2025 Internship Project Overview
Our project focused on improving groundwater management in New Jersey, where increasing demand and climate change make water resources more vulnerable to depletion and saltwater intrusion. Accurate pumping data are critical for planning, but existing records are often too sparse or aggregated to capture local conditions. To address this gap, we developed a machine learning framework to estimate monthly groundwater withdrawals at fine spatial resolution. Our approach used a three-stage ensemble model that first identified whether pumping occurred, then distinguished between low and high withdrawals, and finally applied specialized models to predict pumping volumes. We trained the system on a range of environmental and demographic features, achieving strong accuracy while uncovering how local geography and climate influence water use. By providing more precise estimates of pumping, our work contributes to more effective water management and supports ongoing research in sustainable groundwater use.
ORGANIZATION/RESEARCH GROUP
Integrated GroundWater Modeling Center, Department of Civil and Environmental Engineering, Princeton University
LOCATION
Princeton, New Jersey
Interns

Minor(s): Statistics and Machine Learning
Click to view the end-of-project presentation
Principal Investigator

Mentors
- Lisa Gallagher, Professional Specialist, Civil and Environmental Engineering
- Yueling Ma, Postdoctoral Research Associate, High Meadows Environmental Institute


