Brock B. Bernstein

308 Raymond St., Ojai, CA 93023

Phone: (805) 646-3849

Michael Drennan

MDA, Inc., 17712 Miranda St., Encino, CA 91316

Introduction

Stormwater management and monitoring programs continue to undergo rapid evolution. In a relatively short period of time monitoring programs monitoring program they have progressed from an initial focus on characterizing the nature and extent of stormwater problems to comprehensive efforts to identify sources of stormwater pollution, determine the effectiveness of control measures, and track a myriad of management activities. In that same time period management programs have moved from BMP implementation based strictly on best professional judgment (some have characterized as a shotgun approach), to regionally integrated watershed management efforts based much more solidly on sound science. This accelerated evolution has resulted in constant pressure to develop and implement improved indicators and study designs that can provide the information needed by these developing management initiatives. We suggest that a continuation of the historical emphasis on demonstrating compliance as the primary goal of monitoring and management will be counterproductive in a number of ways. We suggest that as this evolution continues, stormwater managers will need to continually consider the proper balance of monitoring dollars allocated to traditional regulatory (or compliance) monitoring and nonregulatory performance monitoring.

Instead, we suggest that stormwater monitoring programs indicator selection be broadened to include measures indicators of success with stated environmental goals as well as of indicators of compliance with regulatory objectives. Success is a performance issue that measures protection and enhancement of environmental value. Compliance is a legal issue that reduces liability. Both are important. Including measures indicators of success in monitoring programs requires shifts in both methods and mindsets, with attendant shifts in the potential risks and rewards of monitoring and management actions. We demonstrate that the increased rewards typically outweigh any increased risks from broadening indicator selection.

A Brief History

Despite a recognition that contaminated stormwater runoff was an important urban pollution problem (e.g., Weibel et al. 1964), it received little regulatory and management attention throughout the 1960s and 1970s (Barton 1978). Attention instead focused on large and readily identifiable point sources such as municipal wastewater treatment plants and industrial discharges. The relative ease of identifying, treating, and monitoring point sources made this a logical management choice during the early years of intensive water pollution control efforts.

In recent years, significant improvements in point source discharges and a greater understanding of the nature and potential impacts of stormwater runoff (e.g., Hunter et al. 1979, Eganhouse and Kaplan 1981, Cole et al. 1984, Hoffman et al. 1983) have combined to raise the scientific and public profile of this issue. Thus, despite the problems attendant upon dealing with pollution from diffuse arrays of difficult to identify sources, the Water Quality Act of 1987 amended §402(p) of the Clean Water Act to require NPDES permits for stormwater discharges. Rather than identifying definite discharge limits for specific pollutants, stormwater permits require instead that pollutants be reduced to "the maximum extent practicable" (MEP). While this is difficult (some might say impossible) to define, the principal motivation for stormwater management and monitoring remained the need to demonstrate compliance with permit conditions. As a result, the vast majority of parameters measured in typical stormwater monitoring programs are either: 1) the concentrations and loadings of pollutants, or 2) the level of effort applied to various management actions (e.g. miles of street swept, or number of catch basins cleaned). As we argue below, this approach does fulfill one of the essential roles of compliance, protecting against traditional forms of liability resulting from exceeding discharge standards in NPDES permit conditions. It does not, however, demonstrate that beneficial uses are necessarily being protected nor that essential societal values are being enhanced.

Challenges at the Regional Scale

While the goals of the Clean Water Act (CWA) are straightforward ("fishable, swimmable waters", and "protection and restoration of the physical, chemical and biological qualities of the nation’s waters")prohibit discharges to stormwater and reduce stormwater contamination to the maximum extent practicable), achieving them in practice is challenging and complex. Even the more direct stormwater objectives (such as prohibiting discharges to stormwater and reducing stormwater pollution to the MEP) Stormwater runoff must necessarily be viewed at watershed scales. As in municipal wastewater systems, stormwater collects through a network of ever larger channels until it finally reaches a discharge point (i.e., outfall or creek/river mouth). Unlike wastewater systems, however, natural channels can support beneficial uses along their entire length. Thus, monitoring must not only account for a range of diffuse and poorly defined sources, it must also address potential impacts throughout the entire network of channels rather than around a single terminal discharge point.

In recognition of these and other ways in which stormwater systems differ from traditional large point sources, many municipalities have grouped themselves into larger stormwater agencies that encompass entire watersheds or sub-watersheds. This approach to organizing stormwater programs helps to coordinate similar activities (e.g., loadings estimation, trend monitoring) within a watershed. It leaves unresolved, however, the issue of coordinating the full range of agencies and activities that potentially affect both pollutant loads and beneficial uses within watersheds. Beneficial uses are often affected, for example, by channel construction and maintenance activities, land use patterns and changes in these, wildlife management practices, and regional air pollution control strategies, to name but a few. Recently increased attention to this broader array of impacts has put increased pressure on stormwater management programs to consider a wider set of activities than simply pollution control and monitoring (Environmental Statutes 1987, NRC 1990, U.S. EPA 1991).

Such developments complicate the search for reliable and information-rich indicators upon which to base monitoring programs. In regional contexts in general, and for stormwater programs in particular, indicator selection is beset by difficult challenges. Ecology has only recently begun to address processes at the landscape and regional scale, with the result that the development of reliable indicators of pattern and process at these scales is still in its early stages. In addition, larger ecological units such as watersheds are often strongly influenced by intermittent but intense events such as extremes of rainfall or temperature. Most monitoring and research datasets are not long enough to provide adequate insight into how watersheds respond to such events, which in turn hampers our ability to select robust indicators that will be useful over the long term.

At a more pragmatic level, historic data that can help establish reference conditions is often difficult to find and, once found, hard to integrate at the needed spatial scale. A recent National Academy of Sciences study (NRC 1995) documented a wide variety of incompatibilities that can impede attempts to create regionally consistent background datasets. Obtaining the agreement among relevant agencies needed to address these and other similar problems can be frustrating and time-consuming.

Redefining Risk and Defining What Counts

Overcoming these sorts of scientific, technical, and administrative barriers is crucial and we do not intend to minimize their importance in the least. However, focusing only on these constraints can cause us to overlook another set of equally important concerns. We thus believe monitoring design efforts could benefit from stepping back and viewing indicator selection in a larger context.

As mentioned above, most indicators selected for routine monitoring programs reflect straightforward regulatory compliance, i.e., "Are levels of constituent X decreasing?" or "Are streets being swept?" The assumption underlying this approach is that achieving this kind of compliance ensures that broader goals related to habitat and water quality protection and restoration are being met. That is, if pollutant levels decline, then water and habitat quality will necessarily improve, with attendant improvement in a wide range of beneficial uses. This sort of monitoring, and the approach to compliance it supports, certainly reduces the risk that programs will be accused of not meeting their permit requirements. Used properly, this kind of monitoring can also provide useful feedback about whether management programs are in fact reducing levels of pollutants.

While this focus on formal compliance reduces one kind of program risk or liability, it increases another and potentially more far-reaching kind. As mentioned above, beneficial uses can be degraded by a wide range of insults and activities. Some of these are unrelated to pollution but are directly related to the way stormwater is managed. By focusing monitoring narrowly on traditional compliance measures, we can lose sight of the core societal values that motivate environmental protection and regulation in the first place, making compliance instead an end in itself. This is problematic because environmental programs of all kinds throughout the U.S. are facing increasing pressure to show attention to and tangible improvements in these core values. These programs’ constituencies are therefore more and more likely to hold them accountable to a much broader definition of performance. An overly narrow focus on formal compliance runs the risk of breaking the connection to these core values with a resulting loss of support for current environmental management programs.

We therefore suggest that indicator selection be broadened to include measures of success as well as of compliance. These measures of success should be more closely tied to the actual beneficial uses of the receiving waters the regulations are intended to protect, rather than to the quality of a particular discharge to those waters. For example, if a creek has been designated to support specific beneficial uses such as cold water aquatic life, recreation, or drinking water supply, indicators such as the presence of trout, the availability of stream habitat, or the ability of the water to support wading and swimming can be defined. For example, indices of biologic integrity (IBIs) have been used to assess biological resources by integrating information about the elements of biological systems and the processes that generate and maintain them. The IBI evaluates the full spectrum of human disturbances (water quality, habitat structure, energy source, flow regime, and biotic interactions) in an ecological rather than a laboratory-based chemical toxicity context (Karr 1998).

As an example of this approach, the Rouge River National Wet Weather Demonstration Project in Michigan has identified a series of indicators for public use which convey clear, succinct information to non-technical users about the overall health of the aquatic resource. These indicators include public uses (fishing, wading/swimming, canoeing/boating, aesthetics) and river conditions (dissolved oxygen, river flow, bacteria, aquatic life [based on an IBI], and stream habitat [also based on a field method which evaluates several factors]). Each of these indicators was rated as good, fair, or poor and plotted on multi-color maps for the public’s use.

While there may be those who criticize this approach as too indirect, and not relevant to the narrow interests of compliance with stormwater regulations, the public is demanding that the resources spent on environmental protection be correlated with actual results. It is no longer acceptable to merely measure compliance with permit conditions and assume the goals of the CWA are satisfied. As a crude example, stormwater dischargers can currently demonstrate they are in compliance with their permit merely by documenting the amount of debris removed from their catch basins. Nothing requires them to demonstrate that the amount of debris removed is actually significant relative to the threat that debris poses to the receiving waters. While a discharger may be satisfied with mere compliance with permit conditions, the public only cares if the actual goals of the CWA are being reached.

Success is a performance issue that measures protection and enhancement of environmental value. Compliance is a legal issue that reduces liability. Both are important. However, including measures of success in monitoring programs will require shifts in both methods and mindsets. It demands a willingness to hold programs accountable to performance standards that are often harder to define and/or measure. It means asking agencies and their staff to look beyond their existing expertise with pollutants and engineering-oriented management practices to issues of habitat, ecology, land use, and broader public policy. Together these will result in changes to the existing calculus for estimating the potential risks and rewards of monitoring and management actions.

Programs that demonstrate a good-faith effort to address the concerns that matter most to their constituencies by protecting core environmental values build up a reservoir of good will that can provide an effective buffer against accusations that strict compliance standards are not being met. More importantly, they protect against the even more risky charge that programs are out of touch with what really matters and therefore not worth continuing.

References