绿色环境暴露对居民心理健康的影响研究——以南京为例

doi:10.18306/dlkxjz.2020.05.007

摘要/Abstract

摘要：

绿色环境暴露对心理健康的影响长期以来都受到国内外不同领域学者的关注,但从个体与环境交互的角度关注衡量个体对环境实际感知的研究较少。论文试图基于对居民视觉感知和时空活动等因素的考虑,结合机器学习等技术扩展绿色环境暴露的测度方式,并构建绿色环境暴露对心理健康影响的概念框架。同时,以南京为实证对象,运用结构方程模型对比分析绿地率、绿视率、绿色视觉暴露对心理健康影响的差异。研究发现,3种绿色环境暴露测度指标对心理健康均有显著正向影响,但影响程度和路径存在差异,建立更加综合的绿色环境暴露评价指标体系至关重要。主观建成环境可以作为绿视率和绿色视觉暴露对心理健康影响的中介变量,身体活动仅作为绿色视觉暴露影响心理健康的中介变量。研究拓展了绿色环境暴露对心理健康影响的研究框架,并对城市绿地系统的规划管理具有参考价值。

关键词: 环境暴露, 绿色视觉暴露, 绿视率, 心理健康, 机器学习, 南京

Abstract:

The impact of greenspace exposure on mental health has long been the focus of scholars in different fields in China and internationally. Most studies are based on a single individual perspective, such as activity time, range of activities, and so on, or a single environmental perspective, such as the number of green spaces, green space accessibility, among others. Few studies have measured individuals' perception of the environment from the perspective of the interaction between the individuals and the environment. This study constructed a conceptual framework for the impact of green environment exposure on mental health based on considerations of residents' visual perception and spatiotemporal activities. In addition, this study also proposed a green visual exposure measurement method based on green view index and individual spatiotemporal activity, and estimated the green rate using convolutional neural network model and machine learning. This study took Nanjing City as an empirical research object, and used structural equation modeling to compare and analyze the differences in the impact of green rate, green view index, and total green visual exposure on mental health. In addition to observing the direct influences, environmental perception, physical activity, and sense of belongingness were selected as mediating variables to analyze the pathways of different indicators that affect mental health. The results show that the three greenspace exposure measurement indicators have a significant correlation with mental health, but the degree of influence and pathway are different. It is important to establish a more comprehensive green environment exposure evaluation index system. Subjective built-up environment perception can be used as a mediator of the impact of green view index and total green visual exposure on mental health. Physical activity only serves as a mediator of the impact of the total amount of green visual exposureon mental health. This study expands the research framework of the impact of greenspace exposure on mental health, and has important reference value for the planning and management of urban greenspace system.

Key words: environmental exposure, green visual exposure, green view index, mental health, machine learning, Nanjing City

李智轩, 何仲禹, 张一鸣, 金霜霜, 王雪梅, 朱捷, 刘师岑. 绿色环境暴露对居民心理健康的影响研究——以南京为例[J]. 地理科学进展, 2020, 39(5): 779-791.

LI Zhixuan, HE Zhongyu, ZHANG Yiming, JIN Shuangshuang, WANG Xuemei, ZHU Jie, LIU Shicen. Impact of greenspace exposure on residents’ mental health: A case study of Nanjing City[J]. PROGRESS IN GEOGRAPHY, 2020, 39(5): 779-791.

图/表 13

图1

图2

表1

表2

图3

表3

图4

表4

表5

表6

居民个人及家庭属性与模型其余变量的交叉分析

变量名称	心理健康	绿色视觉暴露	身体活动总量	归属感		主观建成环境
变量名称	心理健康	绿色视觉暴露	身体活动总量	社区	南京	美观度	安全感	舒适度	安静感
性别
男	24.24(4.93)	0.47(0.15)	0.28(0.24)	2.57(0.88)	3.02(0.93)	2.80(1.17)	2.56(1.07)	2.35(0.97)	2.70(1.13)
女	24.64(4.98)	0.41(0.13)	0.29(0.26)	2.35(0.94)	2.92(1.06)	2.57(1.08)	2.37(1.07)	2.53(0.96)	2.53(1.13)
年龄
18岁及以下	23.32(5.59)	0.34(0.18)	0.33(0.26)	2.35(1.11)	2.70(1.01)	2.30(1.20)	2.35(1.20)	2.50(1.16)	2.75(1.22)
19~59岁	24.34(4.84)	0.45(0.20)	0.26(0.23)	2.51(0.88)	2.98(0.96)	2.70(1.08)	2.46(1.03)	2.48(0.94)	2.70(1.60)
60岁及以上	25.29(5.10)	0.44(0.20)	0.39(0.30)	2.23(0.94)	3.04(1.12)	2.77(1.30)	2.53(2.24)	2.23(1.03)	2.18(1.11)
学历
本科以下	24.65(4.97)	0.44(0.19)	0.35(0.28)	2.36(0.96)	2.91(1.07)	2.78(1.19)	2.50(1.14)	2.46(1.03)	2.62(1.17)
本科及以上	24.25(4.92)	0.43(0.15)	0.23(0.20)	2.55(0.86)	3.02(0.91)	2.61(1.07)	2.43(1.01)	2.43(0.92)	2.61(1.09)
家庭年收入
10万元以下	24.19(4.63)	0.45(0.18)	0.30(0.27)	2.50(0.92)	2.97(1.04)	2.69(1.15)	2.46(1.09)	2.43(0.93)	2.47(1.11)
10万~20万元	24.84(4.94)	0.47(0.17)	0.31(0.25)	2.36(0.85)	2.92(0.95)	2.66(1.11)	2.44(1.06)	2.48(1.04)	2.77(1.21)
20万~50万元	23.68(5.46)	0.37(0.14)	0.24(0.20)	2.43(0.93)	2.95(0.94)	2.60(1.10)	2.52(1.10)	2.37(0.97)	2.60(1.00)
50万元以上	27.11(4.70)	0.39(0.13)	0.23(0.21)	2.94(1.09)	3.41(1.00)	3.24(1.20)	2.47(1.07)	2.59(0.87)	2.88(1.11)
是否拥有汽车
有小汽车	24.70(4.81)	0.44(0.20)	0.32(0.27)	2.51(0.90)	2.99(1.00)	2.60(1.08)	2.47(1.05)	2.43(0.94)	2.70(1.13)
无小汽车	24.03(5.15)	0.43(0.19)	0.26(0.24)	2.38(0.92)	2.94(0.99)	2.82(1.20)	2.45(1.12)	2.47(1.02)	2.49(1.12)
总体	24.44(4.95)	0.44(0.20)	0.29(0.25)	2.46(0.91)	2.97(0.99)	2.69(1.13)	2.46(1.07)	2.44(0.97)	2.62(1.13)

表6

表7

模型拟合指数

模型	$χ 2 / DF$	GFI	CFI	RMESA
模型1	2.610	0.908	0.842	0.067
模型2	2.768	0.905	0.832	0.070
模型3	2.596	0.910	0.849	0.067
参考值	1~3	>0.90	>0.90	<0.08

表7

表8

图5

参考文献 68

[1]	李小平 . 一个新的交叉学科:环境暴露学 [J]. 国外医学地理分册, 2016,37(2):81-84.
	[ Li Xiaoping . Environmental exposure science: A new inter discipline. Foreign Medical Science Section of Medgeography, 2016,37(2):81-84. ]
[2]	Nieuwenhuijsen M J . Urban and transport planning, environmentalexposures and health-new concepts, methodsand tools to improve health in cities [J]. Environmental Health, 2014,15:161-171.
[3]	马静, 柴彦威, 符婷婷 . 居民时空行为与环境污染暴露对健康影响的研究进展 [J]. 地理科学进展, 2017,36(10):1260-1269. doi: 10.18306/dlkxjz.2017.10.008
	[ Ma Jing, Chai Yanwei, Fu Tingting . Progress of research on the health impact of people's space-time behavior and environmental pollution exposure. Progress in Geography, 2017,36(10):1260-1269. ] doi: 10.18306/dlkxjz.2017.10.008
[4]	张莉, 王五一, 廖永丰 . 城市空气质量健康风险评估研究进展 [J]. 地理科学进展, 2006,25(3):39-47. doi: 10.11820/dlkxjz.2006.03.005
	[ Zhang Li, Wang Wuyi, Liao Yongfeng . Research progress on urban air quality health risk assessment. Progress in Geography, 2006,25(3):39-47. ] doi: 10.11820/dlkxjz.2006.03.005
[5]	Sugiyama T, Carver A, Koohsari M J , et al. Advantages of public green spaces in enhancing population health [J]. Landscape and Urban Planning, 2018,178:12-17. doi: 10.1016/j.landurbplan.2018.05.019
[6]	王兰, 廖舒文, 王敏 . 影响呼吸系统健康的城市绿地空间要素研究: 以上海市某中心区为例 [J]. 城市建筑, 2018,24(3):10-14.
	[ Wang Lan, Liao Shuwen, Wang Min . The impact of spatial factors of urban green apace on respiratory health: A case of a central district in Shanghai. Urbanism and Architecture, 2018,24(3):10-14. ]
[7]	Boers S, Hagoort K, Scheepers L , et al. Does residential green and blue space promoterecovery in psychotic disorders? A cross-sectionalstudy in the province of Utrecht, The Netherlands [J]. International Journal of Environmental Research and Public Health, 2018,15:1-8. doi: 10.3390/ijerph15010001
[8]	Schwartz R M, Christina N, Gillezeau, et al. Longitudinalimpact of hurricane sandy exposure onmental health symptoms [J]. International Journal of Environmental Research and Public Health, 2017,14:1-12. doi: 10.3390/ijerph14010001
[9]	Dzhambov A M, Markevych I, Hartig T , et al. Multiple pathways link urban green- and bluespace to mental health inyoung adults [J]. Environmental Research, 2018,166:223-233.
[10]	Wang W, Lin Z Y, Zhang L W , et al. Building visual green index: A measure of visual green spaces for urbanbuilding [J]. Urban Forestry & Urban Greening, 2019,40:335-343.
[11]	Huang J, Deng F, Wu S W , et al. Comparisons of personal exposure to PM_2.5 and CO by different commuting modes in Beijing, China [J]. Science of the Total Environment, 2012,425:52-59.
[12]	陈曦, 冯建喜 . 基于步行性与污染物暴露空间格局比较的建成环境健康效应: 以南京为例 [J]. 地理科学进展, 2019,38(2):296-304.
	[ Chen Xi, Feng Jianxi . Health effects of built environment based on a comparison of walkability and air pollution: A case study of Nanjing City. Progress in Geography, 2019,38(2):296-304. ]
[13]	Semenza J C, M. P. H, Bubin C H, et al. Heat-related deaths during the July 1995 heat wave in Chicago [J]. The New England Journal of Medicine, 1996,335(2):84-90.
[14]	Hajat S, Kovats R S, Lachowycz K . Heat-related and cold-related deaths in England and Wales: Who is at risk? [J]. Occupational and Environmental Medicine, 2007,64:93-100.
[15]	谢盼, 王仰麟, 彭建 , 等. 基于居民健康的城市高温热浪灾害脆弱性评价: 研究进展与框架 [J]. 地理科学进展, 2015,34(2):165-174.
	[ Xie Pan, Wang Yanglin, Peng Jian , et al. Health related urban heat wave vulnerability assessment: Research progress and framework. Progress in Geography, 2015,34(2):165-174. ]
[16]	Bowler D E, Buyung-Ali L M, Knight T M, et al. A systematic review of evidence for the added benefits to health of exposure to natural environments [J]. BMC Public Health, 2010,456:1-10.
[17]	Coon J T, Boddy K, Stein K , et al. Does participating in physical activity in outdoor natural environments have a greater effect on physical and mental wellbeing than physical activity indoors? A systematic review [J]. Community Health, 2011,65:A1-A40.
[18]	Capaldi C A, Dopko R L, Zelenski J M . The relationship between nature connectedness andhappiness: A meta-analysis [J]. Frontiers in Psychology, 2014,5:1-15.
[19]	Bodin M, Hartig T . Does the outdoor environment matter for psychologicalrestoration gained through running? [J]. Psychology of Sport and Exercise, 2003,4:141-153.
[20]	Kerr J H, Fujiyama H, Sugano A , et al. Psychological responses to exercising in laboratoryand natural environments [J]. Psychology of Sport and Exercise, 2006,7:345-359.
[21]	Johnson B S, Malecki K M , et al. Exposure to neighborhood green space and sleep: Evidence from the survey of the health of Wisconsin [J]. Sleep Health, 2018,4:413-419.
[22]	Gidlow C, Smith G, Kempen E V , et al. Development of the Natural Environment Scoring Tool (NEST) [J]. Urban Forestry & Urban Greening, 2017,29:322-333.
[23]	谭冰清, 武书帆, 苏世亮 , 等. 城市公共绿地供给与居民健康的空间关联 [J]. 城市建筑, 2018,24(8):57-61.
	[ Tan Bingqing, Wu Shufan, Su Shiliang , et al. Spatial association of urban public green space supply and residents' health. Urbanism and Architecture, 2018,24(8):57-61. ]
[24]	姚崇怀, 李德玺 . 绿容积率及其确定机制 [J]. 中国园林, 2015,31(9):5-11.
	[ Yao Chonghuai, Li Dexi . Study on the Green Volume Ratio (GVR) and its calculating mechanism. Chinese Landscape Architecture, 2015,31(9):5-11. ]
[25]	肖希, 韦怡凯, 李敏 . 日本城市绿视率计量方法与评价应用 [J]. 国际城市规划, 2018,33(2):98-103.
	[ Xiao Xi, Wei Yikai, Li Min . The method of measurement and applications of visible green index in Japan. Urban Planning International, 2018,33(2):98-103. ]
[26]	青木陽二. 視野の広がりと緑量感の [J]. 造園雑誌, 1987,51(1):1-10.
	[ Yoji Aoki . Relationship between percieved greenery and width of visual fields. The Japanese Institute of Landscape Architecture, 1987,51(1):1-10. ]
[27]	古澤, 加藤. ~真夏日の不快感を緩和する都市の緑の景観・心理効果について~都市の緑量と心理的効果の相関関係の社会実験調査について [EB/OL]. 国土交通省. 2005- 08- 12 [2020-02-13]. http://www.mlit.go.jp/kisha/kisha05/04/040812_3/01.pdf.
	[ Furusawa, Kato. About the green landscape and psychological effects of the city to alleviate the discomfort of midsummer sun: Social experiment research on the correlation between urban greenness and psychological effects. Ministry of Land, Infrastructure, Transport and Tourism, 2005-08-12[2020-02-13]. http://www.mlit.go.jp/kisha/kisha05/04/04 0812_3/01.pdf. ]
[28]	矢川麻紀子, 田村明弘. 感覚時間による場と人との交感作用の指標化 [J]. 日本建築学会枝術報告集, 1999(8):155-158.
	[ Makiko Vagawa, Akihiro Tamura . Human's sense of time: As an index of the sympathy between man and environment. Architectural Institute of Japan, 1999(8):155-158. ]
[29]	藤井健史, 山田悟史, 廣瀬徳郎等. CG モデルによる全方位緑視率の計量手法とその応用可能性 [R]. 日本建築学会枝術報告集, 2013,19(43):1067-1072.
	[ Takeshi Fujii, Satoshi Yamada, Tokuro Hirose , et al. Measurement of the ratio of visible green spaces in the omnidirectional field of vision using CG models and their potential applications. Architectural Institute of Japan, 2013,19(43):1067-1072. ]
[30]	赵庆, 唐洪辉, 魏丹 , 等. 基于绿视率的城市绿道空间绿量可视性特征 [J]. 浙江农林大学学报, 2016,33(2):288-294. doi: 10.11833/j.issn.2095-0756.2016.02.014
	[ Zhao Qing, Tang Honghui, Wei Dan , et al. Spatial visibility of green areas of urban greenway using the green appearance percentage. Journal of Zhejiang Agriculture and Forestry University, 2016,33(2):288-294. ] doi: 10.11833/j.issn.2095-0756.2016.02.014
[31]	李凤霞, 石辉, 撒利伟 , 等. 微分法三维激光点云数据单木三维绿量测定研究 [J]. 西安建筑科技大学学报(自然科学版), 2017,49(4):530-535.
	[ Li Fengxia, Shi Hui, Sa Liwei , et al. 3D green volume measurement of single tree using 3D laser point cloud data and differential method. Journal of Xi'an University of Architecture & Technology, 2017,49(4):530-535. ]
[32]	崔喆, 何明怡, 陆明 . 基于街景图像解译的寒地城市绿视率分析研究: 以哈尔滨为例 [J]. 中国城市林业, 2018,16(5):38-42.
	[ Cui Zhe, He Mingyi, Lu Ming . An analysis of green view index in cold region city: A case study of Harbin. Journal of Chinese Urban Forestry, 2018,16(5):38-42. ]
[33]	徐磊青, 孟若希, 黄舒晴 , 等. 疗愈导向的街道设计: 基于VR实验的探索 [J]. 国际城市规划, 2019,34(1):38-45.
	[ Xu Leiqing, Meng Ruoxi, Huang Shuqing , et al. Healing oriented street design: An exploration based on VR experiments. Urban Planning International, 2019,34(1):38-45. ]
[34]	Kwan M . From place-based to people-based exposure measures [J]. Social Science & Medicine, 2009,69:1311-1313.
[35]	Grahn P, Stigsdotter U A. Grahn P, et al. Landscape planning and stress [J]. Urban Forestry & Urban Greening, 2003,2(1):1-18.
[36]	Nielsen T S, Hansen K B . Do green areas affect health? Results from a Danish survey onthe use of green areas and health indicators [J]. Health & Place, 2006,13:839-850.
[37]	Berg M V D, Poppel M V P, Smith G , et al. Does time spent on visits to green space mediate the associations betweenthe level of residential greenness and mental health? [J]. Urban Forestry & Urban Greening, 2017,25:94-102.
[38]	Zhang L, Zhou S H, Kwan M , et al. Impacts of individual daily greenspace exposure on health based on individual activity space and structural equation modeling [J]. International Journal of Environmental Research and Public Health, 2018,15:1-18.
[39]	Gascon M, Triguero-Mas M, Martínez D , et al. mental health benefits of long-term exposure to residential green and blue spaces: A systematic review [J]. International Journal of Environmental Research and Public Health, 2015,12(4):4354-4379.
[40]	Paluska S A, Schwenk T L . Physical activity and mental health [J]. Sports Medicine, 2000,29(3):167-180.
[41]	Evans W P, Owens P, Marsh S C , et al. Environmental factors, locus of control, and adolescent suicide risk [J]. Child and Adolescent Social Work Journal, 2005,22(3):301-319.
[42]	Rappe E, Kivelä S, Rita H . Visiting outdoor green environments positively impacts self-rated health among older people in long-term care [J]. Hort Technology, 2006,16(1):55-59.
[43]	Akpinar A, Barbosa-Leiker C, Brooks K A . Does green space matter? Exploring relationships between greenspacetype and health indicators [J]. Urban Forestry & Urban Greening, 2016,20:407-418.
[44]	Helbich M, Klein N, Roberts H , et al. More green space is related to less antidepressant prescription rates in the Netherlands: A Bayesian geoadditive quantile regression approach [J]. Environmental Research, 2018,166:290-297.
[45]	Liu Q Y, Zhang Y J, Lin Y W , et al. The relationship between self-rated naturalness of university green space and students' restoration and health [J]. Urban Forestry & Urban Greening, 2018,34:259-268.
[46]	Thompson C W, Roe J, Aspinall P , et al. More green space is linked to less stress in deprived communities:Evidence from salivary cortisol patterns [J]. Landscape and Urban Planning, 2012,105:221-229.
[47]	Yuan L, Shin K, Managi S . Subjective well-being and environmental quality: The impact of air pollution and green coverage in China [J]. Ecological Economics, 2018,153:124-138.
[48]	Pietilä M, Neuvonen M, Borodulin K , et al. Relationships between exposure to urban green spaces, physicalactivity and self-rated health [J]. Journal of Outdoor Recreation and Tourism, 2015,10:44-54.
[49]	Lachowycz K, Jones A P . Towards a better understanding of the relationship between greenspaceand health: Development of a theoretical framework [J]. Landscape and Urban Planning, 2013,118:62-69.
[50]	Reid C E, Kubzansky L D, Li J , et al. It's not easy assessing greenness: A comparison of NDVI datasets and neighborhood types and their associations with self-rated health in New York City [J]. Health & Place, 2018,54:92-101.
[51]	Dadvand P, Bartoll X, Basagaña X , et al. Green spaces and general health: Roles of mental health status, socialsupport, and physical activity [J]. Environment International, 2016,91:161-167. doi: 10.1016/j.envint.2016.02.029
[52]	Lee S, Ellis C D, Kweon B , et al. Relationship between landscape structure and neighborhoodsatisfaction in urbanized areas [J]. Landscape and Urban Planning, 2008,85:60-70. doi: 10.1016/j.landurbplan.2007.09.013
[53]	Lu Y . Using google street view to investigate the association between streetgreenery and physical activity [J]. Landscape and Urban Planning, 2018,8:2-9.
[54]	Rzotkiewicz A, Pearson A L, Dougherty B V , et al. Systematic review of the use of google street view in health research: Major themes, strengths, weaknesses and possibilities for future research [J]. Health & Place, 2018,52:240-246.
[55]	小宫佑登, 須﨑純一. 航空機計測を用いた市街地における緑視率推定 [J]. 土木学会論文集D1, 2015,71(1):1-9.
	[ Yuto KOMIYA, Junichi SUSAKI . Estimation of visible green index in urban area using airborne measurement. Journal of Japan Society of Civil Engineers, Ser. D1, 2015,71(1):1-9. ]
[56]	Li X J, Ratti C, Seiferling I. Mapping urban landscapes along streets using google street view[C]// Peterson M P. Advances in cartography and GIScience: Selections from the international cartographic conference 2017. Washington D C, USA: Springer, Cham, 2017: 341-356.
[57]	Helbich M, Yao Y, Liu Y , et al. Using deep learning to examine street view green and blue spaces and their associations with geriatric depression in Beijing, China [J]. Environment International, 2019,126:107-117.
[58]	彭锐, 刘海霞 . 城市道路绿视率自动化计算方法研究 [J]. 北京规划建设, 2018(4):61-64.
	[ Peng Rui, Liu Haixia . Research on automatic calculation method of green line rate of urban roads. Beijing Planning Review, 2018(4):61-64. ]
[59]	卢官明, 何嘉利, 闫静杰 , 等. 一种用于人脸表情识别的卷积神经网络 [J]. 南京邮电大学学报, 2016,36(1):16-22.
	[ Lu Guanming, He Jiali. Yan Jingjie , et al. Convolutional neural network for facial expression recognition. Journal of Nanjing University of Posts and Telecommunications, 2016,36(1):16-22. ]
[60]	崔璐, 张鹏, 车进 . 基于深度神经网络的遥感图像分类算法综述 [J]. 计算机科学, 2018,45(6A):50-53.
	[ Cui Lu, Zhang Peng, Che Jin. Overview of deep neural network based classification algorithms for remote sensing images. Computer Science, 2018,45(6A):50-53. ]
[61]	何宛余, 李春, 聂广洋 , 等. 深度学习在城市感知的应用可能: 基于卷积神经网络的图像判别分析 [J]. 国际城市规划, 2019,34(1):8-17.
	[ He Wanyu, Li Chun, Nie Guangyang, et al. The latent application of deep learning in urban perception: Image discrimination analysis by convolutional neural network. Urban Planning International, 2019,34(1):8-17. ]
[62]	解自来 . 提高主城“绿视率”是建设“绿色南京”的重要环节 [J]. 现代城市研究, 2003(6):83-86.
	[ Xie Zilai . Raising the proportion of greenness of main city is an important tacheto build greening Nanjing. Modern Urban Research, 2003(6):83-86. ]
[63]	Arango C M, Paez D C, Reis R S , et al. Association between the perceived environment and physical activity among adults in Latin America: A systematic review [J]. International Journal of Behavioral Nutrition and Physical Activity, 2013,10:122. doi: 10.1186/1479-5868-10-122.
[64]	冯建喜, 黄旭, 汤爽爽 . 客观与主观建成环境对老年人不同体力活动影响机制研究: 以南京为例 [J]. 上海城市规划, 2017(3):17-23.
	[ Feng Jianxi, Huang Xu, Tang Shuangshuang . Comparing the influences of objective and subjective built environments on physical activities participation among the elderly: A case study of Nanjing, China. Shanghai Urban Planning Review, 2017(3):17-23. ]
[65]	姜玉培, 甄峰, 王文文 , 等. 城市建成环境对居民身体活动的影响研究进展与启示 [J]. 地理科学进展, 2019,38(3):357-369.
	[ Jiang Yupei, Zhen Feng, Wang Wenwen , et al. Influence of urban built environment on residents' physical activity: Review and implications. Progress in Geography, 2019,38(3):357-369. ]
[66]	王宁, 杜豫川 . 社区居民适宜步行距离阈值研究 [J]. 交通运输研究, 2015(2):20-24, 30.
	[ Wang Ning, Du Yuchuan . Resident walking distance threshold of community. Transport Research, 2015(2):20-24, 30. ]
[67]	温忠麟, 张雷, 侯杰泰 , 等. 中介效应检验程序及其应用 [J]. 心理学报, 2004,36(5):614-620.
	[ Weng Zhonglin, Zhang Lei, Hou Jietai , et al. Testing and application of the mediating effects. Acta Psychologica Sinica, 2004,36(5):614-620. ]
[68]	温忠麟, 叶宝娟 . 中介效应分析: 方法和模型发展 [J]. 心理科学进展, 2014,22(5):731-745.
	[ Wen Zhonglin, Ye Baojuan . Analyses of mediating effects: The development of methods and models. Advances in Psychological Science, 2014,22(5):731-745. ]

层级结构	卷积核尺寸	输出图片尺寸
Input		224×224×3
Conv2D	3×3	222×222×32
MaxPooling2D	2×2	111×111×32
Dropout		111×111×32
Conv2D	3×3	109×109×64
MaxPooling2D	2×2	54×54×64
Dropout		54×54×64
Conv2D	3×3	52×52×128
AveragePooling2D	2×2	26×26×128
Dropout		26×26×128
Conv2D	1×1	26×26×128
Dropout		26×26×128
Flatten		86528
Dense		256
Dropout		256
Dense		1

小区名称	建设年代	房价/元	小区属性	距市中心距离/km	楼层	有效问卷数
五福里	1949年前	3万以下	房改房/公房	0~5	低层	35
芳草园	1949—1998年	4万以上	商品房	0~5	高层	65
居安里	1949—1998年	3万~4万	房改房/公房	0~5	多层	63
海棠园	1999年后	4万以上	商品房	5~10	多层	64
莲花新城	1999年后	3万以下	安置房/经适房	5~10	高层	50
云燕园	1949—1998年	3万以下	房改房/公房	5~10	多层	40
银城东苑	1999年后	3万~4万	商品房	5~10	高层	42

小区名称	小区周边绿地率/%	小区周边绿视率/%	有效观测点数量/个	有效图片数量/张
五福里	15.07	24.00	2095	4282
芳草园	6.60	28.14	1017	2092
居安里	14.28	21.37	4305	10505
海棠园	12.36	29.53	1879	4096
莲花新城	15.51	16.50	1349	2128
云燕园	22.17	21.36	825	1663
银城东苑	10.26	35.81	855	1379

居民活动路线	出行时长/min	路线平均绿视率/%
路线A	10	24.12
路线B	15	25.73
路线C	6	23.15
路线D	13	22.23

变量类别	潜变量	测量变量	测量变量类型及取值
解释变量	绿色环境暴露	小区周边绿地率(模型1)	连续变量
		小区周边绿视率(模型2)	连续变量
		居民绿色视觉暴露(模型3)	连续变量(0~1)
中介变量	主观建成环境	我认为小区及周边绿化充足	有序分类变量(1~5)
		我认为在小区及周边步行是安全的
		我认为在小区及周边步行是舒适的
		我认为小区及周边环境安静
	身体活动	IPAQ中文版短表周身体活动总量	连续变量(0~1)
	归属感	我对现在居住的小区有归属感	有序分类变量(1~5)
		我对南京有归属感	有序分类变量(1~5)
因变量	心理健康	SWEMWBS中文版心理健康评分	有序分类变量(7~35)