Snow Damage Prediction Model Using Socioeconomic Factors

Abstract

Due to the climate change, the natural disasters have been occurred more frequently and the amount of damage has been increased as well. In South Korea, the number of snow disaster has been increased recently. In last 20 years, from 1994 to 2013, total snow damage was 1.3 billion dollars. In this study, the snow damage was estimated using historical damage data to response the possible heavy snow and mitigate the damage. The historical snow damage data from Annual Natural Disaster Report for the last 22 years were used to develop a multiple linear regression model. Input data for the model were daily maximum snow depth (or daily maximum fresh snow depth), daily highest temperature, daily lowest temperature, daily average temperature, relative humidity as meteorological factors, and also regional area, greenhouse area, number of farmers, number of farmers over age 60 were considered as socioeconomic factors. The developed model was applied in Jeolla-do, Chungcheong-do, and Kangwon-do which have the largest snow damage in the history. As the results, the model showed over 70% of accuracy in all of three cities.