Assessment of Land Degradation and Improvement in China Accounting for Soils,Terrain and Land Use Change

摘要

Caveats about use of NDVI time series data as an indicator of land degradation and improvement include inherent issues of cloudiness and auto-correlation, and the lack of compatible data for soils and land use. For China, removal of residual cloud effects from the 1981-2006 GIMMS dataset by harmonic analysis made almost no difference to trends or areas affected. Non-parametric analysis that is unaffected by auto-correlation shows much the same patterns of greening and browning as the linear trends, which suggests that the linear trends are robust. 34% of the country shows climateadjusted trends of vegetation productivity that may be attributed to human activities: 3/4 negative, mostly in the humid south east; 1/4 positive, mostly in dry lands. Combining NDVI time series with soil and terrain indicates that these two factors influence 25% of the productivity trends. Detailed study in Zhejiang Province, using Landsat imagery, reveals a direct influence of land use change on land degradation but, also, some contradictory trends that may be attributed to changes in management without change in the kind of land use and disturbances like as forest fires and outbreaks of pests and disease that cannot be identified from our data.