Spatiotemporal patterns of street-level solar radiation estimated using Google Street View in a high-density urban environment

Abstract

This study presents a method for calculating solar irradiance of street canyons using Google Street View (GSV) images and investigates its spatiotemporal patterns in a high-density urban environment. In this method, GSV images provide a unique way to characterize the street morphology from which the diurnal solar path and solar radiation exposure can be estimated in a street canyon. Verifications of our developed method using free-horizon HKO observations and street-level field measurements show that both the calculated clear-sky and all-sky solar irradiance of street canyons well capture the diurnal and seasonal cycles. In the high-density urban areas of Hong Kong, we found that (1) the lowest monthly averaged solar irradiations in winter are 6.6 (December) and 4.6 (February) MJ/m2/day, and the highest values in summer are 17.3 (July) and 10.8 (June) MJ/m2/day for clear-sky and all-sky calculations, respectively; (2) The spatial variability of solar irradiation is closely related to sky view factor (SVF). In summer, the irradiation in a low-rise region (SVF≥0.7) on average is about three times that in a high-rise region (SVF≤0.3), and they differ by about five times in winter; (3) Street orientation has a significant impact on the solar radiation received in a high-density street canyon. In general, street canyons with West-East orientation receive higher solar irradiation during summer and lower during winter compared to those with South-North orientation. The generated maps of street-level solar irradiation may help researchers investigate the interactions between solar radiation, human health and urban thermal balance in high-density urban environments. Keywords: Solar radiation; Sky view factor; Street canyon; Google Street View; Deep learning; Hong Kong