ScholarWorks@성균관대학교

초록

Soil moisture is an important factor in the analysis of the water cycle and natural disasters with a catchment area. With the spatio-temporal variability analysis of in-situ soil moisture datasets, robust soil moisture monitoring for the validation of satellite dataset could be conducted. The optimal and minimum necessary sampling size (NSS) using the in-situ soil moisture data at a depth of 10 cm, as measured by time domain reflectometry (TDR) was investigated in the Chengmicheon (CMC) and Seolmacheon (SMC) basins from 21 August 2018 to 20 August 2019 (excluding winter). Specifically, the in-situ soil moisture data that comprising 18 and 19 points were used to detect the spatio-temporal variation pattern of soil moisture in the CMC and SMC demonstration areas. The representing spatial mean value points in the CMC and SMC demonstration areas were identified based on the Temporal Stability of Index (TSI). The mean soil moisture coefficient of variation in the CMC and SMC demonstration areas were 0.20 and 0.37, respectively. In two study areas, the representative points were able to capture the spatial and temporal mean soil moisture with high accuracy (R2 value of greater than 0.88), and the two representative points in the two demonstration areas were both located in areas with soil moisture that was drier than the mean values. In addition, the number of soil moisture points based on the Random Combination Method (RCM) were examined utilising the Coefficient of Determination (R2) and Root Mean Square Error (RMSE) accuracy indicators, with a focus on the time series variation of spatial mean soil moisture. The precision of the time series estimates of spatial mean soil moisture has been improved as the number of soil moisture observations increases, and the requisite minimum number of soil moisture points in the CMC and SMC demonstration areas were 4 with an RMSE of less than 2% and an R2 accuracy of greater than 0.95. In addition, a statistical sampling analysis based on the spatial mean soil moisture was conducted to determine the requisite number of points, indicating that the number of requisite soil moisture points has been increased in proportion to the reduction in absolute error at a given confidence level. The minimum number of soil moisture points required at the 95% confidence level with an absolute error of 4% was determined to be 4 and 28 in the CMC and SMC demonstration areas, respectively, due to the high spatial variability of soil moisture in the SMC demonstration area. © 2025 Korea Water Resources Association.

키워드