Creating a South Korean Hail Risk Map Based on Data Fusion

Abstract:

Hail damage occurs each year, causing enormous losses to outdoor crops and farm facilities such as greenhouses. There is an urgent need for developing a hail information service system customized to the distribution of crop-producing districts and crop growth. This study created a hail risk map by establishing a hail precipitation database including available data from both eye-witness observations from the Korea Meteorological Administration and newspaper articles. The newspaper article data was collected to supplement missing data from regions where the Korea Meteorological Administration was not able to record eye-witness observations. In addition, the frequency and severity of hail occurrences were analyzed in terms of both time and space. The data used for this study included records of hail occurrence from eye-witness observations from the Korea Meteorological Administration from 1904 to 2018 and newspaper articles related to hail from 1990 to 2018, which are available in the BIGKinds service of the Korea Press Foundation. There were 86 Korea Meteorological Administration monitoring stations and 524 hail-related articles registered in BIGKinds. According to the preliminary results, supplementing the data-void areas with newspaper articles greatly reduced the amount of missing data compared with conducting the analysis using article data only. Hail occurred more frequently in the capital region, North Gyeongsang Province, and Jeju Island than any other regions, and the distribution of hail with large diameters increased as the data became more recent. An examination of the monthly hail frequency, excluding the winter season (December, January, and February), based on the Korea Meteorological Administration data showed that the frequency was the highest in November, followed by March. In contrast, the hail frequency analysis using the newspaper article data recorded that the most hail occurred in May and June. The authors plan to complete the creation of the hail risk map of South Korean by supplementing the hail database and considering the overlap of locations and occurrence timing.