Community succession patterns - a function of trophic position, sampling method, and depth of sampling?

Background/Question/Methods

Understanding the mechanisms underlying successional patterns of communities has resulted in research varying from investigating the importance of intrinsic (biotic interactions) versus extrinsic (resource input, disturbances) factors in directing the successional trajectory of a community, to testing community assembly rules and incorporating metacommunity approaches.  However, these research efforts have often resulted in multiple and sometimes conflicting answers, making generalizations even within one type of community still difficult to obtain.  To address such discrepancies in the literature and across sampling methods and study systems, I followed the model Sarracenia purpurea pitcher plant system in replicate communities through an entire growing season in order to determine the changes in abundances, species richness, and diversity throughout the successional trajectory of the community.  These sampling techniques allowed me to explore 1) if community successional patterns differed depending on which trophic level I used as the ‘community’ 2) if community succession differed when the entire food web was the ‘community’, and 3) how different sampling methods in the bottom trophic level affected whether communities converged or diverged in similarity. Specifically, I was interested if agar plate counts gave the same pattern of succession as 16s rRNA gene assessment for the bottom trophic level.

Results/Conclusions

I found that community succession patterns depend on the trophic level examined and if the entire community instead of separate trophic levels is used. Furthermore, culturable bacteria of the bottom trophic level appear to go through pulses of convergence and divergence, with the convergence pattern being highly predictable.  The protozoan species of the intermediate trophic level, however, converge to be similar after the first month and continue to be similar until some unknown force at the end of the season makes the communities diverge. When OTU data was used, I found that communities at the beginning of the season were highly different in both abundance and presence/absence of species when compared to communities at the end of the season.  This pattern is similar when I examined the abundances of culturable bacteria collected from the same samples used for the OTU, but not when comparing the presence/absence of the species in OTU data and culturable bacteria data. It can be concluded from this study that how a community is defined is important for understanding how it is perceived to go through succession and that the sampling method used can alter the result.