Hi, I am Jonathan. I am Assistant Professor of Machine Learning and Artificial Intelligence in Geological Sciences and a Faculty Fellow of the Alabama Water Institute.

I am both a scientist and an engineer. My scientific interests span many scales and many physical processes of Earth. I am particularly interested in what we can learn about the natural world by representing dynamic physical systems with models designed to simulate those processes. My engineering interests are focussed on water resources, particularly the design and life cycle of water resources infrastructure.

For more information, see my CV.

Tackling the big questions in Earth (and planetary) Sciences.

Tackling the big questions in Earth (and planetary) Sciences.

My work focuses on the intersection of water resources, geospatial data, and artificial intelligence, particularly in the context of sustainable data center growth and AI infrastructure expansion. I am both a geoscientist and an applied researcher, with interests in hydrology, geochemistry, and geospatial Earth processes. I focus on modeling water availability and atmospheric dynamics, developing digital twins of regional water systems, and evaluating the impacts of water use, climate, and infrastructure. I am also certified to operate drones (sUAS) for aerial mapping, environmental monitoring, geospatial data acquisition, and remote sensing.

Tackling the big questions in Earth (and planetary) Sciences.

I am working on LSTM-based predictions of streamflow for the Next Generation Water Resources Modeling Framework Prototype . Specifically, I focus on efficient training methods for high temporal resolution (1-hour and sub-hourly) hydrologic forecasts.

Deon is leading a paper on hydrologic extrapolation, bringing his physics background to advance machine learning approaches in Earth sciences.

Check out the preprint here: Link.

I build machine learning and data assimilation tools to strengthen national-scale hydrologic prediction in NOAA’s Next Generation Water Resources Modeling Framework. I’m a technologist and problem-solver who enjoys challenging work and loves coding and logic—where scale, reliability, and rigor make the work exciting. I love building end-to-end solutions: integrating systems, writing reliable code, and running rigorous evaluations that lead to more dependable forecasts.

I build machine learning and data assimilation tools that improve national-scale hydrologic prediction in NOAA's Next Generation Water Resources Modeling Framework. I'm a technologist and experienced software developer who enjoys tough problems - where messy data, real constraints, and performance requirements meet. I love building end-to-end systems: clean pipelines, reliable code, and evaluations that translate into better forecasts for real-world water and environmental decisions.

Jackson worked on a toy experimet to develop machine learning capabilities for surface water terrain routing. See the Deep Routing Lab here: https://github.com/DualEarth/deep_routing_lab