基于NDVI数据的江苏省植被物候变化及其影响因子分析

以气候变暖为主要特征的全球气候变化与生态系统的相互作用成为影响可持续发展的重要因素。植被作为陆地生态系统的主要组成部分，在生态环境评价及碳水循环等方面具有重要作用。以江苏省为研究区，利用长时间序列的GIMMS NDVI3g数据集和气象数据，采用Logistic函数法提取该区域过去34 a(1982~2015年)植被生长期物候(Start Of Season SOS，End Of Season EOS)变化的时空分布特征，并用相关性分析法定量确定主要气象因子(温度、降水)对物候变化的贡献。结果表明：①空间上，从江苏省南部到北部，SOS呈递增趋势，EOS呈递减趋势；②时间上，大部分(83.1%)像元的SOS提前，主要分布在江苏省中部及北部地区，大多提前1~2 d/a，69.2%像元的EOS延后，大多延后0~1 d/a；③植被生长期开始SOS/EOS对温度、降水有明显响应，70.5%像元的SOS与温度呈负相关，主要位于江苏省北部及少部分南部地区，55.5%像元的SOS与降水呈负相关，55.2%像元的EOS与温度呈正相关，71.2%像元的EOS与降水呈负相关。整体上，温度的升高导致生长期提前，降水对SOS具有双向作用，秋季物候的影响因子更为复杂，温度和降水的变化并不能导致EOS的提前或者推迟。本研究加深对气候变化与植被生态系统相互作用过程的认识，为未来植被及气候变化分析提供参考。

Changes of Vegetation Phenology in Jiangsu Province and Its Impact Factors based on NDVI Data

Global climate change characterized by temperature increase has been a hot topic of widespread concern，and environmental protection and governance has become a significant issue affecting sustainable development.As the major component of terrestrial ecosystems，vegetation plays an important role in the aspects of ecologicalenvironment assessment and carbon cycling.based on the logistic function method，the long-term Normalized Difference Vegetation Index(NDVI)from GIMMS3g and meteorological data over 1982~2015 were used to calculate the start(SOS)and the end(EOS)of the season of Jiangsu province.The spatial and temporal characteristics of vegetation phenological changes were also investigated.The effects of main meteorological factors(temperature and precipitation)on phenological changes were explored by correlation analysis.The results showed that：①spatially，SOS showed an increased trend while EOS decreased from south to north，②temporally，SOS for most regions(83.1%)featured advanced trends with a rate of around 1~2 days per year，while 69.2% of pixels showed a delayed EOS by 0~1 day per year，and ③vegetation phenology responded well to air temperature and precipitation，as 70.5% and 55.5% pixels of SOS had significantly negative correlations with air temperature and precipitation.However，for EOS，more than half areas，i.e.55.2% of pixels and 71.2% of pixels respectively demonstrated positive correlation with air temperature and negative correlation with precipitation.Overall，a higher temperature trigged an earlier SOS but increased precipitation did not necessarily advance SOS.Furthermore，climate changes in autumn showed complicate effects on EOS that neither changes in temperature nor precipitation can lead to one directional changes of EOS.These results will deepen the understanding of the interaction between climate change and vegetation ecosystem，and provide a reference for future vegetation and climate change analysis.