A Study On Spatiotemporal Variation Of Surface Urban Heat Island In Xi’an Based On Local Climate Zone

Understanding the seasonal variation of surface urban heat islands(SUHI)in different local climate zones(LCZ)is crucial for mitigating the impact of urban warming.However,this understanding is limited by the lack of high spatial and temporal resolution of land surface temperature(LST),and there is limited seasonal research on semi-arid cities such as Xi’an.In order to address these gaps,this study created a time series LST by fusing Landsat 5/8 satellite data and a gap filled MODIS product,and used LCZ maps and enhanced LST data to study the seasonal and spatiotemporal evolution patterns of SUHI in Xi’an city.The main study findings and conclusions are as follows:(1)This study uses multi-source data and random forest classifier to make a long time series LCZ classification map,which is used to study SUHI of Xi’an,a mega city in semi-arid climate region.The study found that the overall accuracy of LCZ classification is good(above88.3%),demonstrating the feasibility of using Landsat remote sensing images and geospatial data to create high-precision LCZ maps.This method is open-source and can be used for SUHI research in other cities.This study has found that rapid urbanization exacerbates SUHI,and the intensity of SUHI generally increases from suburbs to urban centers.In winter,the opposite is true,and the cold island effect appears in urban centers.However,this phenomenon no longer exists in the winter of 2020,indicating that the increase in urban coverage and population density has a positive effect on SUHI.(2)After using machine learning algorithms to fill the gaps in remote sensing images,this study used remote sensing data spatiotemporal fusion algorithms to produce high spatiotemporal resolution remote sensing images.The inverted LST images can provide a data foundation for in-depth research on SUHI and better capture the seasonal changes of SUHI.This study has shown that the seasonality of SUHI is stronger in warm seasons than in cold seasons.In spring,the SUHI intensity of building LCZ1-10 is higher than that of vegetation LCZA-D,while in winter,it is the opposite.The distribution range of SUHI in Xi’an continues to expand from the city center to the periphery as the urban built-up area expands.In the spring,summer,and autumn of 2005-2020,the SUHI intensity in the central urban area of Xi’an continued to strengthen,while in winter,there was no longer a cold island phenomenon in the urban area.(3)This study used long time series LST data and LCZ framework to investigate the spatiotemporal evolution pattern of SUHI in Xi’an,a semi-arid climate region.This study has found that the ultra-high SUHI intensity is most correlated with compact low-rise(LCZ3)and heavy industry(LCZ10),while sparsely built(LCZ9),water(LCZG),shrubs(LCZC),and dense trees(LCZA)all exist in very low SUHI intensity throughout the four seasons,which has a significant cooling effect on cities.LCZA has the lowest LST,indicating that it has a greater cooling effect compared to other vegetation types.To alleviate SUHI in Xi’an,this study suggests that the floor height of LCZ3 should be appropriately increased.Building more compact high-rise(LCZ1)buildings in low-density vegetation areas;In high-density vegetation areas,the number of open low-rise buildings(LCZ6)or LCZ9 should be appropriately increased.More trees should also be planted to increase the proportion of LCZ areas that have cooling effects on the city,especially LCZA and LCZG.