Characteristics of spatial distribution of daily maximum temperature on hot days and its improving measurements in Shanghai
Based on data from Landsat OLI (Operational Land Imager)and meteorological stations on July 12, August 1 3 and 29 ,20 1 3 ,the spatial distributions of temperature on hot days in Shanghai were evaluated using a spatial regression method.The distribution characteristics of thermal environment for low,medium and high tem-perature zones were analyzed from temperature division,landscape structure,and spatial autocorrelation.On the a-bove basis,the composite characteristics of distribution and land use type of the high-temperature zone in which the daily maximum temperature is larger than 38 ℃were studied combining the land use data in Shanghai.The results show that the prediction model established by a least square regression method using surface temperature,normal-ized water vapor index and solar radiation plays very well in predicting the temperature on hot days with RMSE (Root Mean Square Error)of 1.75 ℃and R2 of 0.92.Using this model we can get a whole and continuous spatial distribution of urban thermal environment.It demonstrates non-equilibrium,variety and crushability characteristics in the urban thermal environment.Meanwhile,there is a spatial aggregation feature in the similar temperature zones.The low-temperature zones,in which the daily maximum temperature is less than 35 ℃,are mainly located in the water area and the southern part,and the high-temperature zones are at the urban or suburb residential areas in Shanghai.It indicates that the urban land use increases the urban thermal environment,while water and vegeta-tion are beneficial to reduce it.Therefore,it suggests that the adverse effect resulted by the extremely high-tempera-ture weather could be mitigated through a series of measurements,such as planning the urban reasonably,control-ling the development level and increasing the greening areas in the future.
Hot dayDaily maximum temperatureSpatial heterogeneityImproving measurement
国家科技期刊平台
NSTL
万方数据
维普
