社区建成环境对老年人活力出行的空间异质效应

The impact of spatial heterogeneity in community built environment on the elderly’s active travel

针对社区生活圈老年人步行友好出行环境理论缺乏与实践不足的问题,从社区建成环境对老年人活力出行影响的空间异质性问题展开了研究.结合老年人出行特征选取了人口密度、土地利用多样性、公交站点密度、绿视率等8个指标刻画建成环境,考虑路网非直线系数对社区生活圈尺度差异化界定,利用兴趣点(Point of Interest,POI)数据、道路网络、街景图像数据等地理空间数据测度建成环境.提取每日步行时间作为活力出行的测度,以重庆市居民出行行为调查数据作为实证研究数据来源,构建了考虑自变量尺度变异的多尺度地理加权回归模型(Multi-scale Geographic Weighted Regression,MGWR).研究结果表明:对比最小二乘回归(Ordinary Least Squares,OLS)模型与传统的地理加权回归(Geographically Weighted Regression,GWR)模型,纳入变量尺度异质性的MGWR模型降低了残差的自相关性,且调整后R2最高,相比于GWR模型与OLS模型分别提高了1.5倍与3.1倍;从标准化系数来看,社区建成环境指标中公交服务水平对老年人步行时间影响最大;社区建成环境指标中街道绿视率接近全局尺度,空间异质性较弱,其他建成环境变量均具有较强的空间异质性,需要进行差异化的空间设计,其中人口密度对老年人步行时间在不同区域呈现相反的影响,土地利用混合度、公交服务水平、蔬菜市场密度、公园广场密度与街道绿视率对老年人活力出行的影响呈现正效应.

This paper investigates the spatial heterogeneity,i.e.,impact of the community built envi⁃ronment on the dynamic travel behavior of elderly people,with a focus on the lack of theory and practice of pedestrian-friendly travel environments in community life circles for the elderly.Combined with the travel characteristics of the elderly,eight indicators,including population density,land use diver⁃sity,bus stop density,and green viewing rate,are selected to describe the built environment.By incor⁃porating the non-linear coefficient of the road network,the community life circle scale difference can be accurately defined.Therefore,geospatial data such as Point of Interest(POI)data,road network data,and street image data can be utilized to accurately measure the built environment.Walking time is used as a measure of dynamic travel.The survey data of Chongqing residents’travel behavior is used as the empirical research data source.A Multi-scale Geographically Weighted Regression(MGWR)model is constructed,which considers the scale variation of independent variables.The re⁃sults show that compared to the Ordinary Least Squares(OLS)model and the traditional Geographi⁃cally Weighted Regression(GWR)model,the MGWR model incorporating variable scale heterogene⁃ity reduces the autocorrelation of the residuals,and the adjusted R2 is the highest,which is 1.5 times and 3.1 times higher than the GWR model and the OLS model,respectively.From the standardized coefficient,the public transport service level has the largest impact on the walking time of the elderly in the community built environment indicators.In the community built environment indicators,the street green vision rate is close to the overall scale and has weak spatial heterogeneity,while the other built environment variables have strong spatial heterogeneity,requiring differentiated spatial design.Among them,population density has an opposite impact on the walking time of the elderly,while the land use mix,public transport service level,vegetable market density,park square density,and street green sight rate have a positive effect on the elderly’s active travel.