Using satellite-observed Normalized Difference Vegetation Index (NDVI) dada and station-observed surface air temperature anomalies for the Northern Hemisphere (NH), we analyze the spatio-temporal characteristics of vege- tation variations in the Qinghai-Tibet Plateau and their correlations with global warming from 1982 to 2002. It is found that the late spring and early summer (May-June) are the months with the strongest responses of vegetation to global warming. Based on the Rotated Empirical Orthogonal Function (REOF) method, the study shows that the first REOF spatial pattern of average NDVI for May-June reveals the northern and southern zones with great inter-annual variations of vegetation, the northern zone from the eastern Kunlun Mountains to the southwestern Qilian Mountain and southern zone from the northern edge of the Himalayas eastward to the Hengduan Mountains. The vegetation, especially grassland, in the two zones increases significantly with global warming, with a correlation coefficient of 0.71 between the first REOF of May-June vegetation and the April-May surface air temperature anomaly in the NH during 1982-2002. A long-term increasing trend in May-June vegetation for the plateau region as a whole is also attributed mainly to global warming although there are considerable regional differences. The areas with low NDVI (grassland and shrubland) usually respond more evidently to global warming, especially since the 1990s, than those with moderate or high NDVI values.
