Assessment of the physical, socioeconomic and climatic constraints on Green Infrastructure

Modeling the impacts of Green Infrastructure (GI) in urban watersheds requires quantification of various physical, socioeconomic, climatic, and other uncertainties. The hypothesis put forth in this dissertation asserts that of all of these, socioeconomic factors are most important drivers of the impacts that GI can have in urban watersheds. This study is conducted using an online GI assessment model, the Low Impact Development Rapid Assessment model (LIDRA), to estimate the cost-effectiveness of GI in reducing runoff and manage Combined Sewer Overflows (CSOs) under a variety of conditions. Physical, socioeconomic, and climatic uncertainty are investigated independently. An actual urban watershed, HP009, in Bronx, NYC, is employed as a case study to perform the assessment of various uncertainties.;Physical and socioeconomic uncertainties are mostly modeled using Monte Carlo (MC) models using symmetric triangular distributions. This distribution is either defined by the maximum and minimum values of uncertainties such as GI dimensions and cost, or a certain range whose mode is equal to the default or user defined value of uncertainties (e.g. soil type, implementation rate, interest and inflation rates). LIDRA results reveals that the cost-effectiveness of GI performance is less sensitive to physical factors than socioeconomic ones. The cost analysis suggests that social/institutional uncertainties associated with the rate of implementation of GI exert a highly significant influence on GI program cost. The analysis of the effectiveness of GI in reducing runoff also underscores the importance of implementation rate, but in addition reveals the importance of GI strategy selection, another socially/institutionally determined factor. An optimal solution can be achieved by pursuing a balance between cost and effectiveness, though the feasibility of such a scenario in a particular watershed is socioeconomically determined and may not actually be feasible.;The importance of climatic uncertainty is tested both by modeling historical precipitation variability and by incorporating non-stationary future expected changes to precipitation as forecast by global climate models (GCMs). Historical precipitation variability is represented using a non-parametric single variable bootstrapping Markov model. Future non-stationary changes in precipitation are simulated using GCM-projected changes in future temperature, along with historical relationships between pressure change and precipitation. After a thorough analysis of the relationships between Pressure Change Event (PCE) occurrence and Average Monthly Temperature (AMT), an association is found to relate hourly precipitation series and AMT projections by mapping PCEs to its AMT under the same season and similar climate condition. A stochastic algorithm is then developed as a combination of a "Moving Temporal Window", a "Moving Temperature Window" and a multi-variable bootstrapping.;The relative importance of GI implementation rate, GI strategy decision, and climatic conditions are compared in a sensitivity analysis. The results indicate that the runoff reduction and CSO mitigation achievable through GI is most sensitive to the socioeconomic factors that determine implementation rate and GI strategy decisions. Socioeconomic factors are thus expected to most significantly determine the overall cost-effectiveness of GI as a strategy for reducing urban runoff at the watershed scale.