GRUN: Global Runoff Reconstruction (GRUN_v1)

Abstract

The dataset contains a gridded global reconstruction of monthly runoff timeseries. In-situ streamflow observations from the GSIM dataset are used to train a machine learning algorithm that predicts monthly runoff rates based on antecedent precipitation and temperature from the Global Soil Wetness Project Phase 3 (GSWP3) meteorological forcing dataset. We thank Prof. Dr. Hyungjun Kim for developing the GSWP3 dataset and providing us with early access to the data. The data is stored in a single NetCDFv4 file at monthly resolution covering the period 1902-2014. The dataset is provided on a 0.5 degrees (WGS84) grid in units of mm/day. The runoff time series correspond to the ensemble mean of 50 reconstructions obtained by training the machine learning model with different subsets of data. Users interested in using the individual ensemble members of the reconstruction are invited to contact the authors directly. When using this dataset, please cite: Ghiggi, G., Humphrey, V., Seneviratne, S. I., Gudmundsson (2019), GRUN: An observations-based global gridded runoff dataset from 1902 to 2014, Earth Syst. Sci. Data, XXX, doi: XXXXXXXXXXX The complete collection of in-situ streamflow observations from the GSIM archive can be found at: - https://doi.pangaea.de/10.1594/PANGAEA.887477 - https://doi.pangaea.de/10.1594/PANGAEA.887470 For further information on the GSIM dataset see: - https://doi.org/10.5194/essd-10-765-2018 - https://doi.org/10.5194/essd-10-787-2018 For further information on GSWP3, see: - https://doi.org/10.20783/DIAS.501 - https://hyungjun.github.io/GSWP3.DataDescription - http://hydro.iis.u-tokyo.ac.jp/GSWP3/exp1.html