Spatial autocorrelation of three nature disasters in China

China is a country which is strongly struck by disasters, and three natural disasters occurr frequently and distribute widely. Studying their spatial distribution and spatial aggregation is significant to disaster prevention, reduction and relief supplies configuration. The spatial distribution and aggregation of disasters can be showed by spatial autocorrelation analysis on the three natural disasters. In this paper, the spatial distribution of the three natural disasters were got from epicenter distribution map, fault zone distribution map, snow and ice disasters frequency map and flooding frequency map. The spatial autocorrelation distribution maps were got from spatial autocorrelation analysis on the spatial distribution of the three natural disasters. The results showed that the three natural disasters reached significant spatial autocorrelation level because the three Z indices of the global correlation coefficient (Moran’s I) were greater than 2.54 at the 0.01 confidence level. Local autocorrelation analysis showed that the spatial distributions of three natural disasters mainly showed high-high positive autocorrelation(HH) or low-low positive autocorrelation(LL) and little high-low negative autocorrelation(HL) or low-high negative autocorrelation(LH). Three natural disasters occurred in such regions and had their own centers. The earthquake occurred mainly in the southwest region of China and Taiwan Province, the snow and ice disasters mainly in the Inner Mongolia Autonomous Region and Xinjiang Uygur Autonomous Region, floods mainly in the southeast region of China.