Abstract:The Pingtan Comprehensive Experimental Community in Fujian Province is an experimental community for cross-strait exchanges and cooperation. As the main coastal protection species in Pingtan, Casuarina equisetifolia is widely distributed in the main Pingtan Island. In this study, normalized difference vegetation index (NDVI) was used to quickly estimate the biomass of C. equisetifolia, which could provide a reference for rapid and non-destructive estimation of single species biomass at the regional scale. Meanwhile, it could provide a basis for evaluating the production potential and management of the main Pingtan Ialand ecosystem. Using RS and GIS technology, the OLI image of the main Pingtan Island in April 2017 was processed. Based on field survey data, the C. equisetifolia information was extracted by knowledge base system (KBS). NDVI was used as the independent variable and C.equisetifolia biomass obtained from the C.equisetifolia formula was used as the dependent variable. Inversion models between NDVI and biomass were established to estimate the C. equisetifolia biomass in the main Pingtan Island. Cluster analysis was used to analyze the C.equisetifolia of each town in the main Pingtan Island. The results revealed that: (1) The overall accuracy of the C.equisetifolia extraction based on KBS was 81.3%. (2)The five common models established passed the significance test. The R values of the five models were all above 0.67, which indicated that there was a correlation between NDVI and the C.equisetifolia biomass. It was feasible to use NDVI for estimating regional biomass. The polynomial fitting effect was the best (R=0.72, MRE=23.00%, RMSE=21.62 t·hm-2, rRMSE= 24.94%). (3) According to the binomial model, the C. equisetifolia biomass amounted to 738 235.98 t,