长三角城市土地利用格局与PM2.5浓度的多尺度关联分析

Multi-scale correlation analysis of urban landscape pattern and PM2.5 concentration in the Yangtze River Delta

PM2.5引发的雾霾污染对人体健康和社会可持续发展产生了严重威胁,已成为中国经济快速发展地区共同面临的问题。长三角是中国城市化进程最快、空气污染最为严重的地区之一,探寻该地区土地利用景观格局变化对PM2.5的影响规律,有助于对PM2.5“源”“汇”景观的空间格局进行合理配置,也可以为污染防治决策提供科学依据。本文运用重心模型、冷热点分析和景观指数,探讨了该区域1995—2015年PM2.5浓度的时空分布特征以及景观格局的变化规律,并使用岭回归方法分析了建设用地、林地、耕地和水体四种土地利用类型的景观格局在行政区尺度和外接圆尺度上对PM2.5浓度的影响。结果显示:①1995—2015年长三角地区PM2.5浓度总体呈上升趋势,并且具有“北高南低”和“南缓北急”空间分异特征。②长三角区域内建设用地面积大幅上升,且呈聚合状发展,而林地和耕地面积却在不断减少,并呈破碎状分布。③建设用地和林地分别是PM2.5的“源”景观与“汇”景观,耕地对PM2.5的“源”“汇”作用交错,水体对PM2.5无明显的净化作用。④相较于行政区尺度,外接圆尺度下林地PLAND、ED与PM2.5浓度的负相关更为显著,可见对城市周边地区进行景观格局优化能收到更好的效果。研究表明:控制建设用地合理有序增长并采用多中心发展模式,有利于缓解城市主城区的环境压力;提高城市周边区域林地的比重和聚集度或加大林地与建设用地的接触面积,可以有效地减少城市PM2.5浓度;对耕地进行整理使其形成连片化景观,并通过科学的耕作方式减少耕地上农业生产所带来的PM2.5前体物,有助于发挥其对PM2.5的“汇”作用。

Haze pollution caused by PM2.5 has threated human health and sustainable development in many rapidly developing regions of China over the past decade. The Yangtze River Delta( YRD) region is one of the rapidly urbanizing regions in China with severe PM2.5 pollution. Therefore,it is critical to explore the impact of urban landscape dynamics on PM2.5 variations in this region to optimize the spatial arrangement of the ‘source’and ‘sink’elements in the landscape. In this paper,we used the gravity model,hotspot analysis,and landscape indices to explore the spatio-temporal variations of PM2.5 concentration and landscape pattern in the YRD region during 1995-2015. We also used the ridge regression model to analyze the effects of landscape pattern of four land use types( including construction land,woodland,cropland,and water body) on PM2.5 concentration at multiple spatial scales. The results showed that ① the PM2.5 concentration increased significantly from 1995 to 2015,especially in the northern part of the YRD.Moreover,the PM2.5 concentration also increased more rapidly in the northern part than in the southern part of the region.②During1995-2015,construction land had an increasing area and aggregation,while woodland and cropland have been decreasing,and their distribution is fragmented.③Construction land and woodland are the‘source’and‘sink’landscape for haze pollution,respectively.Cropland has both effects of‘source’and ‘sink’on PM2.5 concentration. Water body has no obvious effect on haze pollution.④Higher PLAND and ED values of woodland were related with lower PM2.5 concentration at the smaller spatial scale. Based on these findings,we recommend the following land management measures for mitigating PM2.5 pollution for the YRD region: Encouraging the‘leap frogging’pattern of urban growth;distributing and managing woodland near the construction land to increase the contact area between the two land use types;and preserving large cropland patches.