Articles | Volume 26, issue 11
https://doi.org/10.5194/hess-26-2969-2022
https://doi.org/10.5194/hess-26-2969-2022
Research article
 | 
15 Jun 2022
Research article |  | 15 Jun 2022

A two-step merging strategy for incorporating multi-source precipitation products and gauge observations using machine learning classification and regression over China

Huajin Lei, Hongyu Zhao, and Tianqi Ao

Data sets

Integrated Multi-satellitE Retrievals for GPM (IMERG) Technical Documentation G. J. Huffman, D. T. Bolvin, E. J. Nelkin, and J. K. Tan https://gpm1.gesdisc.eosdis.nasa.gov/data/GPM_L3/GPM_3IMERGDF.06/

GSMaP (Global Satellite Mapping of Precipitation) JAXA http://sharaku.eorc.jaxa.jp/GSMaP/index.htm

A quasi-global precipitation time series for drought monitoring C. C. Funk, P. J. Peterson, M. F. Landsfeld, D. H. Pedreros, J. P. Verdin, J. D. Rowland, B. E. Romero, G. J. Husak, J. C. Michaelsen, and A. P. Verdin https://data.chc.ucsb.edu/products/CHIRPS-2.0/

NOAA Climate Data Record (CDR) of Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks (PERSIANN-CDR), Version 1, Revision 1 UC-IRVINE/CHRS - Center for Hydrometeorology and Remote Sensing, University of California, Irvine https://www.ncei.noaa.gov/data/precipitation-persiann/access/

NOAA CPC Morphing Technique (CMORPH) Global Precipitation Analyses Climate Prediction Center https://ftp.cpc.ncep.noaa.gov/precip/CMORPH_V1.0/CRT/

ERA5-Land monthly averaged data from 1950 to present NCEP/NCAR - National Centers for Environment Prediction and National Center for Atmospheric Research - and ECMWF - European Centre for Medium-Range Weather Forecasts https://cds.climate.copernicus.eu/cdsapp#!/dataset/10.24381/cds.68d2bb30

NASA/GSFC/HSL, GLDAS Noah Land Surface Model L4 3 hourly 0.25x0.25 degree V2.1 H. Beaudoing and M. Rodell https://doi.org/10.5067/E7TYRXPJKWOQ

Download
Short summary
How to combine multi-source precipitation data effectively is one of the hot topics in hydrometeorological research. This study presents a two-step merging strategy based on machine learning for multi-source precipitation merging over China. The results demonstrate that the proposed method effectively distinguishes the occurrence of precipitation events and reduces the error in precipitation estimation. This method is robust and may be successfully applied to other areas even with scarce data.