Study On The Dynamic Change Of Land Use/Cover In The Area Of Tumen River Cross Border Region Based On Multi-Source Data

In this paper, the Tumen River Area as a large-scale area was studied. Using 1976,1991,2000 and 2010 MSS, TM, ETM and other remote sensing images, and in the support of geographic information systems technology, this paper analyzed land use/cover change process and driving dynamics of the Tumen River Cross Border region (Include border area of North Korea and Russia) in the last 34 years.The first part of the main contents is about the remote sensing image classification based on the decision tree classification method——classification and regression trees (CART) which is compatible with the classification of texture images, spectral characteristics, and geographic ancillary data and so on. First select a small range of the study area as a test, two classification methods were used and the classification results showed that the classification accuracy of decision tree classification had greatly improved than the result of the maximum likelihood supervised classification. So the decision tree classification method was applied to the classification of all remote sensing images in the whole study area. Verification accuracy of the entire study area showed that better classification results can be analized in the further study.The second part is mainly about regional land use/cover dynamics analysis using the classification results above based on ARCGIS, ERDAS remote sensing software. The land use/cover change of the study area was analized from the aspects of area change, structural change, changes in terms of speed and transition matrix analysis, and we can see that Tumen River Area Land Use/Cover have a great change in all land use types from 1976 to 2010. Dry land has the largest increase in area, and construction land's area has a big increase too. While the area of coniferous forests have reduced 406062hm2, and mixed forest, grassland and broadleaf forest have much reduce in area. In addition, water and unused land has been reduced. Dynamic degree show that coniferous forests declined and construction land grew constantly. The land use/cover of the study area changed drastically. Transfer matrix showed that the land use types of major changes are dry land, coniferous forest, broadleaf forest, mixed forest and grassland. Overall, we can see the human factor in land use/cover change plays a very important role, highlighting a significant reduction in forest land and substantial increase in dry land and construction land. It has great difference in land use/cover changes in the three countries in the study area, and we can see the North Korea side area has greater changes in land use/cover than the sides of China and Russia, the Russia land use had the least change.The third part of the main content is the driving forces analysis of the land use change. Natural driving factors include topography, natural disasters and so on; Human driving factors include population, economic, technology, institutional and other aspects of institutional policy.Finally, the paper summarized the research work and presented in depth in future research and efforts that should continue to improve.