| The rapid development of urbanization has led to a large increase in urban population and construction area.The natural surface has gradually been replaced by a large number of artificial impervious forms such as asphalt pavement,concrete and so on.Due to the change of the type of urban surface cover,the exchange of water and water in the surface and air has changed,and the local microclimatic effect of urban temperature above the suburban temperature is formed.In recent years,the impact of urban heat island on the impact of human living environment is becoming increasingly evident.Remote sensing data can obtain a large area of urban ground temperature,is a fast and effective technical means,in this study to remote sensing data to study karst city Guilin thermal environment changes and its impact factors,the purpose is to improve the Guilin living environment,Scientific approach to environmental management to provide theoretical and technical support.The images of the Landsat 5 TM satellite in the main urban area of Guilin were selected from 2006,2009 and 2010,the surface temperature was reflected and the parameters of the remote sensing of the impervious surface,water body and vegetation were described.The surface temperature of Guilin was normalized,and the change of surface heat condition was analyzed.The vegetation indices of GVI,NDVI,PV,RVI,MS AVI,SAVI,and S AVI were analyzed,and the vegetation coverage was compared with the mean and standard deviation.The vegetation parameters were used as the vegetation parameters to analyze the surface temperature.Regional differences are not sensitive to other vegetation parameters.Quantitative analysis of the relationship between vegetation coverage and surface temperature,respectively,the statistical coverage of different levels of vegetation coverage area of the average temperature,found that vegetation coverage is relatively high,the average temperature is relatively low.Temporal and spatial analysis of vegetation changes and regression analysis with the surface temperature,found that vegetation and surface temperature was negatively correlated.Quantitative analysis of impervious surface,found that the impervious surface and the surface temperature was significantly positive correlation,the use of NDBBI model to extract the construction land,regression analysis found NDBBI and vegetation water was negatively correlated.Quantitative analysis of the relationship between water body and surface temperature in Guilin,and found a negative correlation between water body and surface temperature.The results show that the green weight component is closely related to the surface temperature,and the effect of the surface water temperature on the surface temperature is higher than that of the vegetation and the water temperature.The results show that the temperature of the city is close to the surface temperature.Heat island effect will be more obvious.In view of Landsat 5 satellite thermal infrared band ground resolution of 120m ×120m,inversion can only get the resolution of the surface temperature.In order to obtain the surface temperature of 30m x 30m ground resolution,the neural network model of 120m x 120m surface temperature and related remote sensing parameters is constructed,and the model obtained by learning training is applied to input remote sensing parameters of 30m × 30m.According to the correlation coefficient between the surface remote sensing parameters and the surface temperature and the coefficient of regression fitting with the surface temperature,the green vegetation index,the normalized vegetation index,the modified vegetation index,the ratio vegetation index,the vegetation coverage,Modified normalized difference water index,normalized difference bare land and building land index,impermeable surface rate as a genetic neural network model for training and testing input.In this paper,the method of selecting input data is validated,and it is proved that the correlation coefficient and regression analysis coefficient are feasible. |
PDF Full Text Request