회귀크리깅을 이용한 AMSR2 토양수분자료의 다운스케일링

Soil moisture is an important meteorological factor to understand hydrological circulation and is closely associated with disasters such as drought, flood and wildfire. However, the spatial resolution of satellite-based soil moisture data is too coarse to be applied directly to local analysis. To solve the problem of the restricted spatial resolution of soil moisture data retrieved from microwave satellite sensors, this study presents a downscaling method that combines spatial statistical models with various land surface variables. Regression-kriging, which is known as the most elaborated downscaling technique, was employed to downscaling of the daily soil moisture data from AMSR2 (Advanced Microwave Scanning Radiometer 2) at the resolution of 10km and 25km for the period between April and October, 2013-2014. The downscaled result at the resolution of 2km and 4km showed very good consistency with the original data, which means the spatial patterns and the data properties were well preserved even after downscaling. Our approach will be applied to other meteorological factors and can be a viable option for overcoming the problem of limited spatial resolution in satellite images and numerical model data.