The distribution of mangroves is influenced by the environment. We aimed to understand the ecological adaptability of various mangrove species within the range of the exotic species, Sonneratia apetala Buch.-Ham., in Dongzhai Harbor, Hainan Island, China. We used three niche breadth indexes (Simpson, Levins, and Shannon-Weiner) and two niche overlap indexes (Pianka and Levins) to quantitatively determine the niche characteristics of nine mangrove species. The results showed that the order of the niche breadth values of mangrove species was as follows: Aegiceras corniculatum (Linn.) Blanco > Kandelia obovata Sheue et al. > Bruguiera gymnorrhiza (L.) Poir. > Avicennia marina (Forsk.) Vierh. Hailanci > S. apetala > S. caseolaris (L.) Engl. > Rhizophora stylosa Griff > Ceriops tagal (Perr.) C. B. Rob. > B. sexangula (Lour.) Poir. Pearson correlation analysis revealed that the niche breadth of each population was significantly correlated with the importance value of the population in the whole sample (R₁ = R₂ = 0.771, R₃ = 0.644, P < 0.05). The nine mangrove species were divided into three groups by Bray–Curtis cluster analysis; the groups were similar to the distribution of mangrove species in the natural state as determined by tide level. Niche similarity analysis showed that the niche similarity of most mangroves ranged between 0.5 and 0.8 and that the species pairs A. corniculatum–B. gymnorrhiza, A. corniculatum–Avicennia marina and K. obovata–S. caseolaris were characterized by large niche similarity ratios. Although it had a moderate niche breadth, S. apetala had a relatively broad niche overlap with mangroves in the mid- and low-tide zones (S. caseolaris, A. corniculatum, K. obovata, and Avicennia marina), a moderate overlap with B. gymnorrhiza and R. stylosa, only a slight overlap with C. tagal, and no overlap with B. sexangular. There was no obvious linear relationship between niche width and niche overlap of mangroves. Due to its inefficiency in utilizing certain resources and relatively high degree of resource selection, it seems likely that S. apetala will not pose a threat to the survival of native plants, let alone completely replace native species.

The experimental plots used in this study were located in the Sanjiang Mangrove Area of Dongzhai Harbor Nature Reserve (the area of mangroves here is approximately 1000 ha). In 1989, approximately 1 ha of S. apetala plantation was located in this forest area and by 1996, the area of S. apetala plantation had increased to approximately 10 ha. S. apetala dispersed primarily upstream and downstream along both sides of the Yanzhou and Meipo rivers. The range of S. apetala dispersal and colonization has occurred within an area characterized by a successional series of Kandelia obovata Sheue et al. + Aegiceras corniculatum (Linn.) Blanco communities. The successional community types in this area and their substitution order are as follows: K. obovata + A. corniculatum (Community I) → A. corniculatum + K. obovata (Community II) → A. corniculatum + K. obovata + Bruguiera gymnorrhiza (L.) Poir. (community III) → A. corniculatum + B. gymnorrhiza + K. obovata (community IV).

The experimental plots (measuring 10 m × 10 m) were all located within these communities, among which 32, 10, 10, and 14 plots were established in communities I, II, III, and IV, respectively. Mangroves in the quadrat that were over 2 cm of ground diameter were measured and the name of species, number of individuals per species, coverage, diameter at ground, and tree height were recorded. Although small numbers of Acanthus illicifolius L. and A. illicifolius var. ebracteatus (Vahl) Williams were also distributed in each community, these were not included in data collection owing to their small size and low numbers.