Robust Köppen-Geiger (KG) Climate Classification Maps
The nc file contains two global Köppen-Geiger (KG) Climate Classification maps (resolution 0.5°). The robust maps are developed by using monthly temperature and precipitation data from UDEL, CRU, JRA55, NCEP/CFSR, MERRA2, ERA5, and WFDEI gridded products over the period 1980 to 2017. MasterMap 1 shows the most frequently observed climate types among all KG maps for each grid point, after directly merging the climate maps produced by the seven datasets. MasterMap 2 results from estimating the medians of precipitation and temperature among the seven datasets at each grid point.
Steps to reproduce
This study uses precipitation and 2-meter temperature data from gauged-based and reanalysis products, and products combining various sources of information (e.g., satellites, reanalysis, gauges, etc.). The two Gauge-based products are (1) the University of Delaware Air Temperature & Precipitation (UDEL, version 5.01), and (2) the Climate Research Unit, University of East Anglia, (CRU TS, version 4.04). The four reanalysis products are, (1) the National Centers for Environmental Prediction (NCEP)/Climate Forecast System Reanalysis (NCEP/CFSR, versions 1 and 2), (2) the Japanese 55-year Reanalysis (JRA-55), (3) the fifth-generation of the European Centre for Medium-Range Weather Forecasts (ERA5), (4) and the Modern-Era Retrospective analysis for Research and Applications (MERRA, version 2). The multi-source product used in this study is the WFDEI Meteorological Forcing Data. To keep consistency in terms of their gridded systems, all datasets were regridded to 0.5° by using the bilinear interpolation method. Also, to have a common temporal coverage in all datasets, a common period is chosen, that is, from 1980 to 2017. Finally, seven KG climate maps were derived using monthly precipitation and temperature from CRU, UDEL, MERRA 2, JRA-55, WFDEI, NCEP/CFSR, and ERA5 products. Two robust KG master maps (0.5° resolution) are created utilizing the characteristics of the seven gridded products. By merging the KG classes, MasterMap1, presents the most frequently observed climate subtype among all KG maps. Note that regional analysis of studying 8 surrounding grids was made to determine the climate subtype of each ambiguous grid in case of detecting equally frequent climate types. Also, some grids in the reanalysis-based maps were discarded due to their lack of representation for specific climate types (e.g., Dfd, Dwd, and Dsd) in some regions such as Yakutia. MasterMap2 is developed by using a single dataset from the medians of precipitation and temperature among all products. The two master maps are 95% similar, with most climate types showing less than a 0.5% difference in land coverage. Although KG maps derived from several products do not usually include the climate subtypes of Dsd, Dwd, Dfb, and Csc, these types are presented in the two master maps. Not that no notable differences were observed in the similarity of each dataset’s KG map with the two master KG maps for every latitudinal zone.