As an industry, we’ve acquired a vast amount of knowledge about stormwater, its adverse impacts, and the best management practices (BMPs) implemented to mitigate them. However, the spirited debate with regard to whether the field or the laboratory is the best arena for evaluating BMP performance refuses to yield to consensus. Here are some of the arguments for and against laboratory testing.
Why Laboratory Testing is Preferred
Laboratory testing allows for relatively quick evaluations of BMP performance across a wide range of operating conditions with increased accuracy and at less cost. Variables can be isolated and controlled allowing numerous pollutant loading scenarios to be simulated during a single set of trials.
Laboratory testing is also the ideal vehicle for making side by side BMPs comparisons since it is repeatable. Flow based BMPs have proven particularly difficult to compare since sizing is adjusted based on the required treatment rate. Standardized laboratory testing allows for the creation of a simple BMP comparison matrix.
Laboratory Testing Pitfalls
Typical Particle Size Distribution
Critics of laboratory testing assert that the complex nature of stormwater runoff is impossible to synthesize in a laboratory (Bryant 1999). The highly dynamic nature of stormwater runoff retards our ability to replicate field conditions. Variables such as particle size distribution, rainfall intensity/runoff rate, and pollutant concentration are in a constant state of flux in the field. To conduct laboratory trials we attempt to estimate “typical” conditions and hold them relatively constant in order to evaluate how the system performs given a specific set of circumstances. If assumptions made in the laboratory grossly misrepresent field conditions then the results are not likely to be a good predictor of field performance.
One of the most widely debated issues relative to laboratory testing is what constitutes a representative sediment particle size distribution (PSD). In the field, PSD is constantly changing and largely a function of rainfall intensity and site specific solids loading. As rainfall intensity increases larger particles are mobilized and during lower intensity events only finer particles tend to be mobilized. However in the laboratory it is common for a single PSD to be injected into the test system regardless of operating rate. This practice likely inflates performance estimates at lower operating rates as a result of the presence of coarse particles. Most laboratory studies are conducted using sediment such as crushed silica that is of uniform density/specific gravity, which also oversimplifies real world conditions. A significant fraction of the sediment load transported by stormwater runoff is often made up of organic matter and other material with lower settling velocities than silica particles.
A typical stormwater laboratory study involves stabilizing flow at a target rate and then collecting a series of samples to assess performance at that rate. The process is then repeated for each flow rate to be studied. Ideally the target flow rates should span the full range of the BMP’s operating capacity. This approach is an obvious departure from typical field conditions where inflow rates are in a constant state of flux during any given storm event. Testing under stable flow conditions likely over simplifies pollutant transport and delivery mechanisms as compared with field conditions, but the potential impact on BMP performance is not well understood.
During laboratory studies it is common to isolate a single pollutant, most often solids, without regard for how the presence of other pollutants may alter performance. It’s difficult to know precisely what the impact of excluding trash, debris, oil, grease and other pollutants from laboratory trials is on solids retention, but it is unquestionably a simplification of field conditions. Large volumes of trash and debris can alter flow patterns and potentially contribute to trapping solids via crude filtration. Oils and greases can coat solids and other materials present in runoff changing their settling characteristics and creating opportunities for particles to stick together. More complex pollutants such as nutrients and metals that exist in both dissolved and solid phases, and depending on water chemistry may transition between the two, are predominantly excluded from laboratory trials.
The majority of laboratory studies conducted to date neglected to investigate BMP longevity. Focus is placed exclusively on pollutant removal without concern for long term operation or pollutant storage capacity. Many laboratory trials are conducted with the BMP in “like new” condition which is not representative of the field. The New Jersey Department of Environmental Protection made an effort to address this in their laboratory protocol by requiring that BMPs be filled to half of their sediment storage capacity before performance trials are conducted, but only a small number of studies currently comply. Failing to assess longevity is particularly concerning for filtration BMPs. Filters typically perform very well, but can rapidly occlude with solids and other pollutants rendering them inoperable.