Abstract: By using the 1982-2003 NASA data of Global Inventory Modeling and Mapping Studies (GIMMS) Normalized Difference Vegetation Indices (NDVI) with a spatial resolution of 8 km as well as the historical climate data, and based on the extraction and analysis of pixel-to-pixel discrete information with the help of regression and correlation analysis, this paper studied the inter-annual variation of vegetation cover in Guizhou of Southwest China in 1982-2003 and the relationships between this variation and main climate factors, aimed to reveal the basic characteristics of Guizhou vegetation cover change and its responses to climate change. In the recent 22 years, the NDVI, air temperature, and precipitation in Guizhou exhibited an increasing trend, with a linear tendency being 0.001, 0.302 ℃, and 12.776 mm per decade, respectively. The mean monthly NDVI increased  with air temperature, and had a noticeable parabola relationship with mean monthly precipitation. There existed a threshold value in the effects of precipitation on NDVI. Air temperature and NDVI showed a synchronized inter-annual variation trend, while yearly precipitation displayed a certain lagging with the inter-annual variation of NDVI. Different types of land cover had different responses to climate change, and air temperature change had greater effects on the NDVI, compared to precipitation change.

Key words: Citrus unshiu, Light intensity, Chlorophyll fluorescence, State transition, PQ pool