The Case Against Representative Stormwater Runoff Monitoring

One often stated objective of stormwater runoff monitoring programs is to characterize the quality of stormwater runoff from one or more "representative" areas, that is, areas having predominantly residential, commercial, industrial, roadway, or open space land use. Representative urban runoff monitoring has been the focus of many sampling efforts over the past 30 years, including the Nationwide Urban Runoff Program (NURP) study. In the early 1990's, many large municipalities were required to conduct representative urban stormwater runoff monitoring in order to obtain a National Pollutant Discharge Elimination System (NPDES) stormwater discharge permit, and a number of NPDES permits now in effect require some form of representative urban stormwater runoff monitoring. Concentrations of pollutants in stormwater are notoriously variable, and in the case of representative runoff monitoring projects, a large number of sample data is required in order to perform any kind of in-depth statistical analyses. This makes representative urban stormwater runoff monitoring programs extraordinarily expensive, and this commitment of resources is not always—some say, not ever—rewarded with results that substantially support the needs of decision makers. The purpose of this paper is to put forward a case for discontinuing representative stormwater runoff monitoring. First, a simple method for estimating the costs involved in stormwater runoff monitoring is proposed. Next, an example of the type of analyses that can be performed using data already available is provided. Finally, using the cost-estimating model and the results of the statistical analysis, it is shown that an exorbitant amount of money would be required to further "refine" estimates of representative runoff concentrations.