Soil water content (SWC) plays a pivotal role in agriculture. The use of atmospheric data to predict soil water content in real-time will greatly facilitate agricultural management and improve agricultural production. In this study, we applied an SWC model that includes three components (i.e., seasonal variations, environmental factors, and long-term trends) and used meteorological data as inputs to predict SWC at different depths (20, 40, 60, 80, and 100 cm) at a soil water content monitoring location in Shimabara, Japan. The prediction model employs singular spectrum analysis (SSA) and nonlinear least squares fitting (NLSF). This model achieved accurate predictions at soil depths of 20, 40, and 60 cm. We concluded that the SWC in the study area exhibits strong seasonal variations and that precipitation has a more significant impact than other variables. The impact of global warming in this area is not dramatic over a short period of time. The results of this study demonstrate the promising application of the model for predicting the SWC necessary for agriculture management and production.