Abstract
Florida regulations require stormwater runoff from residential developments to be captured in wet detention ponds for the purpose of removing nutrients in runoff prior discharging to waters of the U.S. The design and operation of the ponds, however, is weakly regulated: regulations are in place only for ponds’ dimensions at the time of construction, leaving an open question about the ponds’ effectiveness over time at their intended purpose. Regulations also do not require monitoring of pond discharges so there is no regulatory database about their effectiveness, and few data have been produced by independent studies of water quality in these systems – none of it over lengthy periods of time that would readily support assessment of the ponds’ effectiveness. This research used three kinds of data to assess to the extent possible the effectiveness of a sample of south Florida residential stormwater ponds, and equally importantly, to investigate the capability of available data to conduct such an assessment. The first was a regulatory analysis to critically evaluate the rules to which the ponds are subject; to determine if those rules effectively specify the initial conditions; and to determine if those rules are being followed in permits filed for ponds in the target region. The second was a field investigation of pond geometry (surface area, length/width ratios, and bathymetry) of a sample of ponds to determine whether ponds in the target region, as built, conform to the rules, which supports an assessment of the regulatory effectiveness as implemented. The third was an analysis of the best available data on historical trends in water quality, for the largest possible sample of ponds, to assess whether those ponds that conform to the dimensional requirements of regulations are in fact more successful at removing nutrients than ponds that do not. Taken together these three analyses assess the effectiveness of the sample of ponds at their stated purpose, to the best extent possible given important limitations in available information. The research encompassed a sample of 14 ponds, which constituted the largest number of ponds with a reasonably long historical record of water quality data: at least seven years’ reported data over the period 2000-2015 from a limited, volunteer-based monitoring program in Lee County FL. The first research question was whether the regulatory system has been effective in attaining compliance by this sample of ponds. The answer is that it has not. One reason is that a pond in any one location is subject to rules from multiple agencies, with overlapping jurisdictions, have specified inconsistent, sometimes conflicting, standards and regulatory requirements. Compliance was irregular, as 11 of 14 target ponds failed to complete one or more essential paperwork requirements. Further, 12 of the 14 ponds failed to meet the regulations’ stated requirements for pond dimensions: the bathymetry analysis revealed that more than half of the sample of ponds far exceeded the regulatory-maximum depth. As other results of this research suggest, pond depth may be important in the ponds’ performance, so this is a crucial failure. The second research question was whether the geometry of the ponds, as measured for this research, displays any statistical relationship with water quality. That relationship, if demonstrated, supports the regulatory presumption that specifying ponds’ dimensions is a valid surrogate for specifying the proportion of nutrients removed. Water quality constituents analyzed included four compounds of phosphorous; six compounds of nitrogen; and chlorophyll-a. This research question requires the imperfect assumption that ponds with the highest measured nutrient concentrations are those achieving the least nutrient removal. A number of confounders could interfere with this assumption, but if a correlation were identified in the face of these confounders it could be meaningful. Simple regressions and simple correlations were conducted for each of the 11 measured constituents against 16 geometric parameters, for each of three statistical analyses. The analyses produced a total of 528 simple regression/correlation results. The one-constituent-to-one-geometric-parameter analyses showed none of the geometric measures, by themselves, explained more than a small amount of variation in constituents: only 72 analyses (14% of those attempted) achieved correlation coefficients (R2) with absolute value greater than 0.16. , Those that did attain this very modest correlation were nearly all associated with some measure of pond depth, either average depth; maximum depth; or proportion of the pond deeper than the regulatory target. Even those findings are questionable because of those 72 relationships, only 32 showed statistical significance of p < 0.10, and only 16 showed significance of p < 0.05, a proportion that cannot be distinguished from random variation. The data do not show evidence that ponds’ geometry has statistically significant relationships with concentration of nutrients or chlorophyll-a, suggesting that a regulatory approach that specifies pond geometry may not succeed at creating ponds that effectively remove nutrients. The third research question was whether ponds whose dimensions are the farthest from the regulatory requirements are also those with the greatest nutrient concentration, and thus, by our assumption, those achieving the least nutrient removal. If that were true it would suggest that regulations on pond geometry might at least prohibit construction of the least-effective ponds, if those regulations were followed. Here also the answer was they do not. The correlations that failed to support the second research question also failed to support this third question, but direct inspection of the individual ponds conclusively showed they do not conform to this hypothesis. Ponds deeper than the regulatory requirement were by far the most frequent violation, and the data show that deeper ponds do not have the highest concentrations of nutrients or chlorophyll. The deepest ponds in the sample included both some ponds with the highest, and some ponds with the lowest, concentrations of each of the nitrogen, phosphorous, and chlorophyll constituents. The ponds with the highest concentration of each of the eleven analyzed constituents included some with the greatest, and some with the least, depth divergence from the regulatory requirement. Even in this small sample, the findings were sufficient to answer this research question in the negative. The research concluded the available data do not justify the assumption, enshrined in Florida’s regulations, that as-built pond dimensions currently specified lead to removal of nutrients from residential stormwater runoff, and a sizable proportion of this small sample behaved as if there is no such relationship.