이상기상과 지형 조건이 신갈나무 고사에 미치는 영향

Abstract

This study quantitatively analyzed tree mortality associated with abnormal weather and topographic conditions in a Quercus mongolica stand located in Minjujisan, Muju, South Korea. Daily weather data from the Mountain Meteorological Network, spanning from November 2020 to February 2024, were used to define mixed precipitation, extreme wind, extreme low temperature, and compound risk conditions. These meteorological indicators were integrated with high-resolution digital elevation model (DEM)-derived topographic variables, including elevation, slope, aspect, topographic wetness index (TWI), topographic exposure index (TOPEX), and solar radiation. The results revealed that dead trees were predominantly distributed in high-altitude and steep-slope areas with low TWI and limited solar radiation. Mortality was concentrated during the 2023–2024 winter, which coincided with frequent compound abnormal weather conditions. Among the predictive models tested, the Random Forest model showed the highest performance (AUC = 0.855), while Generalized Additive Models (GAM) were employed for their interpretability. Interaction analyses demonstrated that combinations such as slope × wind speed and TWI × solar radiation significantly influenced the probability of mortality. These findings highlight that tree mortality can be significantly amplified by the interaction between abnormal weather and topographically vulnerable conditions, and provide a scientific basis for predicting forest damage and formulating climate change adaptation strategies in mountainous regions.