基于归一化差值植被指数(normalized difference vegetation index,NDVI),提出了一种新的物候遥感监测方法,基于森林植被的年NDVI时间累积曲线,利用Logistic模型对NDVI累积曲线进行拟合,依据曲率极值方法提取森林植被物候期的关键参数(生长季开始日期,SOS; 生长季结束日期,EOS),并对森林植被的生长季长度(length of season,LOS)进行分析,探讨近10年东北森林物候的时空变化。主要结论为: (1)2001年~2010年间,东北森林生长季开始日期集中在110~140天但在10年间没有明显变化;(2)第260~290天,森林逐渐停止生长,生长季结束日期从北向南逐渐推迟,但在十年间几乎没变化; (3)与生长季开始和结束日期相对应,东北森林生长季长度集中在120~160天之间,但存在空间差异,大兴安岭地区森林生长季长度较短,集中在120~140天之间,小兴安岭、长白山、辽东半岛地区的森林生长季长度可达到160天,对整个研究区来讲,近10年间变化的区域仅占研究区的14.9%,变化趋势集中在1d/10年。研究结果与物候观测数据及已有的研究具有较好的一致性,说明利用遥感数据动态监测东北森林植被物候期具有一定的可靠性。
英文摘要:
Plant phenology is the best indicator of terrestrial ecosystem response to climate change and it becomes a hot issue in the study of global change. The forestin northeast of China plays an important part in global forest ecosystem. In this paper, yearly integrated Normalized Difference Vegetation Index (NDVI) of forest vegetation in northeast China was calculated based onSpot Vegetation datasets from 2001~2010, which has been filtered using Savtiky-Galoy method. And then, the yearly integrated NDVI profile was fitted using a logistic model. Two key parameters of forest phenology (start of season, SOS; end of season, EOS) were extracted according to the greatest rate of curvature of fitted cumulative NDVI and the length of forest phenology (length of season, LOS) was also analyzed. The main conclusions of this paper are (1) SOS mainly occurs in the 110th~140th day and EOS in 260th and 290th day. SOS displays a marked delayed from south to north while EOS gradually advances. However, the changes of SOS and EOS in ten years are not obvious. (2) Corresponding to the SOS and EOS, LOS of forest in study area mainly occurs in the 120th~160th day; however, it is spatiallyheterogeneous. LOS of forest in Greater Khingan Mountains is shorter (about 120~140 day) than forests in Xiao Hinggan Ling and Changbai Mountains (about 160 day).The results in this paper are concordant with records of phenology in situ measurements and previous researches in the same area. It indicates that forest phenophases using method in this paper from Spot Vegetation dataset is feasible.