Diverse responses of vegetation phenology to changes in temperature and precipitation in Northern China

The start and end of the vegetation growing season (SOS and EOS) determine the length of the vegetation growing season and are necessary parameters to assess the primary productivity and carbon stocks of vegetation as well as the key dates to determine the start and end of soil wind erosion in arid zones. However, how SOS and EOS of different types of vegetation respond to changes in temperature and precipitation is still unclear. In this study, Moderate-resolution imaging spectroradiometer (MODIS)-normalized difference vegetation index data from 2001 to 2015 were used to estimate SOS and EOS for each vegetation type, the temperature and precipitation data recorded by the meteorological stations in the study area were used to calculate the average temperature and precipitation in 19 periods from June of the previous year to November of the following year, Spearman's rank correlation analysis was used to analyze the response relationships of SOS and EOS to average temperature and precipitation in each period. The results showed that SOS, EOS, the average timing of SOS (AvSOS) and the average timing of EOS (AvEOS) all had strong spatial heterogeneity. Precipitation is the key factor controlling SOS and EOS, an increase causing SOS to advance and EOS to be delayed, while higher temperature has the opposite effects. Overall, the average advancement rate of SOS was 0.88 day/year, the average delay rate of EOS was 0.33 day/year. From southeast to northwest, AvSOS gradually changed from 121st day to 141st day, whereas AvEOS was the opposite, advancing from 271st day to 282nd day. We divided the vegetation into 12 types in the soil wind erosion area of northern China according to latitudinal zonation and dominant vegetation species, the results avoided t a shortcoming of previous studies in which SOS and EOS were identified without distinguishing vegetation types.

植被生长季起始（Start of Growing Season，SOS）与终止（End of Growing Season，EOS）决定了植被生长季的时长，既是评估植被初级生产力与碳储量的必要参数，亦是确定干旱区土壤风蚀起止时间的关键节点。然而，不同植被类型的SOS与EOS如何响应气温与降水变化，目前仍不明确。本研究采用2001至2015年的中分辨率成像光谱仪（Moderate-resolution Imaging Spectroradiometer，MODIS）归一化植被指数（Normalized Difference Vegetation Index，NDVI）数据，反演得到各植被类型的SOS与EOS；利用研究区气象站点观测的气温与降水数据，计算了从前一年6月至次年11月共19个时段的平均气温与降水量；并通过斯皮尔曼秩相关分析（Spearman's Rank Correlation Analysis），探究了各时段内SOS与EOS对平均气温、降水量的响应关系。研究结果显示，SOS、EOS以及SOS平均起始时间（Average Start of Season，AvSOS）、EOS平均终止时间（Average End of Season，AvEOS）均存在显著的空间异质性。降水是调控SOS与EOS的关键因子：降水量增加会使SOS提前、EOS延后，而气温升高则会产生相反的影响。整体而言，SOS的平均提前速率为0.88天/年，EOS的平均延后速率为0.33天/年。从东南向西北方向，AvSOS的平均起始时间从第121天逐步过渡至第141天；而AvEOS则呈现相反趋势，其平均终止时间从第271天提前至第282天。本研究根据纬度地带性规律与优势植被物种，将中国北方土壤风蚀区的植被划分为12类；该结果规避了以往研究未区分植被类型即提取SOS与EOS的缺陷。