Disentangling the effects of the surrounding environment on street-side greenery: Evidence from Hangzhou.

• People are more exposed to street greenery than to parks. • The Green View Index (GVI) is used to quantify the daily exposure to street-side greenery. • Development of a multidimensional indicator set to explore the association of the micro-scale surrounding environment characteristics and GVI. • The context surrounding environment exhibits a strong role on physical appearance of eye-level greenness of GVI. • Findings can shed lights on developing urban green design and planning policy. Human-scale greenery has been heavily featured in the development of planning-related theories and research. Daily exposure to street greenery is proportionately greater than exposure to parks. Several case studies have estimated the green view index (GVI), which quantifies daily exposure to street-side greenery, at the city-scale. However, the street-side greenery distribution is uneven. The relationship between the GVI and the surrounding environment has rarely been discussed. Here, we investigate the differentiation of street greenery from the physical features of a street and examine how the surrounding environment affects the physical appearance of eye-level greenery. In this work, we analyzed the GVI using the internet data crawling approach and obtained 12,232 panoramic street view images from Baidu Map for the urban zones of Hangzhou, China. A range of 17 surrounding environmental characteristics are integrated with multisource geographic data to analyze their relationship with the GVI at the microscale. Spatial econometric models are explored to provide insights into the underlying mechanisms associated with the GVI. The results showed that the surrounding environment exhibited a strong role in the physical appearance of eye-level greenness. There are two pathways through which surrounding environmental characteristics (the nature of land use and the enclosure of the street) affect the distribution of GVI. Four factors proved to be important: scenic protection policies and ecological renewal projects, vertical and horizontal green characteristics of different land uses, and facade designs dominated by enclosed exterior walls or railing walls. Interesting approaches are presented to rationally implement activities related to these factors. The urban green design process should be facilitated in terms of the trade-off between greenery and optimum land use. These findings can be useful in drafting appropriate policies and increasing eye-level greenery in cities. [ABSTRACT FROM AUTHOR]