| Urban green space is an important public resource that provides multiple ecosystem services to urban ecosystems.Previous researchers have commonly used green coverage rate,the ratio of green space,normalized vegetation index(NDVI),and per capita green area to measure urban greenery.These indicators mainly reflect urban greening on a two-dimensional plane,ignoring the urban greening resources and quality on the three-dimensional space.The green view index(GVI),namely the percentage of green in human vision,can better reflect the three-dimensional greenery of the city than the traditional two-dimensional indicators.In recent years,the development of image processing technology and online street view service has promoted the wider application of GVI.Although GVI has been widely used in the urban green evaluation,there is still a lack of knowledge.First,the relationship between GVI as a new quantitative indicator of urban green space and the commonly used quantitative parameters of urban green space in the past is unclear.Second,the main influencing factors of the GVI are unknown,especially the main influencing factors of the GVI in the intercity,intracity,and the country.Third,regarding the application of GVI,whether GVI can be used to characterize urban biodiversity,there is still a knowledge gap.Based on this,in this research,we used the Baidu Street View big data was to measure GVI and combined remote sensing data,field survey data,and literature data to conduct a systematic study on the above three issues.The specific research content is as follows:Firstly,the relationship between GVI and NDVI and green coverage rate was analyzed.On the road scale,according to the vertical perspective,we divided the GVI into three layers: upper GVI,middle GVI,and lower GVI,which correspond to the tall tree greenery,shrub greenery,and ground cover greenery,respectively,and analyzed their relationship with the tree size,and the woody plant species composition,and physical and chemical properties of the soil.On the subdistrict scale,the Gini index was used to quantify the fairness of GVI spatial distribution,and the relationship between GVI and Gini index and socioeconomic factors,biophysical factors,and landscape patterns were analyzed.At the scale of the two cities,we compared and analyzed the differences in the spatial distribution of GVI between and within the cities of Harbin and Changchun,as well as the relationship with plant composition,tree size,and landscape pattern.On a national scale,31 major cities in China were selected to analyze the relationship between GVI and climate factors and socioeconomic factors.Finally,the relationship between urban bird diversity and its composition and GVI was analyzed.The results of our study are as follows:The relationship between GVI and NDVI and green coverage rate were weaker(r<0.63).GVI,NDVI,and green coverage rates reflect different perspectives of urban greenery.The GVI was mainly driven by socioeconomic factors compared to NDVI and green coverage rate which were mainly driven by biophysical factors.In the future urban greenery evaluation,a comprehensive evaluation index including GVI should be constructed to evaluate the quality of urban greening.On the road scale,the GVI in the vertical direction was ranked as middle GVI>lower GVI>upper GVI.After analyzing the relationship between GVI and tree size,plant composition,and soil physical and chemical properties,we found that largesize trees and a high proportion of Salicaceae could create more average GVI and upper GVI.On the whole,the GVI was negatively correlated with soil p H and electrical conductivity.On the subdistrict scale,the mean GVI and Gini index in Changchun was 5.5 and0.53,respectively.The GVI and spatial distribution equity of GVI of the subdistricts located in the city center were lower than those near the city outskirts.The best BRT model explained 56.8% and 52.5% of the variation of mean GVI and Gini index,respectively.Landscape patterns of urban green spaces were the dominant driving factors for both mean GVI and spatial fairness of subdistricts street-view greenery.Mean GVI and Gini index both have the threshold response to the percentage of landscape(PLAND)and edge density(ED).Green spaces with larger and complex shapes were conducive to improving the GVI and spatial fairness of street view greenery.Subdistricts with high socioeconomic status have more availability and spatial fairness of neighborhood street-view greenery.On the two-urban scale,the mean GVI,middle GVI,and lower GVI in Harbin were significantly higher than those in Changchun(p<0.05).The lower GVI of the two cities showed an upward trend from the city center to the suburbs,while the upper GVI showed a radial downward trend.The GVI was mainly driven by PD,tree height,total green area,and ED in Harbin,while the GVI was mainly by tree height,ED,and aggregation index(AI)in Changchun(p<0.05).In future urban planning and management,the urban GVI can be improved through landscape patterns regulating and protecting urban large-size trees.On the national scale,the mean GVI was mainly influenced by climatic factors rather than socioeconomic factors.However,the maximum value of GVI in the city was influenced by both climatic and socioeconomic factors.The upper GVI was directly affected by sunshine hours,however,other climatic factors mainly indirectly affected upper GVI through sunshine hours.Middle GVI was significantly affected by MAP,followed by sunshine hours,relative humidity,and GDP.The lower GVI was directly influenced by relative humidity significantly(p<0.05),followed by sunshine hours and per capita income but not to a significant level(p>0.05).As an emerging parameter for urban greening measurement,the GVI can be used to indicated urban bird diversity.The number of families,genera,and species of urban bird increased with increasing urban MAT,GVI,GDP,and plant richness;Urban bird composition was mainly affected by spatial distance and temperature difference.The composition of urban birds was mainly affected by the difference in urban space distance and temperature,and the difference in composition increases with the increase in the distance and temperature difference between cities.In summary,street view big data can be used to evaluate urban green space through the calculation of GVI.It is a supplement to traditional field measurement and remote sensing methods.It is possible to explore the differences in the greenery of road,subdistrict,city,and national level and their possible driving factors for geographical climate and social development from a new perspective,and have the potential to be used to evaluate other differences in biodiversity.It is expected that its simplicity,quickness,and big data characteristics can support the evaluation and construction of forest cities and forest city clusters in China. |
PDF Full Text Request