According to the UN's 'World Disaster Report 2000-2019', the number of disasters has increased by 1.7 times compared to the previous 20 years, and the main cause is diagnosed to be climate change. In addition, heavy snow damage, which has a small numb...
According to the UN's 'World Disaster Report 2000-2019', the number of disasters has increased by 1.7 times compared to the previous 20 years, and the main cause is diagnosed to be climate change. In addition, heavy snow damage, which has a small number of damages compared to other natural disasters, is becoming increasingly large-scale due to climate change. It was calculated that 7,348 disasters occurred worldwide over the past 20 years, resulting in 1.23 million deaths and approximately 3,400 trillion won in property damage. In the case of heavy snow, damages amounted to approximately 160 billion won over the past 10 years, making it the third largest natural disaster in Korea. Therefore, research is needed to predict differentiated heavy snow damage considering regional characteristics. In this study, a heavy snow damage risk index was developed using the PSR and DPSIR techniques, and risk levels (Red Zone, Orange Zone, Yellow Zone, Green Zone) were assigned to each administrative district. The heavy snow damage data from 1994 to 2020 were used as dependent variables, and meteorological factors and socio-economic factors were selected as independent variables. A heavy snow damage prediction technology was developed by risk level through multiple regression analysis. The developed prediction technology was evaluated through RMSE and NRMSE, and the prediction technology by risk level that reflected regional characteristics showed excellent performance as a result of the prediction ability evaluation by technology. If a model that predicts the damage range is developed through the results of this study in the future, it is expected that heavy snow damage can be reduced in advance.