EMBRACE-EAR-Seed: The impact of climate change on karst groundwater resources with deep learning approach

Sam Houston State University, Huntsville TX

Investigators

Abstract

Karst aquifers, formed in soluble rock formations like limestone, supply drinking water to approximately 25% of the world's population and 40% of the groundwater used for drinking in the United States. Karst aquifers are characterized by complex networks of fractures, conduits, caves, and sinkholes, which contribute to high complexity in hydraulic properties. These complex groundwater systems are vulnerable to climate change, yet their responses remain poorly understood due to their intricate hydrological processes. This research aims to bridge this knowledge gap by leveraging advanced deep-learning techniques to predict and analyze the impact of climate change on karst groundwater resources. By integrating extensive meteorological and hydrological data with state-of-the-art artificial intelligence models, the research will provide insights into the future sustainability of karst groundwater resources. The project's findings will be valuable for policymakers and stakeholders in water resource management, contributing to informed decision-making processes in environmental conservation and public policy. Moreover, the project will enhance STEM education and Geoscience Learning Ecosystems (GLEs) by actively involving students from underrepresented groups in cutting-edge research, integrating the research components into the curriculum to enhance students’ data analysis and quantitative skills, and raising public awareness about karst aquifers and climate change through seminars, lectures, and collaborations. The overarching goal of this research is to investigate the impact of climate change on karst groundwater resources using advanced deep-learning models. The research will develop innovative deep-learning frameworks specifically designed to capture the highly nonlinear and nonstationary behaviors characteristic of karst hydrology, providing more accurate predictions of spring discharge rates, groundwater levels, and water quality. These models will be trained on comprehensive datasets from various U.S. karst aquifers and integrated with climate projection data from the Coupled Model Intercomparison Project (CMIP) to forecast the status of karst groundwater resources under different greenhouse gas emission scenarios. This project will provide a detailed analysis of how different climate change trajectories might impact groundwater availability, spring discharge rates, and related ecosystems in karst terrains. The insights gleaned from this research are expected to contribute to increased understanding of karst terrains' resilience and vulnerabilities in the face of climate change, thereby informing more effective conservation and management strategies. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.