Reposted by Marc Bierkens
Global Hydrological Models are difficult to calibrate due to computational costs and regionalisation issues. Droppers, @waterbier.bsky.social and Wanders developed a deep learning surrogate model of their global hydrology and water resources model.
agupubs.onlinelibrary.wiley.com/doi/full/10....
agupubs.onlinelibrary.wiley.com
A Multi‐Resolution Deep‐Learning Surrogate Framework for Global Hydrological Models
We introduce a framework for the development of multi-resolution deep-learning global hydrological surrogates To test our framework, we develop and evaluate a surrogate for the PCRaster Global Wa...