global hydrologic simulation dataset

[Chaopeng Shen]

Dear all,

We provide a global-scale high-resolution hydrologic simulation dataset for scientific use and assessment. We welcome inputs and evaluation from the community. Note it is intended as a global seamless simulation on the MERIT network, rather than specialized in any region.

This dataset is derived from this paper: Ji et al., 2025 Distinct hydrologic response patterns and trends worldwide revealed by physics-embedded learning https://www.nature.com/articles/s41467-025-64367-1.

Data for US and Europe can be downloaded from this zenodo https://zenodo.org/records/17417150; for the whole world, due to file size limit, it is uploaded to hydroshare: http://www.hydroshare.org/resource/6c8191d3613c4477b717be41c81a4372. This simulation contains a full suite of physical variables including ET, baseflow and other variables as shown in the paper, but due to file size limit, we only uploaded streamflow now. If you are interested in other variables, let us know. Figure S2c below in the supplementary information shows the training gauges.

Feedback welcomed.

Best

=================

--

Chaopeng Shen

Professor

Department of Civil and Environmental Engineering

281 ECoRE Building
The Pennsylvania State University
University Park, PA 16802

Email: cs...@engr.psu.edu
Office: 814-863-5844

Web: https://water.engr.psu.edu/shen/ 

LinkedIn: https://www.linkedin.com/in/chaopeng/Our group on AGU TV: https://www.youtube.com/watch?v=BIBBlM0BWaU

Shen's interview on the Apple Finch Pudding science podcast and a news 

Check out our national-scale and global-scale high-resolution differentiable hydrologic model and long-term seamless simulation datasets.

Our code collection and benchmarks: https://mhpi.github.io/