Development of a bioretention cell model and evaluation of input specificity on model accuracy

With the implementation of Phase II of the National Pollutant Discharge Elimination System (NPDES), municipalities have new requirements to reduce stormwater quantity and enhance water quality. Bioretention cells (BRCs) are a pollution mitigation option that can address the new regulations. In order to implement BRCs in the landscape, models are needed so stormwater engineers and managers can estimate the impact of the mitigation technique. While several BRC models are available, users must supply input parameters, which are many times poorly understood. The objective of this work was to determine the level of input specificity in hydraulic parameters needed to accurately estimate water movement through a BRC. A water movement model was developed that incorporates infiltration, drainage, and overflow for a single storm event. Then pilot-scale BRCs were constructed and operated to obtain data for model testing. The model was run with four sets of input parameters with increasing specificity: soil type, fraction sand/silt/clay, an adjustment for bulk density, and a macropore routine to serve as a fitting parameter. While the model with the highest input specificity proved to match experimental values closest (drainage volume between 0.7% and 8.8% from observed, and maximum drainage flow rate between 1.4% and 18% from observed), it is unlikely that stormwater managers would have access or time to obtain this information. However, a simulation with the fraction sand/silt/clay and an adjustment for bulk density provided acceptable results (drainage volume between 0.7% and 18% from observed, and maximum drainage flow rate between 30% and 39% from observed).