Assessing the Impacts of Species Composition on the Accuracy of Mapping Chlorophyll Content in Heterogeneous Ecosystems

Chlorophyll is an essential vegetation pigment influencing plant photosynthesis rate and growth conditions. Remote sensing images have been widely used for mapping chlorophyll content in different ecosystems (e.g., farmlands, forests, grasslands, wetlands) evaluating status productivity of these ecosystems. Compared to farmlands forests that are more homogeneous terms species composition, grasslands wetlands heterogeneous with highly mixed various grass, forb, shrub species). Different contribute differently the ecosystem services, thus, monitoring species-specific critical better understanding their status, functions, supporting management control invasive However, previous studies rarely estimated content, which was partially due limited spatial resolution remote commonly past few decades recognizing species. In addition, many one universal model built data all entire study area, did not fully consider impacts composition on accuracy estimation (i.e., establishing models may generate higher accuracy). this study, helicopter-acquired high-spatial hyperspectral were acquired classification estimation. Four models, including a linear regression (LR) species), LR each species, respectively), random forest (RFR) model, RFR compared determine performance evaluate composition. The results show performed than especially fewer samples dataset. best then maps using results. Impacts retrieval further assessed support future management.