Using HHT to Successfully Uncouple Seasonal and Interannual Components in Remotely Sensed Data

摘要

This paper shows how the EMD can successfully uncouple the seasonal and interannual components of NDVI signals from induced satellite artifacts. The known delay in crossing the equator of the afternoon National Oceanic and Atmospheric Administration (NOAA) polar satellites results in changes of illumination that affects measurements made by the advanced very high resolution radiometer (AVHRR). Despite the improved processing, correction of calibration loss and atmospheric effects, one of the standard AVHRR products, the normalized difference vegetation index (NDVI) time series, still may contain variations due to orbital drift or changes in sun-target-sensor geometry. In this study, we identify solar zenith angle (SZA) trends associated to orbital drifts and analize their effects on NDVI. We present the adaptive empirical mode decomposition (EMD) method as a mean to identify and remove the induced SZA trend from NDVI time series. The EMD is based on the local characteristic time scale of the data, and it is usable to give sharp identifications of embedded structures of nonlinear and nonstationary processes. The approach was tested on 1 degree NDVI global data sets. The results show : (1) that regions in the tropics, especially tropical forest, are the most SZA-affected, and (2) the high northern latitudes are less contaminated.