How green are residential areas? An analysis of community greening emerging multi-source geo-data

Urban residential greening provides opportunities for social integration and physical exercise. These activities are beneficial to promoting citizens’ mental health, relieving stress and reducing obesity and violent crimes. However, how to measure the distribution and spatial difference of green resources in urban residential areas have been controversial. This study takes the greening of urban residential units in Shenzhen City as its research object, measures the various greening index values of each residential unit, and analyses the spatial distribution characteristics of residential greening, regional differences and influencing factors. Large sample of street view pictures, urban land use and high-resolution remote sensing image data are employed to establish an urban residential greening database containing 14,196 residential units. This study proposes three greening indicators, namely, green coverage index, green view index, and accessible public green land index, for measuring the green coverage of residential units, the visible greening of surrounding street space and the public green land around, respectively. Results show that (1) the greening level of residential units in Shenzhen City is generally high, with the three indicators averaging 32.7%, 30.5% and 15.1%, respectively; (2) the types of residential greening differ per area; and (3) the level of residential greening is affected by development intensity, location, elevation and residential type. Such findings can serve as a reference for improving the greening level of residential units. This study proposes an accessible public green land index as a measure for the spatial relationship between residential units and surrounding green spaces. It suggest that future green space planning should pay more attention to the spatial distribution of public green land, and introduce quantitative indicators to ensure sufficient public green lands around the walking range of residential areas.

people's behaviour, such as encouraging children to walk to school and people to walk 50 and cycle [8]. Research shows that about 90% of the information received by people in 51 the environment comes from vision. The most commonly seen green landscape 52 amongst residents is 3-D greening, as represented by street greening and 53 neighbourhood trees [10]. The visual evaluation of 3-D greening enables an intuitive 54 understanding of the greening level in residential areas [11]. 55 Extant green space evaluation and research mainly focus on 2-D urban greening 56 [12] but rarely on street greening and neighbourhood trees in residential areas [8]. For 57 many years, China's planning management system evaluated urban greening on the 58 basis of green rate, green coverage index and park land per capita [13]. These 59 indicators can evaluate intuitively and quantitatively the total amount of urban green 60 space [14]. However, they cannot evaluate, guide and control the distribution of 61 greening in residential areas, and the relationship between green space and residential  images and other data to build a green database integrated by multi-source geographic 111 data combined with existing urban greening indicators. It also explores methods for 112 measuring greening level, spatial distribution and regional differences amongst urban 113 residential units. greening of residential areas, as well as serves as a model role in the urbanisation of 129 developing countries and regions. As a young city, Shenzhen is also less constrained 130 by the lack of green space apparent amongst old cities. Therefore, it is selected to 131 explore the difference in green resource supply of urban residential space.

157
The study integrated existing methods to measure urban greening. It proposed 158 three greening indicators, namely, green coverage index (hereinafter referred to as 159 GCI), green view index (hereinafter referred to as GVI) and accessible public green  In terms of GLI, this study integrated accessibility and green rate to quantitatively 201 compare the service level of public green land around different residential units [25]. 202 We proposed an indicator of public green land index of a residential unit within 203 walking distance (GLI). The calculation of GLI was based on current land-use data.

204
The proportion of public green land within a certain distance around the residential   Commercial houses in Shenzhen are mostly located in new residential areas with a 250 high green land rate. The GCI is even higher than the green rate because the land 251 areas of tree projection, roof garden and grass-planting bricks are included in its 252 calculation.

253
The GLI is 15.1% in average, and the average public green area in the 500 m 254 buffer zone of residential units is 11.9 ha, which is higher than the 4.4 ha of a recent    Moreover, their building density is high, leaving few spaces for greening in residential 305 areas. Hence, their average GCI is relatively low, between 25% and 34%.

306
Generally, each district differs slightly in terms of GVI. The average value of 307 each district falls between 22% and 37%. Amongst the districts inside and outside the 308 original SEZ, Dapeng and Pingshan district have relatively high greening levels.

309
Dapeng district has the highest average GVI of 36.5%, followed by Nanshan District 310 and Futian District at 36%. Guangming, Baoan and Longhua districts outside the 311 original SEZ have low average GVI value.

312
The average GLI of residential units in the whole city is 15.1%. Considerable 313 differences were observed amongst districts whose average value falls between 9% 314 and 40%. Dapeng District still has the highest value, with a GLI of 39.9%, followed 315 by Pingshan, Guangming and Yantian with GLI of above 20%. These areas have 316 many forests and country parks, and many residential areas are built on mountainside.  This study shows that the three greening indicators have both commonalities and 404 differences. The average GCI is the largest, whereas the average GLI is the smallest. In addition, this study finds that the difference in spatial distribution of GLI is the 414 largest amongst the several types of greening of residential units, whereas the spatial 415 difference of GVI is the smallest. Accessible public green land (GLI) refers to urban 416 parks, community parks, country parks, mountain parks and other urban public green 417 lands. Relevant planning standards mainly control the total amount of urban public 418 green land but fail to effectively control its spatial distribution, especially its 419 relationship with residential units [26]. Street greening is generally invested, 420 constructed and maintained by government departments at all levels, and thus the 421 spatial difference is small. In terms of green coverage inside the residential unit, due 422 to the stipulation of residential greening in the 'Code of Urban Residential Areas

423
Planning & Design' that 'green rate should not be lower than 30% in new urban areas 424 and should not be lower than 25% in rebuilt old areas', the overall spatial difference is 425 not large.