Prediction of vegetations dynamic changes in central Nujiang watershed based on Markov process model

By using GIS,RS and GPS techniques and the TM images of 1994 and 2004,the changes of vegetation landscape patterns in central Nujiang watershed in 1994-2004 were studied,and the prediction of the changes in next ten years was made by using Markov process model.The results showed that the matrix of the landscape in study area was composed of shrub and grass,with other landscape patches embedded.In 1994-2004,the most transfers among landscape types were between shrub-grass,semi-humid evergreen broadleaved forest,and coniferous-broadleaved mixed forest.The greatest change occurred in shrub-grass and semi-humid evergreen broadleaved forest at low elevation,while no obvious change was observed in coniferous-broadleaved mixed forest and cold temperate coniferous forest at high elevation.It was predicted that in the next ten years,the influence of human interference on the vegetation succession in the whole watershed would be further increased,the areas of shrub-grass and cold temperate coniferous forest would increase while those of semi-humid evergreen broadleaved forest and humid evergreen broadleaved forest would decrease,the landscape as a whole would be fragmented,and the regional eco-environment would be seriously affected.