Abstract: Changes in the magnitude and frequency of extreme rainfall caused by climate change have attracted wide attention around the world. Researching the evolution law of extreme rainfall is of great significance for regional flood control and water resources management. According to rainfall data in 54 years (1960-2013) at 16 meteorological stations in Zhaotong, Yunnan Province and its surrounding areas, we analyzed the spatial evolution of six extreme rainfall indicators recommended by the World Meteorological Organization by linear trend analysis, Mann-Kendall method, and moving-t test, and the spatial distribution pattern by inverse distance interpolation method based on ArcGIS10.2. Results manifest that total rainfall volume indicator PRCPTOT presented a downward trend; three indices reflecting the magnitude of extreme rainfall, namely, SDII, RX5day, and R95p, displayed increasing trends, among which RX5day increased significantly; R10mm and CWD, which reflect the frequency of extreme rainfall, both showed decreasing trends, of which R10mm decreased apparently in particular. In terms of spatial distribution, the six indicators are of high spatial heterogeneity: the magnitude and frequency of rainfall were high in northeast Zhaotong, but alleviated in central Zhaotong, and relatively low in the southwest. In addition, extreme rainfall indicators were negatively correlated with elevation: extreme rainfall mainly happened in area of low altitude.

Key words: climate change, extreme rainfall, spatio-temporal variation, extreme precipitation indices, correlation, Zhaotong City