Zhao Xiao-Feng, Huang Xian-Jin, Zhang Xing-Yu, Zhu De-Ming, Lai Li, Zhong Tai-Yang
Laboratory of Environmental Geography in Land Resources and Tourism Sciences Department, Nanjing University, Nanjing 210093, China.
Huan Jing Ke Xue. 2009 Jun 15;30(6):1580-7.
Spatial autocorrelation is an effective tool of spatial statistics, which is used to disclose the spatial structure of regional disparity. There are two different scales to measure regional spatial dependence: global spatial autocorrelation and local spatial autocorrelation. Based on environmental data of 13 cities in Jiangsu province from 1990 to 2006, the regional disparity of COD, SO2 and TSP emission was discussed by using spatial autocorrelation analysis methods. The results show that total emission of COD and TSP decreased respectively from 596 353 t and 1 101 404 t in 1990 to 291 762 t and 704734 t in 2006, while total emission of SO2 kept steady. In 2006, Global Moran's I of COD, SO2 and TSP emission was 0.465 7, 0.214 2 and 0.510 1 respectively. It is identified that positive spatial autocorrelation is presented and spatial aggregation pattern of COD, SO2 and TSP emission are appeared. However, spatial aggregation pattern of COD emission appears earlier than that of SO2 and TSP, and spatial aggregation degree of COD is also higher than that of SO2 and TSP. There are different spatial patterns between southern and northern Jiangsu. In southern Jiangsu, Global Moran's I of COD, SO2 and TSP emission had increased to 0.499 7, 0.320 2 and 0.298 3 up to 2006, and spatial aggregation pattern appeared remarkably. In northern Jiangsu, most of the Global Moran's I were less than -0.2, and spatial aggregation pattern disappeared accordingly. High cluster region of COD emission is Suzhou, Wuxi and Changzhou, and high cluster region of SO2 emission is Suzhou and Wuxi. However, spatial pattern of TSP emission does not change much and five cities of southern Jiangsu (Suzhou, Wuxi, Changzhou, Zhenjiang, Nanjing) are still the high cluster region. The last, the research provides an important cognition to regional environment disparity and macro-environmental strategy, and a significant means to harmonious society and eco-province construction in Jiangsu province.
空间自相关是空间统计学的一种有效工具，用于揭示区域差异的空间结构。衡量区域空间依赖性有两种不同尺度：全局空间自相关和局部空间自相关。基于江苏省13个城市1990 - 2006年的环境数据，运用空间自相关分析方法探讨了化学需氧量（COD）、二氧化硫（SO₂）和总悬浮颗粒物（TSP）排放的区域差异。结果表明，COD和TSP的排放总量分别从1990年的596353吨和1101404吨下降到2006年的291762吨和704734吨，而SO₂的排放总量保持稳定。2006年，COD、SO₂和TSP排放的全局莫兰指数（Global Moran's I）分别为0.4657、0.2142和0.5101。研究发现存在正空间自相关，且出现了COD、SO₂和TSP排放的空间集聚格局。然而，COD排放的空间集聚格局比SO₂和TSP出现得更早，且COD的空间集聚程度也高于SO₂和TSP。苏南和苏北存在不同的空间格局。到2006年，苏南地区COD、SO₂和TSP排放的全局莫兰指数分别增至0.4997、0.3202和0.2983，空间集聚格局显著出现。在苏北，大多数全局莫兰指数小于 -0.2，相应地空间集聚格局消失。COD排放的高聚类区域是苏州、无锡和常州，SO₂排放的高聚类区域是苏州和无锡。然而，TSP排放的空间格局变化不大，苏南的五个城市（苏州、无锡、常州、镇江、南京）仍是高聚类区域。最后，该研究为认识区域环境差异和宏观环境战略提供了重要依据，也是江苏省构建和谐社会和生态省建设的重要手段。
