CISL Seminar/RAL - Advancing integrated continental-scale hydrologic simulation and forecasting through democratized data and ML-accelerated modeling
1:00 – 2:00 pm MDT
Abstract
Today, water and resource managers face a significant challenge managing systems that are dynamic and rapidly evolving, where historical observations are no longer a reliable guide. Capturing interactions from bedrock to treetops is important for understanding water stresses and represents a critical gap in our current models. Simulations with integrated hydrology models (that solve the 3D Richards' equation and 2D shallow water equations in a globally implicit manner) provide robust results out to continental scales yet are computationally expensive. Groundwater-surface water interactions are tightly coupled and can have a large impact on watershed dynamics yet are challenging for all models to accurately resolve.
We have developed a hybrid physics-based, machine learning digital twin over the entire continental US (CONUS). This proof-of-concept forecast system runs operationally, providing all hydrologic states and fluxes from bedrock to the top of the canopy at hourly timesteps and greater than 1 km resolution. Automated comparison to observations is enabled through the HydroData platform, supporting continuous evaluation and model improvement. This talk will highlight the technical challenges of combining integrated hydrologic modeling with machine learning in a national forecast system, including physics-based approaches that improve solver performance by more than an order of magnitude for continental-scale simulations. Machine learning emulators embedded within integrated hydrology models can also drastically reduce computational burden and provide 30 m spatial resolution for groundwater and surface water. We advance a vision that deploys these models and openly available forcing and parameter datasets to understand future water challenges from local to continental scales.
Here is the public livestream link. Staff members can look for a Google Calendar invitation for the talk.
Please reach out to Sam Scalice (sscalice@ucar.edu) with any questions you may have.
Name
Reed Maxwell
Reed Maxwell is the William and Edna Macaleer Professor of Engineering and Applied Science at Princeton University, where he holds appointments in Civil and Environmental Engineering and the High Meadows Environmental Institute and directs the Integrated GroundWater Modeling Center. His group studies how water moves through the connected hydrologic cycle, from groundwater to the land surface to the atmosphere, and how that system responds to human pressures. Much of this work runs on ParFlow, an open-source parallel model that simulates water across continents at high resolution and scales to the largest machines available.
Reed has published 195 papers and teaches hydrology and fluid mechanics. He was the 2020 Henry Darcy Distinguished Lecturer and is an elected Fellow of the American Geophysical Union. Before Princeton he was on the faculty at the Colorado School of Mines and a member of the Hydrologic Sciences group at Lawrence Livermore National Laboratory. He earned his PhD at UC Berkeley.