Research
Our Groundwater Sustainability and Data Sciences research group combines process-based models with data-driven methods to improve predictive capability and understanding of water resources systems, in particular, under human adaptations and global change.
Ongoing Projects
Predictive Modeling of Arizona Groundwater Quality Using Transfer Learning.
This project is funded by GCWT under AWII and aims to support the ADEQ Groundwater Quality Monitoring Program by developing a predictive model of concentrations of key contaminants using state-of-the-art machine learning (ML) techniques. The model will provide actionable information for efficient and proactive water quality monitoring and management.
Critical Aspects of Sustainability (CAS)-Climate: Actionable Heat and Carbon Mitigation by Urban Greening–Integrating Physical Modeling and Machine Learning for Decision Support
This project aims to develop a transformative platform by integrating the physically based modeling of urban system dynamics and machine learning-based techniques in support of decision-making and urban planning.
Opportunities to Enhance Recharge in Arizona
Our group is working with a tri-university research team consisting of researchers from the University of Arizona (UoA), Northern Arizona University (NAU), and Arizona State University (ASU) on a collaborative project. The tri-university team are assisting the Arizona Department of Water Resources (DWR) to (1) identify areas where water that would otherwise evaporate could be captured for underground storage without affecting surface water flows; and (2) assess the potential to increase water supply availability in Arizona via enhanced recharge of unallocated water sources in both urban and rural areas of the state.
Smart Tree Watering in Arizona’s Urban Environment
In this collaborative project, we will work with researchers from the University of Arizona and Arizona State University to provide mobile, efficient, and scalable urban tree watering solutions through place-based research in Tucson and Phoenix metropolitan area. Specifically, our group is using machine learning-based fast surrogate models to estimate city-scale water savings achieveable by smart watering schemes.
Designing Nature to Enhance Resilience of built infrastructure in Western US Landscapes
This collaborative US Army Corps funded project will develop a modeling toolkit that will allow for rapid, scenario-based assessment of outcomes of combinations of natural and built infrastructure in a hydrologic and water resources context. Our research group is helping to develop groundwater modeling component as well as combining domain science with data science for higher efficiency and reliability of the toolkit.
Quantifying Watershed Dynamics in Snow-Dominated Mountainous Watersheds Using Hybrid Physically Based and Deep Learning Models
Snow dominated mountainous karst watersheds are a primary water supply in many parts of the world. These watersheds are typically characterized by complex terrain, spatiotemporally varying snow accumulation and melt process, and complex flow and storage dynamics due to a high degree of hydrogeological heterogeneity. As a result, predicting streamflow from meteorological inputs has been challenging. We are developing a hybrid modeling approach that integrates an energy balance snow model with a deep learning rainfall-runoff model. Working with our collaborators at Utah State University and Boise State University, we will use various field sampling data to verify, interpret, and constrain the deep learning model.
Developing an irrigation dataset for assessment of anthropogenic impacts on terrestrial-atmosphere energy-water coupling using machine learning-based data fusion
Mechanistic understanding and prediction capability of agricultural irrigation impacts on terrestrial hydrologic cycle and land-atmosphere feedback have been limited due to inadequate representation of irrigation processes in existing models. We are developing a spatially resolved dataset of irrigation amount and timing for the High Plains region by blending in situ, remote sensing, and reanalysis datasets using machine learning-based data fusion.
Advancing Data Science and Analytics for Water (DSAW)
DSAW is a multi-institution project to develop new software that will enhance scientists’ ability to apply advanced data visualization and analysis methods (collectively referred to as “data science” methods) in the hydrology and water resources domain. Our group is develping water-data science applications that demonstrate the usage of exploratory data analsysis and machine learning for bias diagnosis and correction of groundwater models.
Quantifying trade-offs in water quantity and quality in a managed aquifer recharge system
In this collaborative work, we performed detailed modeling and field sampling to investigate the infiltration capacity and groundwater contamination and health risks of recharging treated wastewater.
