Distinguished Lecture Series in Energy: Dr. Ilya Zaslavsky

Dr. Ilya Zaslavsky, the Director of Spatial Information Systems Laboratory at the University of California San Diego, presents “Leveraging AI and Large Language Models to Navigate the Water-Energy-Food-Health Nexus” as a part of the Distinguished Lecture Series in Energy on Wednesday, November 13, 2024.

Abstract
--------------
Large language models (LLMs) are creating new opportunities for exploring the complex Water-Energy-Food-Health (WEFH) Nexus by extracting insights from diverse interdisciplinary sources such as scientific articles, policy documents, and environmental and health datasets. This presentation will showcase findings from several projects that tested LLM approaches for addressing WEFH challenges. We’ll also demonstrate an emerging software ecosystem built on open-source components, designed to support integrated data analysis and decision-making across WEFH domains.

One example is the NIH-funded Global Center on Climate Change, Water, Energy, Food, and Health Systems (GC3WEFH), a multi-institution, international project focused on modeling climate impacts on vulnerable communities in water-scarce regions of Jordan. Working with researchers at Texas A&M University and other partners, we explore how climate change affects health, food security, water quality and quantity, and water and energy use. GC3WEFH’s data hub offers an innovative data catalog that allows users to perform LLM-assisted natural language queries, enabling data access, visualizations, and predictive analytics for modeling environmental and health impacts.

A related NSF-funded project, the WEN-OKN (Water-Energy Nexus Open Knowledge Network), is building an LLM-driven network for data integration across water and energy datasets. Using LLMs combined with structured knowledge graphs, WEN-OKN enables federated queries across knowledge graphs and other sources such as the Urban Flooding Open Knowledge Network, Internet of Water, U.S. Energy Atlas, KnowWhereGraph, and Data Commons, to answer complex questions about water supply, flooding, energy demand, and water and energy infrastructure risks. These projects demonstrate the potential of an AI-driven, open-source ecosystem to deliver actionable insights and enhance decision-making across the WEFH Nexus.