Abstract:
The question of how biotic and abiotic factors interact in determining the structure and function of plant communities has been a central focus in ecological research. Global change such as climate change or land-use changes may shift the balance of importance between abiotic and biotic factors, and thus alter the natural productivity and species composition of plant communities. This may particularly apply to regions already experiencing extreme conditions, e.g. drylands. Drylands, such as the Eastern Mediterranean have been predicted to be severely affected by climate change. In particular, climate scenarios suggest that water availability will decrease substantially due to decreasing precipitation combined with increasing temperatures, and that extreme weather events will become more frequent. However, abiotic factors not only regulate plants directly through their effect on demographic processes, but also indirectly through alterations in the biotic environment of the entire community. This indicates that under climate change, not only the abiotic but also the biotic environment will change, requiring adaptation, plastic response or escape from altered conditions. Unfortunately, the majority of studies predicting the response of single species or whole plant communities have largely overlooked biotic interactions, and they are mostly based on abiotic factors.
In this study, I utilized a Mediterranean plant community which naturally occupies a large climatic range to test the combined effect of abiotic alterations and biotic interactions on shrub seedling dynamics. Since the region has been under grazing for millennia, I further tested how adding a second potential disturbance (here grazing simulated by clipping) alters the response of shrub seedlings to climate change and annual neighbours. I addressed these questions by conducting large field experiments along a natural climate gradient, complemented by Botanical Garden experiments. First, I evaluated if natural (spatial) stress gradients can serve as a proxy for predicting plant dynamics under future climate change. Secondly, I tested the interactive effect of changing abiotic and biotic factors on shrub seedling dynamics and in determining the intensity and direction of plant-plant interactions. Thirdly, I studied what will happen if grazing is added to such a system. Focal species of my experiment was a widely distributed dwarf shrub that occupies large areas in the Mediterranean and that is assumed to encroach if managed incorrectly (e.g. by overgrazing).
Overall I was able to show that abiotic and biotic factors are inevitably linked and in order to give reliable predictions about the future response of species they both need to be considered simultaneously. Particularly because of the complex interplay of abiotic and biotic factors, environmental gradients along an abiotic axis are a poor proxy for predicting the response of plant species to climate change. I furthermore showed that local direct interactions between shrub seedlings and annual neighbours were always more relevant to shrub seedling establishment than climatic conditions. Particularly, simulated climate change never had any direct effect on shrub seedling establishment. However, climate change affected shrub seedling establishment indirectly via its effect on the growing conditions of the neighbouring community. Plant-plant interactions were always important and always negative. Grazing simulated by clipping significantly decreased biomass production by the annual plant community, resulting in a slight release from competitive interactions.
In synthesis, my research emphasizes the overriding importance of biotic interactions for determining plant population and community dynamics and therefore, they must be addressed in ecological climate impact studies.