Abstract:
Spatial and temporal variations in alpine vegetation cover have been analyzed between 1982 and 2001 in the source regions of the Yangtze and Yellow Rivers on the Tibetan Plateau. The analysis was done using a calibrated-NDVI (Normalized Difference Vegetative Index) temporal series from NOAA-AVHRR images. The spatial and temporal resolutions of images are 8 km and 10 days, respectively. In general, there was no significant trend in alpine vegetation over this time period, although it continued to degrade severely in certain local areas around Zhaling and Eling Lakes, in areas north of these lakes, along the northern foot of Bayankala Mountain in the headwaters of the Yellow River, in small areas in the Geladandong region, in a few places between TuoTuohe and WuDaoliang, and in the QuMalai and Zhiduo belts in the headwaters of the Yangtze River. Degradation behaves as vegetation coverage reduced, soil was uncovered in local areas, and over-ground biomass decreased in grassland. The extent of degradation ranges from 0 to 20%. Areas of 3×3 pixels centered on Wudaoliang, TuoTuohe, QuMalai, MaDuo, and DaRi meteorological stations were selected for statistical analysis. The authors obtained simple correlations between air temperature, precipitation, ground temperature and NDVI in these areas and constructed multivariate statistical models, including and excluding the effect of ground temperature. The results show that vegetation cover is sensitive to variations in temperature, and especially in the ground temperature at depths of ~40 cm. Permafrost is distributed widely in the study area. The resulting freezing and thawing are related to ground temperature change, and also affect the soil moisture content. Thus, degradation of permafrost directly influences alpine vegetation growth in the study area.
