A Study on the Segmentation of Heat Wave Warning Zones using Cluster Analysis

Abstract

A heat wave countermeasure period was selected and operated to minimize the damage from heat waves in Korea. However, to this date, there are no clear ideas and methodology for selecting a heat wave countermeasure period, resulting in continuous budget expenditure incurred for its preparation and response activities. This requires an efficient selection of countermeasure periods. This study analyzed the time series data trend of heat index using Mann-Kendall and Pettitt test for cities and counties in Gyeongsangbuk-do, and it confirmed the current characteristics of increasing heat waves compared to- those in the past. In addition, K-means cluster analysis was applied to derive the city-wise heat wave characteristics. The cities and counties were classified into three based on the heatwave characteristics and starting point and ending point of each heat wave characteristic representing the years 2004 to 2021 between July 21 to August 11, July 16 to August 15, and July 11 to August 19. It was also confirmed that there was a significant difference from the heat wave countermeasure period (May 20~Sep 30) currently selected. Hence, in the future, economic losses may be minimized in terms of disaster preparedness and response through the selection of the heatwave countermeasure period reflecting temporal and spatial characteristics.