A multiagent socio-hydrologic framework for integrated green infrastructure and water resource management in a changing environment

Abstract

Green infrastructures (GIs) are a nature-based solution for capturing and using rainwater, and are being developed in cities to complement centralized water supply. However, they also present challenges to urban and watershed water resource management. To address these challenges, this thesis proposes a multiagent socio-hydrologic framework for quantitative analysis of integrated green infrastructure and water resource management (IGWM). The core of this framework consists of two agent-based models to simulate decision-makings of urban water (UWM) and watershed manager (WM) agents for IGWM. By integrating these models with hydrologic and socioeconomic models, various multiagent systems are built to mimic both environment- and agent-to-agent interactions, which can reflect the socio-hydrologic dynamics of a watershed, driven by IGWM, vary across spatiotemporal scales, and analyze IGWM-related issues. The framework is first applied to the Upper Mississippi River Basin to explore case-specific features of IGWM at three spatial scales. Chapter 2 simulates the optimal responses of city-scale IGWM to different hydroclimatic conditions, classifying four type patterns of water supply; demonstrates the dissimilarity in water usage costs between up-and downstream areas in inter-city scale IGWM - an increase of 1%; and finds two optimal water strategies in watershed-scale IGWM, indicating that equitable water allocation strategies can generate different costs. Second, the framework is used in the Colorado River Lower Basin to examine short- and long-term IGWM in a watershed. It is modified for short-term IGWM and further extended for long-term IGWM by introducing an urban water demand prediction model. Chapter 3 and 4 assess the performance of water allocation schemes, indicating that in both the short- and long-term, different policies have varying impacts, i.e., water trading scheme saves costs in water use significantly – a reduction of 14.6% and 18.4%, while water tariff scheme promotes equitable water allocation – an increase of 45.7% and 51.7%, and water quota scheme is a moderate approach; identify the role of GIs in IGWM, denoting their effects that can save costs in water use - a reduction of 1.8%, and hurt equity in water allocation - a reduction of 16.9% - in the short-term, however, these impacts gradually reduce over time. Third, the framework is transformed into an uncertain formulation by setting partial parameters as fuzzy random variables. It is then applied in the Colorado River Lower Basin to simulate IGWM under two uncertain settings - uncertainty in hydroclimatic input and water demands. Chapter 5 examines the ramifications of these uncertain conditions on watershed-scale IGWM, highlighting their adverse consequences, including an increase of 8.4% in cost and a decrease of 10.2% in equitable water allocation; and assesses the effects of water schemes and GIs in mitigating these negative impacts, which reveals the varying impacts of three schemes and the inefficacy of GIs’ influence. In conclusion, this thesis explores the spatiotemporal characteristics of IGWM and can advance our understanding of the role of GIs in water resource management from a macro view - the effect of GIs is local and short-term. The framework can assist water managers in making decisions related to IGWM.