Abstract: This study aimed to establish the minimum area required for surveys of species diversity in forest communities in Langqi Island, Fuzhou, to improve research accuracy and reduce survey costs. Nest sampling of 20 m × 20 m quadrats was employed to investigate species-area curves of forest communities on the island. Three models (logarithmic function, power function, and logistic function) were selected to fit three types of forest communities, namely Acacia confusa Merr., Eucalyptus citriodora Hook. f., and Celtis sinensis Pers., as well as the species-area relationships of the shrub and herb layers. The optimal fitting model was selected using the Akaike Information Criterion (AIC) to calculate the minimum sampling area for the three species in the three communities. Results showed that: (1) Power function fit models were the best fitting models for the species-area relationships of the three target species in the three forest communities. (2) The minimum sampling areas for the A. confusa community, shrub layer, and herb layer were 113.5, 68.9, and 24.4 m², respectively, when sampling precision was set to 50%–90%. The minimum sampling areas for the E. citriodora community, shrub layer, and herb layer were 153.1, 69.6, and 41.4 m², respectively. The minimum sampling areas for the C. sinensis community, shrub layer, and herb layer were 65.5, 38.0, and 15.2 m², respectively. (3) The minimum sampling areas for the forest community, shrub layer, and herb layer in Langqi Island were 10 m × 15 m, 10 m × 10 m, and 6 m × 6 m, respectively. (4) A positive correlation was found between the minimum sampling area and number of species in the community, which was also influenced by community differences, sampling area, human disturbance, occasional species, and other factors.