Research Areas > Stormwater > BMP Modeling

Project: BMP Modeling

Background and Objectives

Structural Best Management Practices (BMPs) have become a common tool for stormwater managers to achieve water quality improvements and regulatory compliance. However, reliance on empirical (field-based) evaluation of BMP performance is costly and has limited ability to predict effectiveness under varying environmental conditions. In contrast, BMP models mathematically simulate the mechanisms of pollutant accumulation and removal. They can be used to simulate BMP performance under a variety of storm conditions and maintenance scenarios, as well as estimating annual pollutant removal, evaluating design options, and assessing performance of BMPs applied in series.

Low Impact Development (LID) has also been gaining popularity as a stormwater management approach. LID employs a combination of structural BMPs and alternative land use planning principles to manage pollutant-containing storm water and dry season runoff. Given the complexity and cost of evaluating potential LID scenarios, growth in this field creates an even greater need for modeling applications.

The main objectives of this project were (1) to apply an existing model for BMP/LID evaluation, and (2) to identify key data gaps that prevented optimization of the model's performance.
 

Status

This project was completed in 2005.

Methods

The Low Impact Development Management Practices Evaluation Computer Module (Tetra Tech, 2006) was used to simulate BMP performance over ten years of rainfall. This model uses land use runoff and loading (from HSPF or a similar model) as input, and then mechanistically simulates pollutant removal processes within the BMP. Existing BMP performance data from the International BMP database was used to calibrate the removal efficiency for the two modeled BMP types. The model was then used to evaluate the performance of the BMPs over a variety of climatic conditions. The model evaluated each BMP alone as well as in series, targeting volume, TSS, and total copper. BMP effectiveness was evaluated based on several measures of performance such as load reduction, event mean concentrations, frequency of exceedence of water quality standards, and peak flow reduction.

Findings

The model predicted over 60% removal of solids and copper throughout most conditions; however, effectiveness was reduced during large storms and wet years. The output for frequency of exceedence of water quality standards was also sensitive to storm size. BMP performance was comparable according to each of the separate measures of performance. This study demonstrates that BMP modeling can help managers understand expected BMP performance and, perhaps more significantly, evaluate BMPs in series, over a range of storms, time periods, and design parameters.

This study provided a comprehensive view of likely BMP performance over multiple storms and BMP configurations. One shortcoming of this effort was reliance on data collected outside of southern California, since there is a paucity of data detailing how stormwater pollutants levels are reduced within BMPs. For example, there are generally only event mean concentration data available for the influent and effluent. Knowing the influent and effluent pollutant concentrations throughout the storm would greatly enhance our understanding of BMP performance as well as the models’ predictive abilities. In addition, there is little information on the particle size distribution within the influent and effluent, which is significant since pollutants are often attached to those particles.

Future SCCWRP research efforts will focus on filling the data gaps noted above, and on modeling combinations of BMPs and LID strategies.