To link regional land use/cover changes with environmental effects, land cover changes are required to reflect vegetation successions, whereas the land cover classification systems commonly used nowadays cannot serve this purpose. In this paper, a new land cover classification system is established in which land covers are classified by the vegetation succes-sions, taking Zamtang County, Barkam County and Jinchuan County in the upper Dadu River watershed as a study area. Using multi-temporal remote sensing images, the land cover data of 1967, 1986 and 2000 are obtained by means of integration of unsupervised classification and visual interpretation methods. The database facilitates the study of land use/cover changes, en-vironmental effects and ecological construction. Land cover changes reflect the main ecological processes in the upper Dadu River watershed. The landscape composed mainly of grasslands, wildwoods and alpine scrubs in 1967 was changed to that of grasslands, secondary forests, al-pine scrubs, fragmentary wildwoods, artificial forests, secondary scrubs in 2000, meanwhile, the landscape got more fragmentized. The total area of the forests decreased by 9.43%. Study results have shown the process of restoration of logged areas in forest centers. From 1967 to 2000, only 6.86 percents of logged areas were converted to shrubs, meadows or crop-lands, and the rest were converted into artificial forests or secondary forests. So the ecological shelter functions will be restored, stage by stage. Firewood collection, charcoal production and overgrazing are the three major triggers for the extensive degradation of alpine oak forests, Sa-bina tibetica forests and meadows. The arid valley grasslands expanded too. The degradation of vegetation in the southern slopes impairs ecological shelter functions and affects livelihood of local residents, so it is essential to find effective measures for ecological restoration and recon-struction. Field investigations have found that the current measures have not concerned wi
YAN Jianzhong1,2,3, ZHANG Yili1, BAI Wanqi1, LIU Yanhua4, BAO Weikai5, LIU Linshan1, 2 & ZHENG Du1 1. Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing 100101, China
