Algorithm Development for Decision Support
Decision Support for the Sustainable Management of California Groundwater
|
|
Motivation
Current models for groundwater resource estimation are relatively accurate but too complicated for decision support. The CA Sustainable Groundwater Management Act requires that groundwater be managed sustainably by local authorities.
Objective
Develop a prototype for decision support capability for groundwater resource management that allows fast what-if scenario analysis and enables timely decisions.
Proposed approach
Train computationally cheap surrogate models on available observation data at the watershed scale. This will enable decision makers to run many what-if scenarios in order to manage groundwater optimally.
Related publications
J. Mueller, J. Park, R.Sahu, C. Varadharajan, B. Arora, B. Faybishenko, D. Agarwal,
"Surrogate optimization of deep neural networks for groundwater predictions",
preprint., 2019.
[link].
|
