Abstract: Accurately quantifying the influence of climate change on NDVI and determining key driving meteorological factors are of great significance for vegetation management in ecologically fragile areas. Based on the meteorological data and MODIS NDVI remote sensing image data of 23 meteorological stations in Northwest Liaoning during the growing season from 1998 to 2020, we analyzed the change trends of NDVI, air temperature, precipitation, sunshine time and evapotranspiration using Theil-Sen Median slope estimation and Mann-Kendall trend analysis. The correlation between NDVI and meteorological elements and the spatial-temporal variations of key climate drivers were examined by using spatial interpolation, correlation analysis, multiple regression, and bandmath. The results showed that NDVI significantly increased with a rate of 0.03 10a^-1 from 1998 to 2020 in Northwest Liaoning (P<0.01), but the change of climate was not significant. The trends of the changes in air temperature, evapotranspiration, sunshine time, and precipitation during 1998-2008 and 2009-2020 were opposite. In Northwest Liaoning, NDVI was negatively correlated with air temperature, evapotranspiration, and sunshine time, but positively correlated with precipitation. The correlations between NDVI and those four climatic factors during 2009-2020 were obviously weaker than that during 1998-2008. The proportion of relative importance value for sunshine time and evapotranspiration was 54.52% and 22.93% during 1998-2008, which were the main driving factors of NDVI in central-western part and eastern part, respectively. During 2009-2020, the proportion of relative importance value for evapotranspiration and air temperature accounted for 38.18% and 26.89%, respectively. Evapotranspiration mainly affected the NDVI of central-northwest part, while the local distribution pattern of NDVI mainly driven by air temperature was relatively scattered. Precipitation had slight effects on vegetation change in Northwest Liaoning. This study clarified the regional vegetation change and the driving climate factors, which could provide scientific basis for vegetation restoration of ecologically fragile areas in Northwest Liaoning.

Key words: NDVI, climate driven, change trend, growing season, Northwest Liaoning