Improving terrestrial evapotranspiration estimation across China during 2000-2018 with machine learning methods

This dataset is produced by the paper (Improving terrestrial evapotranspiration estimation across China during 2000-2018 with mechine learning methods). We used the random forest to merge five process-based ET model.The data in distinct Geotiff file are named as ‘RF_yy_dd_ET.tif’, where ‘yy’ and ‘dd’ denote year, order. The order is from 1 to 36. There are 3 issues per month. Assuming there are N days in this month, the unit of the first 2 issues is mm/10day, and the unit of the third issue is mm/(N-20) day. The scale factor is 0.01.

本数据集源自论文《Improving terrestrial evapotranspiration estimation across China during 2000-2018 with machine learning methods》。我们采用随机森林（Random Forest）算法，融合了5个基于物理过程的蒸散发（Evapotranspiration，ET）模型。所有数据以独立的GeoTIFF（Geotiff）格式文件存储，命名规则为"RF_yy_dd_ET.tif"，其中"yy"代表年份，"dd"代表数据序号。数据序号的取值范围为1至36，每月包含3期数据。假设当月总天数为N，则前两期数据的单位为毫米/10日，第三期数据的单位为毫米/(N-20)日。数据缩放因子为0.01。