Assessing the impacts of social and climate uncertainties in managing stormwater runoff reduction and associated CSO risk using green infrastructure

Abstract:

Existing modeling tools provide, at best, deterministic estimates of green infrastructure’s (GI) effectiveness in managing stormwater runoff. However, due to various uncertain physical, climatic, and socioeconomic conditions that arise over the decades during which GI programs are typically implemented, this problem is perhaps better approached probabilistically. This paper uses the Low Impact Development Rapid Assessment (LIDRA) model to assess GI’s ability in managing stormwater runoff and associated Combined Sewer Overflows (CSOs) risks considering uncertainties from both climate change and socio-physical aspects. A real watershed in New York City is employed for this purpose. The model explicitly propagates known uncertainties in GI size, soil properties and climate (both stationary and non-stationary) stochastic estimates of 30-year program runoff reduction efficiency. Other uncertain parameters (e.g. implementation rate) are user-defined and factored explicitly into the results. Considered together with validated hydrologic and hydraulic modeling of the local watershed system, LIDRA’s runoff reduction results are also used to compute the ability of GI to reduce the CSO risk from the study area. A fast development plan may benefit the watershed hydrologically but with exponential high cost from a long-term perspective. After all the proposed GI is implemented in the case watershed, the annual risk of CSO decreases from 12.5% to 8%. While climate change could raise the initial CSO risk to around 30%, a fully developed GI network could reduce the risk to 23%. Yet, since the climate change makes the environmental challenges more severe, decision makers may have to make compromise with the cost burden brought by social uncertainties.