Exploratory versus Predictive Approach for Proactive Decision-Making in Water Resource Management

Author: ORCID icon orcid.org/0000-0003-0903-3223
Nazemi, Neda, Systems Engineering - School of Engineering and Applied Science, University of Virginia
Louis, Garrick, EN-CEE, University of Virginia

Water resource management (WRM) is a complex and challenging task, as it frequently involves making critical trade-offs under conditions of high uncertainty, complexity, and knowledge constraints. These challenges arise from the numerous interconnected and interdependent factors that influence water resources, the diversity of stakeholders with conflicting interests and values, and the incomplete, uncertain, and often contradictory understanding of the system. In such a context, forward-thinking and adaptive decision-making are essential for effective WRM. The overarching aim of this research is to compare a range of exploratory and predictive foresight methods to assess their potential applicability in addressing various WRM problems.
This research specifically investigates agricultural water management (AgWM) and hazardous algal bloom (HAB) management as two critical and challenging WRM problems that have significant environmental, social, and economic implications. To tackle these issues, the research is divided into two main parts.
In the first part, a formative scenario planning method is employed to explore the adaptation strategies for AgWM in arid and semi-arid regions. This process involves constructing a small set of coherent and distinctive governance scenarios that consider various factors, such as climate change, technological advancements, and social and political developments. These scenarios provide a robust framework for decision-makers to develop proactive and flexible strategies that can effectively respond to the evolving challenges in AgWM.
In the second part of this research, a set of data-driven predictive models is developed to provide near-term forecasts of HABs, which can support the prevention, control, and mitigation of these harmful events. These models leverage advanced machine learning techniques and a variety of data sources to generate accurate and timely predictions of algal bloom occurrences. By providing reliable and actionable information, these predictive models can inform WRM stakeholders and support evidence-based decision-making to minimize the negative impacts of HABs on water quality, public health, and the environment.
In conclusion, this research contributes to the field of WRM by comparing and evaluating a range of foresight methods that can address the complex challenges associated with AgWM and HAB management. The findings of this study can provide valuable insights for policymakers, practitioners, and researchers in WRM, and help inform the development of more effective, adaptive, and sustainable strategies for managing water resources in the face of uncertainty and change.

PHD (Doctor of Philosophy)
Proactive Water resource management, Agricultural water management, Hazardous algal bloom management, Foresight methods, Predictive modeling, Ecological Forecasting, Long Short-term Memory , Convolutional Neural Network
Issued Date: