Comparing SWMM LID Module Results to Field Studies' Data
Citation:
Simon, M. Comparing SWMM LID Module Results to Field Studies' Data. Department of Civil, Environmental, and Ocean Engineering Seminar, Hoboken, NJ, November 13, 2018.
Impact/Purpose:
This presentation is a undergraduate and graduate student seminar for the Stevens Institute of Technology, Department of Civil, Environmental and Ocean Engineering. Many urban infrastructure systems are operating beyond their original design life and will need to be upgraded, costing billions of dollars over many decades. Municipalities rely heavily on software such as the USEPA’s Storm Water Management Model (SWMM) to simulate stormwater and wastewater infrastructure performance as an enhanced decision-making tool. As such, the accuracy of SWMM should be rigorously analyzed to ensure adequate model performance. The main objective of this study was to quantify how accurately SWMM v5.1.10 simulates the hydraulic activity of previously monitored low impact development storm control measures. Model performance was evaluated by mathematically comparing empirical data to model results using a multi-event, multi-objective calibration method. The calibration methodology utilized PEST software, a parameter estimation tool, to determine unmeasured hydraulic parameters for SWMM’s low impact development modules. The calibrated LID modules had a range 0.65-0.94 Nash-Sutcliffe Efficiencies and 0.67-0.97 correlation ratios. This study led to several model improvements, and identified SWMM’s most severe limitation is that it does not model lateral exfiltration of water out of the storage layer of an infiltration trench. Nevertheless, SWMM can often successfully simulate eight of the nine LIDs’ performance that were considered, given accurate model configuration, parameter measurement, and model calibration to site-specific hydraulic data.
Description:
Many urban infrastructure systems are operating beyond their original design life and will need to be upgraded, costing billions of dollars over many decades. Municipalities rely heavily on software such as the USEPA’s Storm Water Management Model (SWMM) to simulate stormwater and wastewater infrastructure performance as an enhanced decision-making tool. As such, the accuracy of SWMM should be rigorously analyzed to ensure adequate model performance. The main objective of this study was to quantify how accurately SWMM v5.1.10 simulates the hydraulic activity of previously monitored low impact development storm control measures. Model performance was evaluated by mathematically comparing empirical data to model results using a multi-event, multi-objective calibration method. The calibration methodology utilized PEST software, a parameter estimation tool, to determine unmeasured hydraulic parameters for SWMM’s low impact development modules. The calibrated LID modules had a range 0.65-0.94 Nash-Sutcliffe Efficiencies and 0.67-0.97 correlation ratios. This study led to several model improvements, and identified SWMM’s most severe limitation is that it does not model lateral exfiltration of water out of the storage layer of an infiltration trench. Nevertheless, SWMM can often successfully simulate eight of the nine LIDs’ performance that were considered, given accurate model configuration, parameter measurement, and model calibration to site-specific hydraulic data.