Validating Korean model by estimating soil available water content under soybean cultivation for upland drought forecasting

Climate change is increasing the frequency of extreme weather events such as heatwaves, heavy rainfall, and drought.This is concerning for agricultural fields that rely on natural rainfall, as drought can significantly impact crop growth and yield.To address the annually recurring upland drought, model for assessing upland drought based on soil available water content (SAWC) has been used to provide weekly drought forecast and warning.This study was conducted to validate the accuracy of model for assessing upland drought.As a result of comparing SAWC estimated at given reference date (SAWC e ) and SAWC predicted one week before for the reference date (SAWC p ) on a weekly basis, it was demonstrated that the prediction of model was reliable, with r 2 = 0.91 (P < 0.001) and NSE = 0.89.The differences between SAWC e and SAWC p ranged from -18.5% to +29.2%, attributed to variations in evapotranspiration and water stress coefficient.Comparing the measured SAWC collected from soybean fields in five regions across the country (SAWC m ) and SAWC e , the r 2 values ranged from 0.54 to 0.78, and the RMSE ranged from 12.9 to 27.4.The differences between SAWC m and SAWC e ranged from -6.3% to +46.4%.Our results suggest that to reduce the error and improve the accuracy of upland drought assessment model, it is important to estimate evapotranspiration more precisely, additionally, to consider wide range of factors including soil structure and organic matter content.