Geological age drives community structure: Arthropod diversity in a space for time chronosequence on the Hawaiian Islands

Background/Question/Methods

In classic community assembly theory, processes driving community species composition and diversity are theorized to be deterministic and take place on relatively short time scales. In reality, contemporary community structure is the result of longer timescales, with historical effects and in situ diversification influencing the regional species pool. Furthermore, the effect of ecological processes such as biotic and abiotic interactions may also vary with succession and community age. To address the question of dynamic drivers of diversity and species composition, we study effects of age on arthropod communities, through geological history and consequent nutrient status and forest structure changes, as well as through successional changes in dominant canopy tree species traits. 

We use a space-for-time chronosequence on the Hawaiian Islands to examine the relative importance of local ecological variables and geological age in the structure of canopy arthropod communities. Local communities within the Hawaiian archipelago span an expanse of substrate ages ranging from historic (<50 years) to Pleistocene (500000+ years). Furthermore, mesic forests on the Hawaiian Islands are dominated by a single tree species Metrosideros polymorpha, allowing us to compare arboreal communities across substrate age. We selected sites in native mesic forest and sampled arthropod abundances using canopy branch clipping.

Results/Conclusions

While preliminary, our findings show that arboreal arthropod abundances are predicted by substrate age in a unimodal fashion, peaking at intermediate aged sites. Multivariate analysis indicates that communities separate in abundances per order along substrate age and geological history, as well as having a characteristic age related community composition at the order level. Sites also separated on volcanic origin of substrate, a factor both associated with age and substrate type. Abundances of the dominant orders were predicted by age and site, irrespective of environmental variables. Results show that local variables such as individual tree height and Metrosideros morphotype explain some of the site variation in arthropod abundances.

 Further analysis will demonstrate changes in arthropod diversity and feeding guild structure across age. Our data on local tree morphology, foliage nutrient status and forest structure will allow direct linking of arthropod community composition and diversity to landscape level differences in geology and substrate age. Understanding the ecological factors that scale with substrate age will allow for insight into the balance between ecological and evolutionary processes in driving arthropod abundances and diversity in early ecosystem development.