Exploring on the climate regionalization of Qinling-Daba mountains based on Geodetector-SVM model

boundary between subtropical coniferous forests and temperate coniferous forests; Group I and Group III were used to delimit the boundaries between subtropical broad-leaved forests and temperate broad-leaved forests. Meteorological data. WorldClim 2.1 climate data were used as meteorological data [43] and included monthly temperature (minimum, maximum and average) and precipitation with a resolution of 1 km2 (downloaded from the website of http://www.worldclim.org/). These data were generated using the interpolation method of thin-plate spline based on 9000-60000 meteorological stations and its covariates, including elevation, distance to the coast and vari- ous covariates obtained with the MODIS satellite platform. The temporal range for station data was between 1970 and 2000. All data were aggregated to monthly climate averages. A total of 12 images are available for each variable. Meteorological background data of China. The data included the annual average tem- perature, the annual average precipitation, the accumulated temperature with �0˚C and the accumulated temperature with �10˚C with a resolution of 500 m2 (downloaded from the web- site of http://www.resdc.cn/). The data were generated by the interpolation method of inverse distance weighted averages based on data from 1915 national meteorological stations. The meteorological background data were then resampled to 1-km2 resolution by calculating the Fig 2. Spatial distribution data of vegetation and plant community survey data. https://doi.org/10.1371/journal.pone.0241047.g002 PLOS ONE

Climate regionalization based on Geodetector-SVM model PLOS ONE | https://doi.org/10.1371/journal.pone.0241047 November 2, 2020 5 / 16