Spatial effects on extreme precipitation in the coastal areas of southeastern China during the raining season

This study investigated the spatial effects of the extreme precipitation during the raining season in the perspective of tropical cyclone-related (TCEP) and non-TC-related extreme precipitation (nTCEP). The seasonal maximum 1-day precipitation (RX1day) data across 94 stations in the coastal areas of southeastern China from 1964 to 2013 were used and partitioned into different homogenous regions. Regional Bayesian models were developed for generalized extreme value (GEV) distribution to quantify the spatial effects on extreme precipitation in each homogeneous region. Four coastal and one inland homogeneous regions were identified, which were consistent with the TC tracks. The extreme precipitation was found to decrease as the distance from the coastline increases, whose sensitivity to the distance from the coastline differs from region to region. For TCEP, the distance from the coastline has effects on the location and the scale parameters of the regional GEV model in all regions. The 50-year return level of TCEP in southern regions is five times more sensitive to the distance from the coastline than that in the northern regions. For nTCEP, the distance from the coastline has effects on the scale parameter in all regions, while the location parameter is affected only in two regions in the south. The sensitivity of 50-year return levels of nTCEP to the distance from the coastline decreases gradually from south to north. Within 200 km from the coastline, the 50-year return level of TCEP is 15.6–78.9% larger than that of nTCEP in the same region.