Spatial Clustering of Water Pollution Based on Spatial Autocorrelation Analysis

The spatial autocorrelation analysis was performed to quantitatively analyze the spatial distribution characteristics of water pollution monitoring indices, based on China Status of Environment in 2012. Eight water pollution indices, including dissolved oxygen, potassium permanganate index, BOD₅, ammonia nitrogen, petroleum, volatile phenol, mercury and lead. The Moran’s I is greater than 0 for all pollution indices, indicating that the spatial distribution of the main pollutants in China has significant spatial autocorrelation. The result of Moran scatter diagram of LISA shows that the pollutants have developed several kinds of spatial patterns. In general, the distribution of the water pollutants has close relationship with regional socio-economic situations, with obviously higher pollution in densely populated areas, such as Beijing-Tianjin-Hebei region. The water pollution in China could be divided into two major patterns “east-west” and “north-south”, namely, the water pollution in China's eastern coastal areas were significantly higher than that in the midwest, and the water pollution in northern China was worse than that in southern region. Different water pollution indices had different agglomeration areas.