欧亚大陆温性草地植被群系时空变异图—植被群系分类图（2010S）

本数据集为欧亚大陆温性草地植被群系时空变异图—植被群系分类图（2010S））。数据为tif栅格格式，空间分辨率为1公里，按5个草地类型进行植被群系分类，草地类型分别为：温性草甸草原、温性典型草原、温性荒漠草原、温性草原化荒漠、温性荒漠。 该数据集的数据基础：1980S中国科学院植物研究所为主持单位《中华人民共和国被图（1∶100万）》数据图集；1980S中国科学院中亚区域研究中心为主持单位《中亚五国植被类型图（1:250 万）》数据图集；1990年中国科学院地理科学与资源研究所为主持单位《蒙古国草地覆盖数据》30米分辨率数据图集；1980年代欧亚大陆区域NOAA和MODIS卫星数据、气象数据（ECMWF-ERA5）、土壤数据、地形数据。分类方法：依据上述数据基础，采用决策树机器分类方法获得欧亚大陆温性草地类型时空变异图。利用Python的scikit_learn包下的决策树进行因地制宜的分区建模，获得各个研究亚区的植被群系分类模型。精度验证：2010s草地类型分类结果是采用野外踏勘验证法。通过2021年对内蒙古部分地区的实地调研，共调查采集的78个样地中有57个样地的优势种或次优势种与模拟结果一致，验证精度达到73%。该数据可为欧亚大陆温性草地类型分布信息以及时空变异分析提供依据。

This dataset is the Spatio-temporal Variation Map of Temperate Grassland Vegetation Formations in Eurasia - Vegetation Formation Classification Map (2010s). The data is in TIFF raster format, with a spatial resolution of 1 km. Vegetation formations are classified into 5 grassland types, namely: temperate meadow steppe, temperate typical steppe, temperate desert steppe, temperate steppe-desert, and temperate desert. Data foundations of this dataset: 1) The "Vegetation Map of the People's Republic of China (1:1,000,000)" data set, hosted by the Institute of Botany, Chinese Academy of Sciences (CAS) in the 1980s; 2) The "Vegetation Type Map of the Five Central Asian Countries (1:2,500,000)" data set, hosted by the Center for Central Asia Research, Chinese Academy of Sciences in the 1980s; 3) The 30 m-resolution Grassland Cover Data of Mongolia, hosted by the Institute of Geographic Sciences and Natural Resources Research, CAS in 1990; 4) NOAA and MODIS satellite data, meteorological data (ECMWF-ERA5), soil data and topographic data over Eurasia from the 1980s. Classification method: Based on the above data foundations, the decision tree machine classification method was adopted to generate the spatio-temporal variation map of temperate grassland types in Eurasia. Decision trees from the scikit-learn package in Python were used to conduct site-specific zoning modeling, and vegetation formation classification models for each research sub-region were developed. Accuracy verification: The 2010s grassland type classification results were validated using the field survey method. Through field investigations in parts of Inner Mongolia in 2021, among the 78 collected sampled plots, the dominant or subdominant species of 57 plots were consistent with the simulation results, with a verification accuracy of 73%. This dataset can provide a basis for obtaining the distribution information of temperate grassland types in Eurasia and conducting spatio-temporal variation analysis.