Direct and indirect disturbance impacts on forest biodiversity

Abstract Understanding the responses of biodiversity to forest disturbance is critical for maintaining ecosystem integrity and key ecological functions. Ecological research in forests after major disturbance typically focuses on the direct responses of individual species or communities, often with only limited consideration of the interactions between forest structure and biodiversity. Despite this, these interactions can mediate indirect disturbance impacts throughout forest ecosystems. Using data from a decade of extensive, empirical research, we employ structural equation modeling to quantify the direct and indirect impacts of multiple stand‐replacing disturbances on elements of forest structure (basal area (BA) of dominant plant lifeforms and number of hollow‐bearing trees) and on biodiversity (plant diversity, arboreal marsupials, and avifauna) in the southeastern Australian montane‐ash forests. The number of stand‐replacing disturbances resulted in a lower species richness and lower functional dispersion of avifauna and had a strong negative influence on the number of hollow‐bearing trees. In contrast, the number of stand‐replacing disturbances increased plant species richness in the understory, after controlling for the influence of stand age. In this ecosystem, we also found evidence of ecological interactions that mediated indirect impacts. For instance, disturbance‐induced declines in the number of hollow‐bearing trees had negative impacts on the richness of arboreal marsupials. Moreover, disturbance‐induced increases in the richness of understory plants had positive impacts on avian richness, partially mitigating direct disturbance impacts. Our novel integrative study provides new insights into the mechanisms that underpin disturbance impacts in forests at the ecosystem level. This has enabled the development of a new conceptual model on forest responses to the number of stand‐replacing disturbances. By exploring interactions between elements of biodiversity, forest structure, and disturbance, our model demonstrates how disturbance‐induced changes in one measure can have cascading effects on another. This understanding will provide for an integrative approach to management and planning.