Sky View Factor Calculation Based On Baidu Street View Images And Its Application In Urban Heat Island Study ——A Case Of Shanghai

As the constant improvement of urbanization in China,the problem of urban heat island is becoming more serious.How to reduce urban heat island(UHI)effect effectively has received much attention.The sky view factor(SVF)is one of the most important indicators to characterize the urban radiation fluxes and urban thermal environment.Therefore it is a key morphological parameter to study UHI effect.How to calculate large-scale SVF quickly and accurately is important to urban morphology and urban climate research.Studies have shown that SVF has a strong relationship with UHI intensity,nevertheless the relationships found can be contradictory.In this paper,Shanghai,which has a high level of urbanization and prominent UHI problem,is selected as the research area.Firstly,we proposed a fast,accurate and automatic SVF calculation method using street view images and deep learning algorithms,and applied the method to the UHI study in the city center of Shanghai.Secondly,based on Landsat-8 OLI/TIRS images and radiative transfer equation,we retrieved land surface temperature of Shanghai to examine the spatiotemporal variation of the surface urban heat island(SUHI)with urban development from 1989 to 2020.Also,we analyzed the influencing factors of urban heat island at the macroscale.Finally,we focused on the impact of urban morphology on SUHI,and took SVF as an effective indicator of urban morphology.Based on the local climate zones(LCZ)scheme,we combined large-scale SVF value with the land use and building morphology to examine the relationship between SVF and SUHI intensity.The specific results are as follows:(1)Deeplabv3+can detect the sky and non-sky range effectively in different scenarios(MIOU=91.64%).The SVF calculated using the proposed method is in good agreement with that calculated using fish-eye photos(R~2=0.8869).(2)Shanghai has different spatial distribution characteristics and change trends of SUHI in different periods.In a word,between 1989 and 2020,the temperature in this city is rising and the area of SUHI is expanding significantly.There are more hot spots in the suburbs.This means that SUHI has a multi-center distribution.However,the downward trend of the SUHI intensity in the central urban area is obvious.With more low temperature centers appearing,the intensity of SUHI in the central urban area has declined.(3)Urban construction land has a strong positive linear correlation with surface temperature.Vegetation coverage has a strong negative linear correlation with surface temperature.Population density has a significant positive exponential correlation with surface temperature.(4)The nocturnal SUHI intensity to each LCZ type in Shanghai is as follows:LCZ 8(large low-rise)>LCZ 7(lightweight low-rise)>LCZ 10(heavy industry)>LCZ 1(compact high-rise)>LCZ 3(compact low-rise)>LCZ 5(open mid-rise)>LCZ E(bare rock or paved)>LCZ 4(open high-rise)>LCZ 6(open low-rise)>LCZ F(bare soil or sand)>LCZ D(low plants)>LCZ B(scattered trees)>LCZ G(water).(5)For different LCZ types,the relationship between SVF and nocturnal SUHI is significantly different.For LCZ 1 and LCZ 3,the correlation coefficient is-0.75 and-0.54 respectively.For LCZ4 and LCZ 5,the correlation coefficient is 0.38 and 0.62 respectively.For LCZ 7?8?10,the correlation coefficient is 0.08.The proposed SVF calculation method is shown to be applicable in high-density and complex urban environment.In addition,the calculation of large scale continuous SVF provides the possibility for zonal understandings of the SUHI effect based on the LCZ scheme.