Evaluation of the Use of Rain Gauge Data for Improving Quantitative Precipitation Estimation in South Korea's Mountainous Areas

Accurate and high-resolution rainfall data are fundamental to effective hydrological modeling and flood forecasting, especially as the frequency and intensity of extreme rainfall events increase due to climate change. In South Korea, radar-based quantitative precipitation estimates (QPE) and rain gauge networks are the primary tools for rainfall observation. However, both systems have limitations. Radar-derived rainfall is susceptible to topographic and beam-blocking biases, particularly in complex mountainous terrain, while the rain gauge network maintained by the Korea Meteorological Administration (KMA) is often sparse in these regions, reducing the ability to calibrate or correct radar data effectively. This study assesses the applicability of supplementary rain gauge data from the Korea Forest Service (KFS) in improving radar rainfall estimates over South Korea's mountainous areas. These areas are hydrologically significant, as many major river systems originate in high-elevation zones. The integration of KFS gauge data allows for a more refined spatial correction of radar rainfall fields by filling observational gaps and enhancing the representativeness of ground truth data. We conducted a comparative analysis using gauge-adjusted radar rainfall and independent rain gauge observations from K-water to evaluate performance. The results showed a notable reduction in radar estimation bias, particularly during high-intensity rainfall events, which are often underestimated in mountainous regions due to radar limitations. The integration of additional gauge data improved both the accuracy and reliability of rainfall inputs for hydrological applications. Our findings underscore the importance of expanding ground-based observational networks in complex terrain and demonstrate the effectiveness of multi-source data integration for enhancing QPE in regions vulnerable to topographic distortion. This approach offers valuable insights for developing more resilient flood forecasting systems and supports adaptation strategies in the context of a changing climate.