RAPID2 static files for global implementation with MERIT Basins v1.0 and GLDAS v2.0

This dataset contains model parameters, network topology, and coupling files necessary for global river routing using version 2 of the Routing Application for Programmed Integration of Discharge (RAPID2) model. The data is formatted as computationally efficient Parquet files and is organized globally by level-2 Pfafstetter basin codes (pfaf_XX).

It builds upon earlier datasets used for global river water storage analysis, transitioning key routing parameters and topological arrays into the modernized .parquet format for improved performance and reduced file sizes.

Corresponding Peer-Reviewed Publication

This dataset shares its coupling methodology, connectivity logic, and parameterization with the study reported below. When making use of any files in this dataset, please cite both the dataset herein and the following article:

Collins, E. L., C. H. David, R. Riggs, G. H. Allen, T. M. Pavelsky, P. Lin, M. Pan, D. Yamazaki, R. K. Meentemeyer, and G. M. Sanchez (2024), Global patterns in river water storage dependent on residence time, Nature Geoscience, DOI: 10.1038/s41561-024-01421-5

Input Datasets

The files in this repository were derived from and built upon the following datasets:

• MERIT-Basins (version 1.0): Derived from MERIT-Hydro (version 0.7), available under a CC BY-NC-SA 4.0 license. https://www.reachhydro.org/home/params/merit-basins

• GLDAS Land Surface Models (V2.0): Monthly 1.0 x 1.0 degree models (VIC, Noah, and Catchment) available under a CC0 license. https://science.nasa.gov/spd-41/.

File Structure and Contents

The repository consists of 61 individual .zip archives. Each archive corresponds to a specific global hydrography region designated by a 2-digit Pfafstetter level 2 code (indicated by XX in the filenames).

Zip File Naming Convention: MERIT_Basins_v1.0_pfaf_XX_GLDAS_v2.0_RAPID_v2.0.zip

Internal File Structure: Within each zipped archive, you will find the corresponding arrays saved as .parquet files. Using Pfafstetter region 76 as an example, each archive contains files similar to the following:

• bas_pfaf_76_topo.parquet: River IDs organized from upstream to downstream following topological sort. 

• con_pfaf_76.parquet: River network connectivity (associating upstream and downstream nodes).

• crd_pfaf_76.parquet: Coordinate data for the routing network.

• cpl_pfaf_76_GLDAS.parquet: Coupling file that maps the GLDAS v2.0 land surface model outputs to the MERIT-Basins catchments.

• kpr_pfaf_76_nrm.parquet: Muskingum routing k parameter (representing normal residence time).

• xpr_pfaf_76_nrm.parquet: Muskingum routing x parameter (non-dimensional).

Known Limitations and Notes

• Missing Catchments: In the original files for MERIT-Basins (version 1.0) derived from MERIT-Hydro (version 0.7), two river reaches do not have a corresponding catchment: COMID=31000001 and COMID=61000003. These two river reaches were removed from the analysis, matching the total number of river reaches to the total number of catchments.

• Connectivity Logic: In the original MERIT-Basins (version 1.0) attribute tables, the reach with COMID=67072432 lists only four upstream reaches, despite having five on the map. This discrepancy is resolved in this dataset, as all connectivity (con_pfaf_XX.parquet) was generated by associating upstream and downstream nodes directly rather than relying on the legacy attribute table.