Identification And Analysis Of Ecosystem Functional Types Based On Sentinel-2 Image

The characteristics of ecosystem functions are of great significance for biodiversity conservation and ecosystem services.Ecosystem function types(EFTs)refers to a kind of ecosystem groupings based on similar ecosystem functions,which are used to represent the spatial pattern and temporal changes of key ecosystem functional traits.Ecosystem functional types(EFTs)differs from the structure and composition of vegetation and represents the spatial heterogeneity of ecosystem functions.Obviously,the identification of EFTs based on low-resolution remote sensing data cannot satisfy the needs of fine-scale characterization of regional ecosystem functionality patterns.To explore how to use remote sensing data sets with high spatialtemporal resolution and the optimized identification methods to realize fine-scale EFTs classification and reveal the diversity pattern of ecosystem function types,is in line with the development direction of quantitative research on regional ecosystem functions.In this study,EFTs at a county scale is characterized based on the subtractive fuzzy cluster means(SUBFCM)and Sentinel-2 time series data during 2016～2019.The normalized difference vegetation index(NDVI),surface albedo,the fraction of absorbed photosynthetically active radiation(FAPAR),and canopy water content(CWC)and their derived variables in the growing season(May to October)were selected as ecosystem functional indicators to characterize regional EFTs diversity pattern.Combined with the statistics of classification indicators,the main characteristics of each EFTs are summarized,and the relationship between EFTs and land cover types is investigated by correspondence analysis(CA)and Minnick’s coefficient.Through the landscape pattern index,richness and rarity index,the pattern characteristics of ecosystem function types are analyzed.Based on soil,climate and topographic elements,and the factors such as soil type,average temperature,elevation,etc.were selected.The spatial distribution of EFTs was preliminary discussed.The results showed that the expanded selection of variables reflecting the carbon benefits of regional ecosystems,surface energy balance and water characteristics are suitable for the functional classification of ecosystems.The SUBFCM algorithm can automatically divide ecosystem functional types with faster convergence speed and reduced subjectivity.The obtained EFTs based on Sentinel-2 images reflected the internal structure of carbon balance well and the distribution pattern of ecosystem functional diversity in the Zhenlai county.The selected influencing factors comprehensively explain the spatial differentiation of EFTs in the Zhenlai county.Among them,as the most basic element,soil type has a very important influence.The aspect,slope,average temperature and average relative humidity in the growing season have little.The cumulative precipitation and elevation have a greater impact.This study provides a reference for the identification of regional EFTs at a fine scale based on highresolution satellite data and optimized clustering methods,as well as a further understanding of the spatial heterogeneity of temperate terrestrial ecosystems.