Low Impact Development micro-solutions increase infiltration into the ground thus maintaining the landscape’s natural runoff index. LID controls, which include a bio-retention cell option, have been implemented in USEPA’s Storm Water Management Model (SWMM) engine to aid in decisions regarding their implementation in various landscapes. Field data collected during 2019 from the monitoring of a bioretention basin pilot project in Montreal was used to verify the ability of the SWMM bio-retention cell module to accurately simulate outflow from basins with a trapezoidal cross-section. Two types of basins were modelled; one with an underdrain pipe and runoff storage potential (bio-retention) and the other one without (vegetated basin). Initially, the area of the bioretention basin was modelled as 2/3 of the field area to ensure the potential runoff storage volume was accurate, whereas the remaining 1/3 was as an impermeable surface. The vegetated basin area was modelled as identical to the field area thus prioritizing soil volume accuracy. An uncalibrated continuous model was generated in which all parameters were chosen using literature values, field data or calculations. The simulated results were acceptable for the vegetated basin (NSE = 0.68). As for the bioretention basin, the outflow volume was markedly overestimated, and the hydrographs not aligned well in time (NSE = -7.06). Adjustments to the conductivity slope and hydraulic conductivity parameters as well as the structure of the basin (the 1/3 of original area was made 100% permeable to amplify soil volume and thus infiltration) improved the calibration. However the simulation often failed to properly reproduce the observed hydrographs (NSE = -0.301). Thus, the module presented better predictive capabilities for a basin with a simpler overall design. Using the results from the calibration process, a general method is suggested to model a bioretention basin with a trapezoidal cross-sectional area in SWMM. Globally, the modelled bio-retention basin was most sensitive to the initial moisture content in the soil, influenced by rainfall characteristics (depth, duration, antecedent dry days), rather than specific user input parameters.
Click here to download a static PDF version of the presentation.