Research on Variation Analysis and Prediction of Wetland Resource in Khanka Lake Nature Reserve

Two periods of TM remote sensing images in 1994 and in 2005 are used to analyze the dynamic variation of wetland's types at first, and then, using the cellular automata (CA) model and Markov model to predict the change tendency of wetland types in Khanka lake nature reserve. The result is as following: (1) from 1994 to 2005, the area of herbaceous swamp & swamp meadow and shrub swamp have a rapid decrease, and the area of paddy field and dry land expand in large scale. (2) CA models and Markov model show a similarity prediction results which presented that natural wetlands continue to decrease and constructed wetland continue to increase. (3) the most important reason of nature reserve wetland change is natural wetland reclamation driven by the agricultural economic development. Although the establishment of nature reserve played an important role to accelerate the recovery of natural wetlands (especially in core area), and relieve the decrease of nature wetlands to some extend, but the overall natural wetland resources still tend to degradation.