2001～2015年中国城镇化与PM_(2.5)浓度时空关联特征

Analysis of the spatial-temporal association between urbanization and PM_(2.5) concentration during 2001～2015 period in China's Mainland

综合运用时空统计、剪刀差、地理加权回归等手段分5个阶段从人口、土地、经济角度研究了2001～2015年中国大陆地区城镇化水平与PM_(2.5)浓度的时空分布以及相互关联特征.结果表明:2001～2015年,我国3a均人口、土地、经济城镇化水平平均值稳步提高,PM_(2.5)浓度波动上升(44.14～50.89μg/m^3),不同经济区之间时序变化趋势相似但强度存在差异.不同阶段PM_(2.5)浓度空间分布趋势基本一致,京津冀、河南与山东北部、新疆西部等地区始终为高值区,3类城镇化水平高值区域逐步扩大.城镇化水平的时序变化趋势与PM_(2.5)浓度随时间变化的趋势存在差异(切线夹角:15.33°～62.92°),难以解释PM_(2.5)浓度在时间上的突变.城镇化水平与PM_(2.5)浓度空间分布关联特征显著,土地城镇化水平与PM_(2.5)浓度在0.01水平上正相关,其相关系数依次为东北(0.609～0.723)>中部(0.572～0.631)>东部(0.218～0.323)>西部(0.079～0.255),除中、西部地区外,经济城镇化水平对PM_(2.5)浓度正效应明显,东北地区人口城镇化水平与PM_(2.5)浓度负相关,GWR模型调整R^2在2001～2003年阶段最高(0.6～0.77),2013～2015年阶段最低(0.08～0.64).

This study analyzed the spatial and temporal variations of PM2.5 concentration and the degrees of urbanization for the five periods of 2001~2003,2004~2006,2007~2009,2010~2012,and 2013~2015 in China's Mainland,on perspectives of population,land and economy.Scissors difference method,correlation analysis and geographically weighted regression(GWR)model were employed to explore their relationships.Results showed that the three-year average degrees of urbanization for population,land and economy increased steadilyand PM2.5 concentration fluctuated in an increase direction(44.14~50.89μg/m^3).However,rates of these variations were different between four economic regions.Spatial distributions of PM2.5 concentrations were similar for the five periods,they were high in Beijing-Tianjin-Hebei(BTH)area,north Henan and Shandong province,and west Xinjiang.The area with high urbanization degree increased gradually.Angles(15.33°~62.92°)between the two tangent lines of temporal variation curves of degree of urbanization and PM2.5 concentrations based on scissors difference method revealed the insignificant relationships between them temporally.In addition,significant association was found between the spatial distributions of PM2.5 concentrations and the degrees of urbanization.The degrees of urbanization for land were positively correlated with PM2.5 concentrations at the 0.01 level.The sequence of correlation coefficients from high to low was Northeast(0.609~0.723) >Midlands(0.572~0.631) >East(0.218~0.323) >West(0.079~0.255).Except for Midlands and West,the degrees of urbanization for economy had significantly positive effect on PM2.5 concentrations,while those for population were negatively associated with PM2.5 concentrations in Eastern area.The results for GWR models further demonstrated this varied spatial association with adjusted R2 were found to be the highest in 2001~2003 period(0.6~0.77)and the lowest in 2013~2015 period(0.08~0.64).