Abstract: Based on ten-day average temperature, precipitation and ten-day NDVI of SPOT data from 1999 to 2010 in north Shaanxi, annual responses of vegetation cover to temperature and precipitation was analyzed by using of the lag cross-correlation method. The results showed that there was strong correlation between NDVI and temperature and precipitation, with the correlation coefficient more than 0.9. The response was promptly over south of Shanbei, with the lag time only ten to twenty days. The correlation coefficient was between 0.75-0.85 across the Maowusu desert sand grass land area, with the lag time thirty to fifty days. NDVI showed a greater correlation and faster response to areas with much water and high temperature. From other perspective, the comprehensively responses to water and thermal factor on basin, loess tableland beam and gully region was better than the gullies rolling loess sub area and typical grassland ecological subregions. The effect degree of temperature on NDVI decreased gradually from south to north, the max  correlation coefficient was 0.947 between bush and temperature, and minimum correlation coefficient was 0.902 in desert. The effect degree of precipitation on NDVI was maximum in farmland, followed by plain grassland and low mountain grassland, and minimum in desert. The correlation coefficient was 0.926 between farmland and  precipitation, and 0.853 in desert. A certain type of vegetation required a higher temperature, also had faster response than others. On the contrary, vegetation depended on low temperature also showed slow response. The response of vegetation cover to annual precipitation was similar to that of temperature. The results could provide a theoretical basis for optimizing vegetation planting types in north Shaanxi.

Key words: Lag cross-correlation method, NDVI, Water and heat conditions, Correlation, North Shaanxi