| In the past 20 years, with the China’s reform and opening up go further and total amount of economy continues to improve, the urbanisation process continues to change the condition of the land surface. As the center of national politics and culture, Beijing, its change of the underlying surface is very significant, and the change has continuity on time and integrity on the development stage. it is very representative for present study. This study put the nine main counties and districts in Beijing as the study area, including the two core area of the capital functions area: Dongcheng and Xicheng(Chongwen and Xuanwu were incorporated into Dongcheng and Xicheng in 2012).Four expansion area of city functions: Haidian,Fengtai, Chaoyang and Shijingshan,and three development area of the city:Tongzhou, Shunyi and Changping. By means of remote sensing and GIS, the analysis to the study area was carried on the characteristics of impervious surface spatial expansion and its effect on the thermal environment, combined with the Logistic regression model,discuss on the internal driving force of expansion was making.the following conclusion was got:1.It is concluded from the section analyse that in the whole time,Impervious Surface’s expansion is mainly concentrated in the north and north-east direction,the north-west of it is quick in the early and faster than the average level generally,but it become slow in the last and stand in the middle level.The east and south-east of it only have a great development in the middle time and decrease in the last.The west only in the early time has a development.The south-west and south is always in a low level in the expansion of Impervious Surface.2.The result of land-use transfer matrix suggests that after the nearly 20 years of development, The classification transformed into the impervious surface has regional difference and dynamic characteristics because of various historical reasons and position.(1)in terms of area,the newly increased impervious surface mainly concentrated in the developing new counties, including Changping,Tongzhou and Shunyi, Their growth rate and the sum of the impervious surface was far beyond the expanding area like Haidian,Chaoyang, Fengtai and Shijingshan. Early in each district, the increasing area and the speed of impervious surface was relatively quite small, all of them became greatly accelerating In the middle and finally slowed down in the late.the impervious surface percentage in the classification showed that the proportion in the expanding area was roughly the same with the cultivated land and green space at the beginning(1992) and became dominant(2007).Then it became overwhelming and exceed 50% in the whole area(2011).this situation indicated that the land-use structure had changed in quality and the function of agriculture has shrunk a lot. Although the proportion of impervious surface in the new district were on the rise in each period, the farmland and green space were the majority in the land-use types.As to the source of the impervious surface, the new district depended on the cultivated land and the expanding area derived from the green space and the farmland from 1992 to 1999.in the stage from 1999 to 2007,the cultivated land conversion percentage has all risen in each county. When it comes to the last period, only the new district depended on the cultivated land as the source of the impervious surface. The trend that farmland and green area were transformed into the impervious surface has been controlled and the policy of ecological environment protection in city initially appeared.3. The factors of influencing the impervious surface expansion of Beijing has a lot, in which the main issues are economic and demographic factors and space factors. Combined with principal component analysis and Logistic regression model, the result shows that: the economic and demographic factors are the most important in factors affecting the expansion of impervious surface, regression coefficients of the first principal component which represents them in the model are up to 1.904,which is the highest of all variables. And the increasing proportion of impervious surface and the increase of social investment in fixed assets as well as the income of the government reached significant or extremely significant correlation. the correlation coefficients are 0.79 and 0.93. it suggests that the expansion of impervious surface is closely related to the economic development. The spatial distance factors are also the key factors which influence the impervious surface expansion. the regression coefficient is-1.003.it concludes that its effects on the expansion of the Impervious Surface is negative and when away from the main traffic line and original impervious surface every 1 units, the occurrence probability of impervious surface expansion will reduce 37%.4. The strong heat island center of Beijing are mainly distributed in the core area and extension area. Nearer to the center of the island, stronger the heat island is. The Thermal Field Variation located in the Dongcheng and Xicheng is ten times than it in Tongzhou. The heat island effect in southern part of the study area was stronger than that in the north. Average surface temperature in different heat island center of regions and has great differences from the average temperature in surrounding area. Overall, the surface temperature of every direction is decreased with increasing distance away from the center, surface temperature along the south profile drop fastest.In the 1/3 range away from the center, the LST value changes slightly, then turns volatility with the increase of the distance. while closer to the center, the high value of LST appeared more frequently. Land Surface Temperature(LST) and the Impervious Surface Percentage(ISP), Normalized Difference Vegetation Index(NDVI) were significantly correlated, the correlation coefficient all reached 0.86. In the regression model fitting LST and ISP, the optimal polynomial model is the best, the coefficient of determination R2= 0.7481. When introducing the NDVI, the accuracy of multi variable regression model is higher than that of single ISP variable model. According to the estimated surface temperature of the multi variable regression model decreased after reducing the rate of impervious surface and raising the same proportion of vegetation cover compared with adjusting before especially in the areas of high ISP. The decreasing value is 1.15 centigrade. |
PDF Full Text Request