基于1990-2015 归一化植被指数的河北植被时间变化特征及其对干旱的响应

基于1990-2015年研究区逐旬归一化植被指数（Normalized Difference Vegetation Index, NDVI）数据和站点逐日降水数据，对河北地区NDVI时间变化特征及其与多尺度标准化降水蒸散指数（Standardized Precipitation Evapotranspiration Index, SPEI）的响应关系进行了深入分析。结果表明：（1）研究区草地、林地面积增加，空间上呈自东向西交替变化，并呈破碎化现象。（2）年均及生长期NDVI呈明显的上升趋势，夏季NDVI最高，冬季最低，月均NDVI时间变化与雨热同期。（3）不同尺度SPEI均呈下降趋势，随着时间尺度的增加，降幅不断加剧。干旱整体上发生的持续时间和干旱程度有所降低，而春季发生干旱明显强于其他季节。（4）年际及夏季NDVI对SPEI变化响应最为显著，夏季NDVI与SPEI06之间的相关性为最高的0.628，春秋次之，冬季最低。年际及季节性NDVI变化与SPEI03及SPEI06变化的相关性较高。年际及季节NDVI变化对SPEI的响应较单独的气候因子（气温、降水）更明显。

Temporal variations of vegetation and its response to drought based on 1990-2015 Normalized Difference Vegetation Index data in Hebei province

Based on the Normalized Difference Vegetation Index (NDVI) data from 1990-2015 and daily precipitation data of 20 meteorological stations, the NDVI time variation characteristics and its response relationship with multi-scale standardized precipitation evapotranspiration index (SPEI) in Hebei Province were analyzed. The results show that: (1) The area of grassland and forest land in the study area increases, and the occupied space alternates from east to west with a fragmentation phenomenon. (2) The annual average and growth NDVI show a clear upward trend. In addition, the NDVI in summer is the highest and that in winter is the lowest. The monthly average NDVI has the same trend to monthly temperature and precipitation. (3) Multi-scalar SPEI shows a downward trend. Specifically, the decline is intensified with the increase of time scales from 3 to 24 months. Overall, the duration and degree of drought is decreased, while the drought in spring is significantly stronger than other seasons. (4) During the interannual and summer periods, the response of NDVI to SPEI is significant. In particular, the correlation coefficient of NDVI in summer to SPEI06 is the highest 0.628, followed by spring and autumn and winter. In addition, the correlation between interannual and seasonal NDVI, and SPEI03 and SPEI06 is significantly higher than SPEI12 and SPEI24, which the shorter time scale is more suitable for the analysis of vegetation dynamics in the study area. Moreover, it is revealed that interannual and seasonal NDVI changes are more pronounced for SPEI than for individual climatic factors (temperature, precipitation).