Sentinel-1/Sentinel-2 derived soil Moisture product at Plot scale (S2MP)

The soil moisture maps were carried out at a plot scale. A map is provided each 6 days (12 days with Sentinel-1A and 12 days with Sentinel-1B). Inversion algorithm for estimating soil moisture was applied for agricultural areas with any vegetation cover. Land cover map is used to extract the agricultural areas. Sentinel-2 images corrected for atmospheric effects are used to calculate the NDVI (Normalized Differential Vegetation Index). Using several in situ measurements of soil moisture, the accuracy on soil moisture estimation was evaluated to be about 5 vol.%. In the provided soil moisture maps (WGS84, EPSG: 4326), the soil moisture values (mv) are multiplied by 5. In order to derive the estimated soil moisture value from the provided maps it is necessary to divide by 5 : volumetric soil moisture (in vol.%) = value obtatiend from the map / 5. Null values in the soil moisture maps = no data (no soil moisture estimation)

本土壤湿度图集以地块尺度生成。该图集更新周期为6天，其中12幅数据由Sentinel-1A生成，另外12幅由Sentinel-1B生成。针对存在任意植被覆盖的农业区域，采用反演算法开展土壤湿度估算；先通过土地覆盖图提取农业区域范围。采用经大气校正的Sentinel-2影像计算归一化差分植被指数（Normalized Differential Vegetation Index）。通过多组原位土壤湿度实测数据对土壤湿度估算结果的精度进行验证，其估算精度约为5体积百分比（vol.%）。本次提供的土壤湿度图采用WGS84坐标系（EPSG: 4326），图中土壤湿度数值（mv）已被放大5倍。若需从本数据集中获取真实的体积土壤含水率估算值，需将图中数值除以5：体积土壤含水率（vol.%）= 图中获取的数值 / 5。土壤湿度图中的空值代表无有效数据（未开展土壤湿度估算）。