| Urbanization lead to replacement of natural vegetation by impervious surface in urban area, causing precipitation reduced, drought enhanced, temperature increases etc negative environmental problems. Impervious surface is not only the important index of urban environment such as water quality, rainfall runoff, urban heat island effect etc, but also urbanization index to study city’s expand changes, and has significance for studying urbanization and urban environmental problems and their interaction.This paper uses Landsat images, contrasts traditional pixel-based classification with object-oriented classification to extract impervious surface information and other land cover components information, and the relationship between land cover and urban heat effect, as well as spatial-temporal change are discussed. The main contents and conclusions are as follows:1.The NDVI, MNDWI and DEM are introduced as new bands in multispectral data to distinguish water from vegetation, mountain area from plain. The results show a better classification precision and the difficulty of Landsat image information extraction is reduced.2. Based on V-I-S model, using Landsat data, impervious surface information and other land cover components information are extracted and classified respectively using pixel-based classification and object-oriented classification. Meanwhile, the classification results is evaluated by field data. It shows that the pixel-based classification is superior to the object-oriented classification. And, pixel-based classification is better and is commended in our study.3. The land cover information is extracted by using pixel-based classification. The results show that, from 2002 to 2014, vegetation area in the study area has reduced 313.5km2. However, the area of impervious surface, soil and water are increased, where impervious area is increased by 225%; soil and water area are increased 56.7km2 and 9.5km2. The impervious patches spread out over the whole plain field, and the main growth area is located in the periphery of the center of the city. And, the proportion of impervious in the Daxin Town, Liangtian Town and Xigang Town, has the largest increase, respectively increased form 26.6%,11%,4.7% in 2002 to 59.6%,33.5%,25.4% in 2014. Wangyuan Town and Yanghe Town take the second place.4. Urban heat island effect is the result of interaction of impervious, soil, vegetation, water etc. mono-window and split-window algorithm are applied to retrival LST, and the level of thermal environment is established by relative LST. The results show that, cool island and weak heat island patches account for large parts in study area, which shows ecology and environment in study are well. The high strength thermal environment patches are located in the Helan piedmont alluvial plain, center city and hill lands on the east of Yellow river, and thermal environment in 2014 is obvious higher. Meanwhile, the reason of strong thermal environment are low vegetation coverage and the desert soil.5. Average LST of soil is highest, and average LST of impervious take the second place, average LST of vegetation land is lower. Because the thermodynamic activity of impervious surface is decreased by green space nearby, impervious land contribute much to the formation of low- and medium-grade thermal patches. Comparatively, soil contribute much to the formation of high-grade and higher-grade thermal patches. And, the LST increased as the vegetation coverage decrease and vegetation land has an absolute advantage in study area which lead to higher contribute to the formation of all-grade thermal patches. |
PDF Full Text Request