Global validation of the SoilClim soil moisture estimates using in-situ and remote sensing observations

Soil moisture is a key factor for plant growth and agricultural production. Therefore, it has become a fundamental part of agricultural drought monitoring systems developed for various spatial scales ranging from local to regional and global. Despite the development of large-scale soil moisture monitoring systems, in-situ soil moisture observations still remain inadequate for precise soil moisture monitoring, especially in remote areas, where there is a limited number of monitoring stations. Together with remote sensing technologies soil moisture modeling may provide an alternative to in-situ measurements that delivers spatially continuous estimates over large geographic areas. SoilClim is a semi-empirical water balance model that, together with other outputs (e.g., reference and actual evapotranspiration, soil temperature), estimates daily soil moisture in various depths of soil profile. The model has previously been validated on a total of 20 sites (5 in Central Europe and 15 in central USA) and is now used for global monitoring and prediction of soil moisture and drought intensity in an operational and interactive web platform (www.windy.com). Our study evaluates the SoilClim soil moisture global measurements with 0.1° spatial resolution using two independent sources of information: i) in-situ soil moisture measurements from the International Soil Moisture Network, and ii) the soil moisture derived from the Metop ASCAT sensors on Metop-A and Metop-B satellites. In the conference presentation we will introduce the SoilClim model and present results of the global validation including statistical spatial analysis and triple collocation.

Acknowledgement: This study was conducted with support of SustES - Adaptation strategies for sustainable ecosystem services and food security under adverse environmental conditions (CZ.02.1.01/0.0/0.0/16_019/0000797).