COS 109-6 - Climate change and habitat loss interact to alter host-parasitoid network structure and function

Wednesday, August 9, 2017: 3:20 PM
D138, Oregon Convention Center
Mark V. Murphy, School of Biological Sciences, University of Western Australia, Jason M. Tylianakis, School of Biological Sciences, University of Canterbury; Department of Life Sciences, Imperial College London, Rachel J. Standish, Environmental & Conservation Sciences, Murdoch University, Perth, Australia and Raphael K. Didham, School of Biological Sciences, The University of Western Australia; Centre for Environment and Life Sciences, CSIRO Land and Water
Background/Question/Methods

Human influence on the natural environment operates via multiple drivers, and at multiple spatial and temporal scales. These global change drivers are still largely studied and managed as independent processes, despite widespread evidence for synergistic interactions, such as between habitat loss and climate change. How such synergies ultimately affect complex communities of interacting species and the ecological services they support is an active area of research, but our understanding is limited by the difficulties in predicting the future effects of rapid climate change from space-for-time substitution studies alone. In the global biodiversity hotspot of southwest Australia, significant declines in annual rainfall since the mid-20thcentury provide an important opportunity to study how a direct temporal shift in climate aridity has exacerbated land-use impacts on ecological networks. In a large-scale mensurative experiment, we conducted a factorial cross of rainfall decline (0-7.5% vs 7.5-15%) and habitat loss (60-, 70-, 80-, 90-100%) across a regional aridity gradient (325-, 375-, 425-, 475-525mm), and sampled networks of native bees, wasps and their parasitoids using 960 trap nests over two years in 48 remnant woodlands. We then used structural equation models to disentangle the direct vs indirect effects of interacting drivers on network structure and function.

Results/Conclusions  

We found strong direct and indirect effects of global change on host-parasitoid networks, dominated by synergistic interactions between drivers. There were direct effects of habitat loss which reduced both the total abundance of parasitoids in networks and their compartmentalised structure (lower modularity). Negative effects of habitat loss on network species richness were compounded by both regional aridity and high rainfall decline, interactions which caused cascading changes to network structure (connectance, modularity, functional complementarity) and to the rate and temporal stability of community wide-function (parasitism). Furthermore, high rainfall decline strongly decreased host-bee dominance of network interactions, an effect exacerbated in the historically-driest areas, and which again caused downstream impacts (direct and indirect) to network structure and parasitism rates. Finally, importantly, the presence of complex three-way interaction effects between rainfall decline and habitat loss across regions of increasing aridity point to the challenges in predicting and managing ecological networks under intensifying anthropogenic impacts. Our findings have uncovered a multitude of different pathways by which global change can alter species networks, while also providing valuable direct evidence of the impacts of a drying climate. This work has important global relevance for the conservation of fragmented habitats under increasing drought pressure with climate change.