Monitoring Ecosystem Functional Heterogeneity Pattern and Changes from Lands at Imagery

Satellite-based Ecosystem functional types (EFTs) have been used to characterize the spatial patterns of ecosystem functioning at the global and regional scales. Most current studies of EFT identification were conducted using coarse spatial resolution images and paid little attention to its change. Herein, the Normalized Difference Vegetation Index (NDVI), land surface temperature (LST) and land surface albedo retrieved from Landsat images were used as the ecosystem functional indicators.The combination of self-organizing map (SOM) and fuzzy c-means (FCM) clustering was employed to identify EFTs. We used correspondence analysis method to investigate the consistency of EFTs and land cover types. The results showed that the combination clustering algorithm improved the objectivity in EFTs identification. The spatial pattern of ecosystem functioning varied in different years. The association between EFTs and land cover was complicated.