基于MODIS时序数据北回归线(云南段)地区植被物候时空变化及其对气候响应分析

植被物候作为自然界规律性、周期性事件,对开展全球气候变化、植被长势观测等研究具有重要价值。以北回归线(云南段)穿过的县域为研究区,基于长时间序列MODIS EVI(Enhanced Vegetation Index,EVI)、土地利用类型和气候因子数据,采用S-G滤波、动态阈值、相关分析等方法分析19 a(2001—2019年)间植被物候的时空分布特征及其对水热因子的响应。结果表明,(1)海拔和地势起伏在物候地域分异中作用显著,植被物候存在明显的垂直地带性分布特征。山地与河谷、坝子、低海拔区的物候值差异较大,山地地区的植被生长季开始期(Start of Season,SOS)在192—240 d,生长季结束期(End of Season,EOS)在次年144—192 d,生长季长度(Length of Season,LOS)为272—304 d;河谷、坝子、东部低海拔地区的植被SOS在80—112 d,EOS在337至次年17 d,LOS在224—256 d。(2)19 a间植被物候年际变化总体特征为SOS显著提前(R²=0.51,P=0.001<0.05),平均提前1.14 d·(10 a)^?1;EOS推迟(R²=0.01,P=0.756>0.05),平均推迟0.07 d·(10 a)^?1;LOS显著延长(R²=0.47,P=0.001<0.05),平均延长1.07 d·(10 a)^?1。(3)不同植被类型的物候期及其变化趋势不同,研究区森林植被生长期最长,草地次之,耕地最短;19 a间SOS、EOS、LOS变化最大的分别是常绿阔叶林(?1.68 d·(10 a)^?1)、耕地(1.25 d·(10 a)^?1)、木本热带稀树草原(1.28 d·(10 a)^?1)。(4)水热组合对植被生长影响显著,河谷、坝子、东部低海拔地区的植被SOS、EOS分别主要受2月降水(负相关)和4月气温(正相关)、9月降水和气温(正相关)影响,山地地区植被SOS、EOS分别主要受6月降水(正相关)和5月气温(正相关)、5月降水(正相关)和4月气温(正相关)影响。

Spatio-temporal Variation of Vegetation Phenology and Its Response to Climate in the Tropic of Cancer(Yunnan section)Based on MODIS Time-series Data

Vegetation phenology,as a regular and periodic event in nature,has an important theoretical significance and practical value for research on global climate change and the detection of vegetation growth.Taking all the counties crossed by the Tropic of Cancer(Yunnan section)as a study area,and based on MODIS EVI time-series data,weather station data,MCD12Q1 data,LST data and TRMM data,we used a Savitzky-Golay filter,dynamic threshold method and Pearson correlation analysis to analyze the spatiotemporal variation of vegetation phenology and its response to climate between 2001 to 2019.The results show that:(1)Elevation and relief play a significant role in phenological regional differentiation in the study area,and vegetation phenology has obvious vertical zonal distribution characteristics.There are significant differences in phenological values between mountain,valley,basin,and low-altitude areas.In mountainous areas,the Start of Season(SOS)is between the 192nd to 240th day,the End of Season(EOS)is between the 144th to 192nd day in the next year,and the Length of Season(LOS)is between 272 to 304 days.In the valley,basin,and low-altitude areas,the SOS is between the 80th to 112th day,the EOS is between the 337th to 17th day in the next year,and the LOS is between 224 to 256 days.(2)The overall characteristics of the interannual change of vegetation phenology in the Tropic of Cancer(Yunnan section)in the past 19 years showed that SOS had a significant advance trend(R²=0.51,P=0.001<0.05),1.14 d·(10 a)?1 in advance on average;EOS has a tendency to delay(R²=0.01,P=0.756>0.05),the average delay was 0.07 d·(10 a)^?1;LOS interannual changes show a significant prolongation trend(R²=0.47,P=0.001<0.05),with an average extension of 1.07 d·(10 a)^?1.(3)Different vegetation types have different growth periods.The growth period of forest vegetation is the longest,grassland vegetation is the second longest,and cultivated land vegetation is the shortest.The greatest variation among SOS,EOS and LOS between 2001 to 2019 was evergreen broad-leaved forest(?1.68 d·(10 a)^?1),cultivated land(1.25 d·(10 a)^?1)and woody savanna(1.28 d·(10 a)^?1).And(4)hydrothermal combination had significant influence on vegetation growth.The SOS was negatively correlated with precipitation and positively correlated with temperature,while the EOS was positively correlated with both temperature and precipitation.Among them,the precipitation in February(negative correlation)and the temperature in April(positive correlation)had the greatest influence on the SOS in the valley,basin and the low altitude area in the east.Precipitation in June(positive correlation)and temperature in May(positive correlation)had the greatest influence on SOS in mountainous areas.Precipitation and temperature in September(both positively correlated)had the greatest influence on EOS in the valley,basin and low-altitude area in the east,while precipitation in May(positively correlated)and temperature in April(positively correlated)had the greatest influence on EOS in mountainous areas.