관측자료 기반 격자 생성자료 생성 및 예측모형 연구

Study for Grid Interpolation Data Generation and Prediction Models based on Observed Data

Natural disasters are increasing due to recent climate change. Rainfall directly or indirectly affects aquaculture and agriculture. Therefore, it is important to anticipate and prepare for this. In this study, a study on the performance comparison between spatial interpolation techniques was conducted based on observed data, and predicted based on the data produced by the spatial interpolation techniques. The spatial interpolation methods compared the inverse distance weighting method, kriging and spatial random forest through MAE (Mean absolute error), RMSE (Root mean square error), correlation coefficient. The verification of predicted data was compared through MAE, CSI (Critical Success Index), FAR (False Alarm Ratio), POD (Probability of Detection) and ETS (Equitable Threat Score).