[目的] 研究黄土高原地区植被变化及其对极端气候的响应，为减缓和应对气候异常提供科学依据。[方法] 基于1982-2017年遥感影像数据和气象数据，采用趋势分析、相关分析等方法，研究黄土高原地区植被时空变化及其对极端气候的响应。[结果] 1982-2017年期间，黄土高原NDVI以每年0.37%的速率呈显著的增加趋势（p<0.01）；空间上，NDVI呈现从西北到东南递增的空间分布格局。极端气候指数变化中，极端气温指数变化趋势较为一致，即表征极高温事件的极端气温指数呈极显著的增加趋势，表征极低温事件的指数呈现显著的下降的趋势，而极端降水指数未发生显著变化。NDVI年际变化与极端气温指数FD0，TMAXmean，TMINmean，TN10p，TN90p，TR20，SU25均呈极显著相关（p<0.01）；四季NDVI变化与极端降水指数均未表现出明显的相关性，但与极端气温指数显著相关且春季和夏季的相关性高于秋季和冬季；月尺度上，NDVI与极端降水指数（RX1day，RX5day）和极端气温指数（TMAXmean，TMINmean，TN90p，TXx，TNn）呈显著的相关性（p<0.01）。NDVI与极端气温指数TMAXmean，TN10p，TN90p，TXx前1个月的相关性大于当月、前2个月、前3个月的相关性。[结论] 黄土高原地区NDVI呈显著增加的趋势，年际和月际NDVI变化与极端气温指数存在相关性，而与极端降水指数均未表现出明显的相关，且黄土高原地区的植被覆盖变化对极端气候的响应存在一定的滞后性。
[Objective] The dynamic changes of vegetation on the Loess Plateau and its response to extreme climate were studied, in order to provide scientific basis for mitigating and adapting to climate anomaly.[Methods] Based on the remote sensing data and meteorological date from 1982 to 2017, the study used Mann-Kendall trend test and correlation analysis method to analyze temporal and spatial changes of vegetation in Loess Plateau and its response to extreme climate.[Results] NDVI exhibited a significant increasing trend with an annual rate of 0.37% (p<0.01) from 1982 to 2017, and increased from northwest to southeast on the Loess Plateau. As for the change of extreme climate, the variation trend of extreme temperature was relatively consistent. The extreme high temperature events showed a significant increasing trend, and the extreme low temperature events showed an extremely significant downward trend. However, the extreme precipitation index did not change significantly. The correlation analysis between extreme climate index and NDVI showed that the interannual variation of NDVI was significantly correlated with extreme temperature index FD0, TMAXmean, TMINmean, TN10p, TN90p, TR20 and SU25 (p<0.01). There was no significant correlation between NDVI changes and extreme precipitation index in all seasons. However, NDVI showed a significant correlation with extreme temperature index, moreover, the correlation of spring and summer was higher than that autumn and winter. On the monthly scale, NDVI was significantly correlated with extreme precipitation index (RX1day, RX5day) and extreme temperature index (TMAXmean, TMINmean, TN90p, TXx, TNn) (p<0.01). The correlation between NDVI and extreme temperature index TMAXmean, TN10p, TN90p and TXx a lag of one month was greater than that current month, a lag of two months and a lag of three months.[Conclusion] The NDVI showed a significant increasing trend. The interannual and monthly NDVI changes were correlated with the extreme temperature index, and both showed a significant correlation with the extreme precipitation index. Moreover, the response of vegetation cover change in the Loess Plateau region to extreme climate has a certain lag.