基于Whittaker滤波的陕西省植被物候特征

Remote Sensing Analysis of Vegetation Phenology Characteristics in Shaanxi Province Based on Whittaker Smoother Method

运用Whittaker滤波重构MODIS NDVI时序数列,利用地理探测器对比滤波前后影像信噪比,采用动态阈值法获取2000—2012年陕西省植被的3个关键物候参数(返青期、枯黄期和生长周期),在此基础上分析了该区植被物候参数空间分布特征。结果表明:(1)Whittaker滤波能够平滑原始NDVI曲线,有效减少原始影像的噪声,提高影像辨识度,并且参数设置简单;(2)陕西省植被物候地区分异明显,不同气候区划类植被物候表现出中温带半干旱区-暖温带半干旱区-暖温带半湿润区-北亚热带湿润区的递变规律:返青期逐步提前,枯黄期逐步推迟;(3)植被物候受高程和纬度影响,并且纬度影响更显著。海拔每升高200m,返青期推迟1.3d,枯黄期提前0.6d;纬度每升高0.5°,返青期推迟3.6d,枯黄期提前1.2d。

This paper used Whittaker smoother to reconstruct MODIS NDVI time series and geographical detection method to compare the recognizable of images before and after filtering.Using dynamic threshold to get three phenological parameters(the start of season-SOS,end of season-EOS,and length of seasonLOS)from 2000 to 2012,and then analyzed the spatial distribution characteristics of vegetation phenology in Shaanxi Province.The result shows that:(1)the Whittaker smoother with simple parameter setting can smooth the NDVI time series,reduce the noise of the original image effectively,and enhance the recognizable of images;(2)the vegetation phenology of Shaanxi Province has significant regional differentiation,and considering for different climate zones,three phenological parameters all performed the order of temperate semiarid region,warm temperate semiarid region,warm temperate semi-humid zones,north subtropical zone:SOS gradually advanced and EOS gradually postponed;(3)vegetation phenology is affected by the altitude and latitude,and the latter was more significant.SOS postponed 1.3 d and EOS advanced 0.6 d for every 200 m higher in elevation.SOS postponed 3.6d and EOS advanced 1.2d for every 0.5 of latitude northward.