Study On Local-scale Urban Climate In The Hot-humid Region Based On Remote Sensing And Surface Energy Balance

With the global climate change and the process of urbanization,the problems of urban climate are increasingly exacerbating,especially in the hot-humid region due to its climatic characteristics.Local-scale is the most important scale for studying urban climate since it is closely related to human life and building energy consumption.Therefore,studying the local climate in the hot-humid region will provide the key base to improve the living environment quality of residents,reduce the urban heat island intensity and building energy consumption,and alleviate the issues of urban climate.Surface energy balance?SEB?is the key method to study the local climate,and there is a dearth of the SEB observations and modelings from the humid subtropical region.Therefore,taking Guangzhou located in the hot-humid region of China as the study object,the present study conducted local-scale SEB observations in the compact high-rise zone which is mostly distributed in Guangzhou by using the eddy covariance technique from September to November2016.The results concluded the basic chracteristics of surface energy exchanges in the hot-humid region,demonstrated that the sky condition played an essential role in urban SEB,and validated the adaptability of LUMPS in the hot-humid region(the predicted Q^* and ?Q_s have highly positive correlations with the measured values,showing an index of agreement higher than 0.95,and the RMSE values of Q_H and Q_E are 36.5 and 42.8 W m^-2 within an acceptable range).Remote sensing technique is the key method to quickly retrieve surface parameters in a large area.Up to now,there is a dearth of a complete methodology of remote sensing which can be applied to the local climate.Therefore,the present study proposed a new SEB-based method for retrieving near-surface air temperature and humidity by using the high-resolution remote sensing data of Landsat and Geo Eye-1.In addition,the present study proposed a complete remote sensing methodology of local climate by combining another meteorological parameter of LST?Land surface temperature?and relevant urban design factors,and validated the precision of the key parameters of land surface temperature,near-surface air temperature,and water vapor pressure that were 1.0??1.0? and 2.06 h Pa,respectively.It provides a solid methodology for large-area observation and quantification of the local climate.The existing local climate prediction models were parameterized from a few cases,which might limit their applications and could not widely be applied to the pratical urban design and planning.Therefore,combining Local Climate Zone?LCZ?system,the present study retrieved the numerous remote sensing data?including 625 LCZs?of the central Guangzhou using the proposed remote sensing methodology mentioned above,confirmed the key urban design factors which affect the major LCZ types,and proposed the urban heat island?UHI?and surface UHI?SUHI?prediction model with the precisions of SUHI and UHI prediction models less than0.95? and 0.04?,respectively.The present study gives suggestions for the climate-sensitive urban designs in the hot-humid region and provides an important reference for other regions of China and global humid subtropical regions.