Thorsten Wiegand1, C.V. Savitri Gunatilleke2, I.A.U. Nimal Gunatilleke2, Xugao Wang3, Zhanqing Hao3, and Andreas Huth1. (1) UFZ Helmholtz Centre for Environmental Research - UFZ, (2) University of Peradeniya, Faculty of Science, (3) Institute of Applied Ecology, Chinese Academy of Sciences
Background/Question/Methods Species rich forests have always fascinated and puzzled biologists. The spatial pattern of tree species distributions could provide significant insights into the mechanisms that maintain high species richness. Our overall aim was to assess if the spatial pattern and interaction structure differed among species rich tropical and temperate forest. To approach this question, we conducted a comparative study including the fully mapped 25ha and 50ha plots of tropical forests at Sinharaja (Sri Lanka) and Barro Colorado Island (BCI, Panamá), respectively, and the 25ha temperate forest plot at Changbaishan (CBS, China). We used point pattern analysis to examine the spatial distribution of tree species (dbh > 10cm) and explored the type and frequency of intra- and interspecific spatial association patterns, approximately disentangling the role of species interactions from that of the environment. More specifically, we (1) tested for any spatial patterning among common species, (2) classified the types of association patterns at different spatial scales, and (3) used specific null models to test selectively for non-random smaller-scale species interactions.
Results/Conclusions (1) We found strong and strikingly similar spatial patterning among forests that changed with scale. (2) Segregation was the dominant association type, but positive associations were rare and occurred mostly at scales below 20m and are replaced by partial overlap at larger scales. (3) Only 5% of all species pairs at the two tropical forests showed significant small-scale interactions, but 33% at the temperate CBS forest. Interactions at CBS were mostly negative and occurred generally between species which shared attributes. Significant spatial interactions were limited to distances not more than 10m away from the stems of the target trees. All forests showed strong segregation patterns, but we found at the temperate CBS forest additional evidence for (competitive) species interactions due to niche overlap not found at Sinharaja and BCI. Segregation may introduce a strong stochastic element and considerably weaken deterministic species interactions. This effect will increase with species richness. We anticipate that our approach will allow for more powerful investigations of spatial structures in plant communities and better understanding species coexistence. An important future research direction would be to decode the rich repository of information encoded in the spatial patterns of species rich forests.