Managing the impacts of habitat loss and invasive mammals on New Zealand forest birds

The greatest challenge for conservation this century will be to preserve biodiversity in the face of extensive, ongoing habitat loss. However, meeting this challenge is complicated by the fact that (1) habitat loss frequently interacts with other threatening processes, making its impacts variable and difficult to predict, and (2) resources are typically insufficient to effectively manage all threats. In this thesis, I use relationships between forest cover, invasive mammalian predators (rats and possums), and native forest birds in New Zealand to examine how variability in the impacts of habitat loss, and the competing need to manage alternative threats, can be incorporated into strategies for preserving biodiversity in the face of ongoing habitat loss. Combining field surveys with existing large‐scale datasets, I ask the following questions: (1) how do the impacts of native forest loss on bird richness vary with the surrounding 'matrix' land use? (2) Does forest loss indirectly affect native birds by altering the abundance of invasive mammals? (3) How effectively can different pest control methods reduce invasive‐mammal abundance? And (4) when can the impacts of forest loss on birds be offset by conducting pest control in remaining forest? Analysing a nationwide dataset of bird distribution, I found that the impacts of native forest loss on bird richness were far lower when the matrix was dominated by exotic pine plantations than by high‐contrast agricultural land uses, suggesting that matrix composition can play an important role in managing the impacts of habitat loss. In contrast, I found little evidence that forest loss indirectly affects birds by altering the abundance of invasive mammals. Although a single‐landscape analysis found that rat capture rates were lower at the forest edge than the forest interior, a regional‐scale analysis found no evidence that native forest cover influenced rat or possum abundance. Instead, abundance varied strongly with local vegetation characteristics, climate and topography, and urbanisation, suggesting that these variables may help managers with limited budgets to prioritise locations for pest control. This analysis also found that high-intensity pest control operations could substantially reduce rat and possum abundance, with abundance indices that were >90% lower in areas of high‐intensity control than in uncontrolled areas. In contrast, there was no measurable difference in abundance between uncontrolled areas and areas under low‐intensity pest control. Low‐intensity operations cost millions of dollars and kill tens of millions of sentient pests each year, and my results raise questions about how often these costs are justified. Finally, I combined data on forest cover, pest control, and bird richness and abundance across the Auckland region to predict when pest control could be used to offset the impacts of forest clearance on bird communities. I found that pest control could offset impacts in some cases, but this depended on the combined effects of several factors: current levels of forest cover in the landscape, proposed amount of forest clearance, and type of pest control employed. Alternative‐threat management is a promising tool for offsetting the impacts of habitat loss, but a framework that can account for these factors, such as the one developed in my analysis, will be necessary to predict when it is appropriate. Understanding relationships between habitat loss, interacting factors, and the management of alternative threats will help to meet the challenge of preserving biodiversity in the face of ongoing habitat loss.