Interspecific associations, which strongly relate to or affect the structure and dynamics of ecosystems, are still poorly understood especially in tropical forests with the greatest plant alpha diversity in the world. In this study, we used long-term census data (census years: 1982, 1985, 1990, 1995, 2000, 2005) from a 50-ha tree plot in the neotropical forest of Barro Colorado Island (BCI), Panama to examine interspecific associations and their relationships with species distributions over the past several decades. The following hypotheses were tested: interspecific associations are species-specific and vary significantly with spatial scale and temporal variation; and interspecific associations are related to species distribution. We used Ripley’s K-function and the related functions (L₁₁(h), L₁₂(h)) to analyze intraspecific and interspecific associations varying with different spatial scales (5, 10, 15, 20, 25, 30, 50, 100, 150, 200, 250 m). A Monte-Carlo simulation was used to test the statistical significance of deviations of L₁₁(h) and L₁₂(h) from zero under the null hypothesis of complete spatial randomness.

Results/Conclusions

Based on the analyses of 42 common tree species (diameter at breast height >100 mm with more than 100 nonjuvenile trees in the 50-ha plot), the results showed: about 30% of all species pairs showed significant interspecific associations (positive or negative) at some spatial scale, which suggests that interspecific associations play very important roles for community assembly in the BCI forest. The ratio of species pairs with significant interspecific associations to all species pairs does not vary greatly with different census years, but species associations of some species pairs at some spatial scales vary with the census. Interspecific associations for most species pairs of the BCI forest vary greatly with spatial scale, and the ratio of species pairs with significant interspecific associations to all species pairs decreases with increasing spatial scale. We also analyzed the relationships between interspecific associations and species distributions in the BCI forest plot. This research will contribute to a better understanding of the mechanisms of species coexistence and community assembly in the neotropical forest.