基于长时序遥感的京津冀森林物候特征分析

Analysis of Forest Phenological Characteristics in Beijing-Tianjin-Hebei Based on Long-time Series Remote Sensing Data

森林是全球生态系统的重要组成部分,由于植被物候对于气候变化具有一定的敏感性,在全球气候变暖背景下,研究森林物候变化对深入理解陆地生态系统对全球气候变化响应方式具有重要意义。研究基于MODIS-NDVI长时间数据,采用动态阈值法提取了京津冀地区2002—2020年森林物候期参数,分析了近19年来京津冀地区森林物候变化时空变化特征,利用同期降水和气温数据,进一步探讨了森林物候对气候要素变化响应。结果如下:①京津冀地区森林物候具有明显的纬向空间分布特征,即纬度每上升1°,生长季开始时间(SOS)、生长季结束时间(EOS)和生长季长度(LOS)分别推迟2.25 d、提前2.1 d和缩短4.35 d。SOS为89~113DOY,EOS为254~273DOY,LOS为143~182 d;②2002—2020年森林SOS年际变化率为-4.5 d/10 a,EOS年际变化率为4.2 d/10 a,LOS年际变化率为8.7 d/10 a;③海拔每上升100 m,SOS推迟0.8 d,EOS提前0.8 d,LOS缩短1.6 d;平原地区的森林物候变化幅度比在其他地貌的变化幅度更大;④森林SOS受冬季(12—次年2月)降水和温度影响,EOS主要受到7—9月温度和降水的影响。

Forest is an important part of global ecosystems.In the context of global warming,due to the sensitivity of vegetation phenology to climate change,studying forest phenological change is of great significance for in-depth understanding how terrestrial ecosystems respond to global climate change.Based on the long-term data of MODIS-NDVI,the dynamic threshold method is used to extract the forest phenological parameters from 2002 to 2020 in the Beijing-Tianjin-Hebei region,and the temporal and spatial changes of forest phenology in the area of the past 19 years are analyzed.The response of forest phenology to climate change is further explored by using the temperature and precipitation data of the same period.The results are:①Forest phenology in the Beijing-Tianjin-Hebei region had obvious zonal spatial distribution characteristics,that is,the Start of Growing Season(SOS),the End of Growing Season(EOS)and the Length of Growing Season(LOS)were delayed by 2.25 d,advanced by 2.1 d and shortened by 4.35 d with every 1°increase in latitude,respectively.SOS mainly occurred on 89~113th Day of Year(DOY);EOS mainly occurred on 254~273th DOY;LOS was concentrated within 143~182 d.②From 2002 to 2020,the interannual change rate of forest SOS was-4.5 d/10 a,the interannual change rate of EOS was 4.2 d/10 a,and the interannual change rate of LOS was 8.7 d/10 a.③For every 100 m increase in altitude,SOS was delayed by 0.8 d;EOS was advanced by 0.8 d;LOS was shortened by 1.6 d.Forest phenological changes were greater in plains than in other landscapes.④SOS was mainly related to precipitation and temperature in winter(from December to February),while EOS was mainly affected by the temperature and precipitation from July to September.