A Remote Sensing-based global 10-day resolution Surface Soil Moisture dataset (RSSSM, 2003~2020)

Based on 11 well-acknowledged global-scale microwave remote sensing-based surface soil moisture products, and with 9 main quality impact factors of microwave-based soil moisture retrieval incorporated, we developed the Remote Sensing-based global Surface Soil Moisture dataset (RSSSM, 2003~2020) through a complicated neural network approach. The spatial resolution of RSSSM is 0.1°, while the temporal resolution is approximately 10 days. The original dataset covered 2003~2018, but now it has been updated to 2020. RSSSM dataset is outstanding in terms of temporal continuity, and has full spatial coverage except for snow, ice and water bodies. The comparison against the global-scale in-situ soil moisture measurements indicates that RSSSM has a higher spatial and temporal accuracy than most of the frequently-used global/regional long-term surface soil moisture datasets. In addition, although RSSSM is remote sensing based, without the incorporation of any precipitation data or records, its interannual variation generally conforms with that of precipitation (e.g., the GPM IMERG precipitation data) and Standardized Precipitation Evapotranspiration Index (SPEI). Moreover, RSSSM can also reflect the impact of human activities, e.g., urbanization, cropland irrigation and afforestation on soil moisture changes to some degree. The data is in ‘Tiff’ format, and the size after compression is 2.48 GB. The relevant data describing paper has been published in the Journal ‘Earth System Science Data’ in 2021.

基于11个全球尺度微波遥感地表土壤水分产品，并融入了9个主要基于微波的土壤水分获取质量影响因素，我们通过复杂的神经网络方法开发了基于遥感的全球地表土壤水分数据集（RSSSM，2003~2020）。RSSSM的空间分辨率为0.1°，时间分辨率为大约10天。原始数据集覆盖了2003~2018年，现已更新至2020年。RSSSM数据集在时间连续性方面表现卓越，除雪、冰和水体外，实现了全覆盖的空间覆盖。与全球尺度的原位土壤水分测量结果相比，RSSSM在空间和时间精度上优于大多数常用的全球/区域长期地表土壤水分数据集。此外，尽管RSSSM基于遥感，但未包含任何降水量数据或记录，其年际变化通常与降水量（例如，GPM IMERG降水量数据）和标准化降水量蒸散指数（SPEI）相符合。此外，RSSSM还能在一定程度上反映人类活动（如城市化、农田灌溉和植树造林）对土壤水分变化的影响。数据以‘Tiff’格式存储，压缩后大小为2.48 GB。相关描述论文已于2021年发表在《地球系统科学数据》期刊上。