Empirical evidence of disturbance interactions: examples from Australian ecosystems

<ol> <li>Ecosystems are shaped by a range of drivers including natural and human disturbances. Many natural disturbances such as fire, insect attack and windstorms are increasing in frequency and severity. At the same time, human disturbances like logging, land clearing and plantation establishment are expanding. Co-occurring disturbances can interact with far-reaching consequences, including altered ecological processes and losses of biodiversity. Whilst the importance of interacting disturbances is increasingly recognized, the majority of empirical studies focus only on a single driver of change.</li> <li>We evaluate evidence for interacting drivers of change, including natural and human disturbances, in 57 empirical studies completed within two markedly different ecosystems in southern Australia that had been subject to large-scale, long-term research programs. These programs have documented the ecological effects of fire, native forest logging, post-fire (salvage) logging, browsing by native herbivores, and invasive species control.</li> <li>We found effects of interactions among disturbances on every vertebrate class that we examined (mammals, birds, reptiles and amphibians), as well as on vegetation and ecosystem processes (e.g. tree germination and tree mortality). Interactions between disturbances were often statistically significant when tested for. Similarly, where tested, environmental context often significantly modified disturbance effects. Finally, we found evidence of one disturbance affecting the likelihood or intensity of a subsequent disturbance (<i>viz</i>: an interaction chain or linked disturbances), and of the order in which disturbances occurred altering biotic responses.</li> <li> <a> <b>Synthesis and applications</b></a>. Interactions between disturbances warrant more attention in empirical studies because they can be common and have substantial impacts over large spatial scales, and prolonged timeframes. Approaches to limit adverse impacts of interacting disturbances include eliminating some drivers if possible (particularly where interaction chains are present), reducing the spatial overlap of disturbances (e.g. by protecting key refugia), reducing the potential for temporal overlap of multiple disturbances, and identifying key leverage points where actions are most effective.</li> </ol>