Ever since Diamond’s (1975) treatise on community assembly, ecologists have debated whether the assembly process is stochastic or follows deterministic rules. This debate has focused largely on determining if communities exhibit random or non-random patterns in species co-occurrences, where a non-random pattern implies that deterministic processes, such as competition or environmental filtering, are responsible for community structure. Although several qualitative fish community structure models have been developed, it is still uncertain whether fish communities are generally structured randomly or non-randomly. A recent large-scale study on temperate freshwater fish communities found that a majority of communities are structured non-randomly and that the processes most likely influencing this structure are environmental filtering and predation. However, it is still uncertain if tropical freshwater fish communities exhibit similar patterns. To investigate this question, we collected fish community data from tropical streams throughout Costa Rica and conducted co-occurrence null model analyses to look for significant patterns of community structure. Additionally, we developed a trait dataset for the species investigated and used linear regression to see if traits related to environmental filtering, competition, or predation could predict segregation scores between species.
Results/Conclusions
Unlike temperate freshwater fish communities, tropical fish communities in Costa Rica were structured randomly. Preliminary analyses show that on average, only 13.6 percent of pairwise co-occurrences were significantly non-random. Of the non-random associations, 74 percent were positive whereas 36 percent were negative. This suggests that non-random associations are largely regulated by environmental habitat requirements. This result was supported by the regression analyses. Final regression models only included dissimilarity scores of environmental traits for two of the three metacommunities investigated. Our results indicate that fish communities in tropical streams in Costa Rica are largely structured by stochastic processes. This is likely due to wet-dry season dynamics where wet seasons allow increased dispersal between communities followed by ecological drift. Future work should see if these results are general for all tropical freshwater fish communities and look closer at the random processes underlying community structure and assembly.