. K. Taulbee. 2005. Stream ecosystem function in urbanizing landscapes. Journal of the North American Benthological So- ciety 24:602â612. Miltner, R. J., White, D., and C. O. Yoder. 2004. The biotic integrity of streams in urban and suburbanizing landscapes. Landscape and Urban Planning 69:87â100. Morisawa, M., and E. LaFlure. 1982. Hydraulic geometry, stream equilibrium and urbanization. Pp. 333â350 In: Adjustments of the Fluvial System. D. D. Rhodes and G. P. Williams (eds.). London: Allen and Unwin. Moscrip, A. L., and D. R. Montgomery. 1997. Urbanization, flood frequency, and salmon abundance in Puget Lowland streams. Journal of the American Water Resources Association 33:1289â1297. Naiman, R. J., and M. G. Turner. 2000. A future perspective on North America's freshwater ecosystems. Ecological Applications 10(4):958â970.
44 URBAN STORMWATER MANAGEMENT IN THE UNITED STATES Neller, R. J. 1988. A comparison of channel erosion in small urban and rural catchments, Armidale, New South Wales. Earth Surface Processes and Landforms 13:1â7. Nelson, E. J., and D. B. Booth. 2002. Sediment budget of a mixed-land use, urbanizing watershed. Journal of Hydrology 264:51â68. Nehrke, S. M., and L. A. Roesner. 2004. Effects of design practice for flood control and best management practices on the flow-frequency curve. Jour- nal of Water Resources Planning and Management 130(2):131-139. NOAA (National Oceanic and Atmospheric Administration). 1992. 1990 Shell- fish Register of Classified Estuarine Waters. Data supplement. Rockville, MD: National Ocean Service. NRC (National Research Council). 2000. Watershed Management for Potable Water Supply: Assessing the New York City Strategy. Washington, DC: Na- tional Academies Press. NRC. 2005. Regional Cooperation for Water Quality Improvement in South- western Pennsylvania. Washington, DC: National Academies Press. NRDC (Natural Resources Defense Council). 2007. Testing the Waters: A Guide to Water Quality at Vacation Beaches (17th ed.). New York: NRDC. Palmer, M. A., and J. D. Allan. 2006. Restoring Rivers. Issues in Science & Technology. Washington, DC: National Academies Press. Paul, M. J., and J. L. Meyer. 2001. Streams in the urban landscape. Annual Review of Ecology and Systematics 32:333â365. Poff, N. L., J. D. Allan, M. B. Bain, J. R. Karr, K. L. Prestegaard, B. D. Richter, R. E. Sparks, and J. C. Stromberg. 1997. The natural flow regime: a para- digm for river conservation and restoration. BioScience 47(11):769â784. Poff, N. L., B. P. Bledsoe, and C. O. Cuhaciyan. 2006. Hydrologic variation with land use across the contiguous United States: geomorphic and ecologi- cal consequences for stream ecosystems. Geomorphology 79 (3â4):264â 285. Pozzi, F., and C. Small. 2005. Analysis of urban land cover and population density in the United States. Photogrammetric Engineering and Remote Sensing 71(6):719â726. Prysch, E. A., and J. C. Ebbert. 1986. Quantity and quality of storm runoff from three urban catchments in Bellevue, Washington. USGS Water- Resources Investigations Report 86-4000, 85 pp. Puget Sound Action Team. 2005. Low Impact Development: Technical Guid- ance Manual for Puget Sound. Available at http://www.psat.wa.gov/Pro- grams/LID.htm. Last accessed September 23, 2008. Raskin, L., A. DePaoli, and M. J. Singer. 2005. Erosion control materials used on construction sites in California. Journal of Soil and Water Conservation 60(4):187â192. Roesner, L. A., B. P. Bledsoe, and R. W. Brashear. 2001. Are best-management- practice criteria really environmentally friendly? Journal of Water Re- sources Planning and Management 127(3):150-154.
INTRODUCTION 45 Roy, A. H., M. C. Freeman, B. J. Freeman, S. J. Wenger, W. E. Ensign, and J. L. Meyer. 2005. Investigating hydrological alteration as a mechanism of fish assemblage shifts in urbanizing streams. Journal of the North American Benthological Society 24:656â678. Schueler, T. 1987. Controlling Urban Runoff: A Practical Manual for Planning and Designing Urban Best Management Practices. Washington, DC: Met- ropolitan Washington Council of Governments. Simon, A. 1989. A model of channel response in disturbed alluvial channels. Earth Surface Processes and Landforms 14:11â26. Steedman, R. J. 1988. Modification and assessment of an index of biotic integrity to quantify stream quality in Southern Ontario. Canadian Journal of Fisheries and Aquatic Sciences 45:492â501. Strecker, E. W., M. M. Quigley, B. Urbonas, and J. Jones. 2004. Analyses of the expanded EPA/ASCE International BMP Database and potential impli- cations for BMP design. In: Proceedings of the World Water and Environ- mental Congress 2004, June 27âJuly 1, 2004, Salt Lake City, UT. G. Sehlke, D. F. Hayes, and D. K. Stevens (eds.). Reston, VA: ASCE. Sutherland, R. 1995. Methods for estimating the effective impervious area of urban watersheds. Watershed Protection Techniques 2(1):282-284. Ellicott City, MD: Center for Watershed Protection. Thomson, J. D., G. Weiblen, B. A. Thomson, S. Alfaro, and P. Legendre. 1986. Untangling multiple factors in spatial distributions: lilies, gophers, and rocks. Ecology 77:1698â1715. Trimble, S. W. 1997. Contribution of stream channel erosion to sediment yield from an urbanizing watershed. Science 278:1442â1444. Trout Unlimited. 2006. Eastern Brook Trout: Status and Threats. Eastern Brook Trout Joint Venture. Arlington, VA: Trout Unlimited. USGS (U.S. Geological Survey). 2006. The quality of our nationâs waters: pes- ticides in the nationâs streams and ground water: 1992â2001. National Wa- ter Quality Assessment Program. USGS Circular 1291. Reston, VA: USGS. van Roon, M. 2007. Water localisation and reclamation: steps towards low im- pact urban design and development. Journal of Environmental Manage- ment 83(4):437â447. Vitousek, P. M., H. A. Mooney, J. Lubchenco, and J. M. Melillo. 1997. Human domination of Earth's ecosystems. Science 277(5325):494â499. Walsh, C. J., T. D. Fletcher, and A. R. Ladson. 2005a. Stream restoration in urban catchments through redesigning stormwater systems: looking to the catchment to save the stream. Journal of the North American Benthological Society 24:690â705. Walsh, C. J., A. H. Roy, J. W. Feminella, P. D. Cottingham, P. M. Groffman, and R. P. Morgan. 2005b. The urban stream syndrome: current knowledge and the search for a cure. Journal of the North American Benthological Society 24(3):706â723.
46 URBAN STORMWATER MANAGEMENT IN THE UNITED STATES Wang, L., J. Lyons, P. Kanehl, and R. Gatti. 1997. Influences of watershed land use on habitat quality and biotic integrity in Wisconsin streams. Fisheries 22(6):6â12. Wang, L., J. Lyons, P. Kanehl, and R. Bannerman. 2001. Impacts of urbaniza- tion on stream habitat and fish across multiple spatial scales. Environ- mental Management 28(2):255â266. Wang, L., J. Lyons, and P. Kanehl. 2003. Impacts of urban land cover on trout streams in Wisconsin and Minnesota. Transactions of the American Fisher- ies Society 132(5):825â839. White, R. J. 1996. Growth and development of North American stream habitat management for fish. Canadian Journal of Fisheries and Aquatic Sciences 53(Suppl 1):342â363. Whitlow, J. R., and K. J. Gregory. 1989. Changes in urban stream channels in Zimbabwe. Regulated Rivers: Research and Management 4:27â42. Williams, G. P., and M. G. Wolman. 1984. Downstream Effects of Dams on Alluvial Rivers. U.S. Geological Survey Professional Paper 1286. Wolman, M. G., and Schick, A. 1967. Effects of construction on fluvial sediment, urban and suburban areas of Maryland. Water Resources Research 3:451â464. Wright, T., J. Tomlinson, T. Schueler, and K. Cappiella. 2006. Direct and indi- rect impacts of urbanization on wetland quality. Wetlands and Watersheds Article 1. Ellicott City, MD: Center for Watershed Protection.