The Land Use And Land Cover Change Dynamic Monitoring And Driving Force Analysis Of JingJinTang Of China

The study on the landuse and landcover may date back to five decades ago in the field of Earth sciences, however, it is still challenging in timely identifying the dynamic changing information and improving the accuracy of the monitoring. The main reason is that there is no uniform definition of the land use and land cover adopted and there is still some shortfalls in the methodology of land cover information monitoring at large area and dynamic information identifying. It has been the major method of land cover and land use monitoring taking combined advantages of geophysical information system and remote sensing technology.This thesis presented the study on the change of land cover and land use in the Beijing-tianjin economic circle over the period of 20 years. The thesis adopted the Object-oriented classification that is based on multi-scale segmentation and comparative fuzzy mathematic classification to classify remote sensing images in 1990 and 2000 with the reference to the thematic data of the land use and land cover in 2010 developed by Chinese academy of sciences. This thesis further analyzed the driving forces based on the analysis of the economic development, demographic and political influences on the change of land cover and land use. The major conclusions are as follows:(1)The keys for the classes and the classification system were set in the study area with the reference to the thematic data of the land use and land cover in 2010 developed by Chinese academy of sciences. The fuzzy mathematical method was explored to conduct the multiple scale segmentations, feature selection and the object-oriented classification on the Landsat TM / ETM +imageries at30 meter resolution in 1990 and 2000 in the Tianjin-Tangshan Economic circle. The preliminary land cover and land use maps in 1990 and 2000 were obtained finally.(2) The above mentioned land cover and land use maps were further corrected with the reference of the ground truth data in 2010 based on the backtracking algorithm that was developed in this paper.(3)The validation showed that the overall classification accuracies of the object-oriented classification in 1990 and 2000 before and after the correction improved from1.7856 and 0.8124 to 0.8524 and 0.8739, respectively and the Kappa coefficientsincreasedfrom0.7253 and0.7596 to 0.8106 and 0.8381,respectively.(4)The change information over two decades was made available under the support of Arcgis software. Based on the change information, it found that the woodland, grassland, arable land and unused land decreased from 1990 to 2010 while the wetland and buildup area increased. In type conversion analysis, the wetland increased mainly from arable land and unused land, artificial surface transformed from wetland and farmland. The woodland decreased into farmland mainly, the grassland reduce into wetland, the farm land had a significant reduction into artificial surface. The unused land decreased mainly into farmland and wetland. In terms of the changing rate, the increasing speed of wetland was the highest and followed by the buildup area while the decreasing speed of arable land was the highest and followed by the unused land. The overall changes of land cover and land use were 0.29% for 1990 to 2000 and 1.02% for 2000-2010, respectively.(5)The thesis analyzed the driving forces of the change of the land use and land cover change based on the statistical data, land use and land cover dynamic change info. The relationship among land utilization, land cover change dynamic, economy, population and politics of the Beijing-tianjin economic circle was further analyzed.