黄土高原植被物候变化及其对季节性气候变化的响应

受气候变化影响，全球范围内植被物候发生了显著变化，而目前针对不同植被分区类型下（荒漠草原区、典型草原区、森林草原区、落叶栎林区、落叶栎林亚区）植被物候变化及其对季节性气候变化响应的研究尚少。因此基于MODIS遥感归一化差值植被指数（MODIS NDVI：MOD13Q1）数据、中国植被区划数据及135个气象站点插值数据，利用Sen''s斜率估计、Hurst指数和高阶偏相关分析等方法，研究黄土高原2001-2018年植被物侯变化及其对季节性气候变化的响应。结果表明：（1）黄土高原植被生长季始期（SOS，Start of Growing Season）主要集中在第96-144天，子植被分区由西北向东南方向，逐渐呈现提前趋势，71.0%的像元植被SOS整体提前0-2 d/10a （α=0.05），且在未来一段时间66%的像元植被SOS继续呈现提前趋势；植被生长季末期（EOS，End of Growing Season）主要集中在第288-304天，各子植被分区植被EOS变化基本保持一致，87.6％的像元植被EOS整体延迟0-3 d/10a （α=0.05），且在未来一段时间有80%的像元植被EOS继续呈现推迟趋势。（2）黄土高原植被SOS主要受各季节温度的影响；当年春季降水导致植被SOS提前，主要分布在黄土高原中部；上年夏季和上年秋季降水增加会导致植被SOS推迟；当年春季、上年秋季和年初冬季的温度升高均会导致植被SOS提前；各子植被分区植被SOS对不同季节降水的响应存在差异，而对不同季节温度的响应具有一致性。（3）黄土高原植被EOS主要受各季节降水和秋季温度的影响；不同季节降水增加均会导致大部分植被EOS推迟；当年秋季温度导致整体区域植被EOS推迟，且各子植被区植被EOS对当年秋季温度响应具有一致性。该研究可为大尺度植被物候影响因素提供新的认识，也为植被适应未来气候变化提供借鉴。

Vegetation phenology change and its response to seasonal climate changes on the Loess Plateau

Due to the influence of climate change, the global vegetation phenology has undergone significant changes, but there are few studies on vegetation phenology change and its response to seasonal climate change for different vegetation zoning. Therefore, based on MODIS remote sensing normalized difference vegetation index data (MODIS NDVI:MOD13Q1), vegetation zoning map of China and interpolation data from 135 meteorological stations, this paper divided the Loess Plateau into five vegetation subzones, and used Sen''s slope estimation, Hurst index and high-order partial correlation analysis to study vegetation phenology change and its response to seasonal climate in different vegetation zoning during 2001 to 2018 over the Loess Plateau. The results showed that:(1) The start of growing season (SOS) of vegetation in the Loess Plateau is mainly concentrated in the period 96-144 days. For different vegetation zones, the advance trend of vegetation phenology was gradually enhanced from northwest to southeast. The SOS is advanced by a rate of 0-2 d/10a over 71.0% of the pixels, and continued to show an advance trend in the future over 66% of the pixels. The end of growing season (EOS) mainly concentrated in the 288-304th day of autumn. The EOS in different vegetation zone was basically consistent. EOS was delayed by a rate of 0-3 d/10a over 76.5% of pixels, and continued to show a trend of postponing in the future over 80% of pixels. (2) The vegetation SOS of the Loess Plateau is mainly affected by the temperature of each season; the precipitation in the spring of that year leads to the advancement of the vegetation SOS, which is mainly distributed in the middle of the Loess Plateau; the increase of precipitation in the summer and autumn of the previous year will cause the vegetation SOS to be delayed; the temperature rise in spring of the same year, last autumn and early winter will lead to the advancement of vegetation SOS; the response of vegetation SOS in each sub-vegetation zone to precipitation in different seasons is different, and the response to temperature in different seasons is consistent. (3) The EOS of the vegetation in the Loess Plateau is mainly affected by the precipitation of each season and autumn temperature; the increase of precipitation in the different seasons will cause the delay of most vegetation EOS; the autumn temperature of that year caused the delay of the vegetation EOS of the whole area, and the vegetation EOS of each sub-vegetation area is affected. The temperature response in the fall of that year was consistent. This study can provide a new understanding of the influence factors of largescale vegetation phenology and provide reference for the adaptation of vegetation to future climate change.