File(s) under embargo
until file(s) become available
Protected area planning for land-use change : optimising outcomes for biodiversity
Habitat destruction and degradation due to human land uses are core drivers of biodiversity loss, and area-based conservation efforts (e.g., protected areas) are a primary means of addressing these threats. The upcoming conference of the parties to the Convention on Biological Diversity is likely to endorse an increase in the protected area (PA) coverage target to 30% by 2030, with the goal of better mitigating ongoing biodiversity loss. However, the effectiveness of an expanded global PA network is threatened by problems like PA-impermanence, non-target ‘leakage’ effects, poor spatial prioritisation, and historically entrenched biases like unequal representation of habitat types. While each of these issues has been discussed in the literature, there are no clear guidelines for addressing leakage nor for coordinating local actions to build globally effective, expanded PA networks.
As framed in Chapter 1, the aim of my thesis was to provide real-world insights into the relative importance of the two emergent problems – leakage and globally uncoordinated prioritisation – in the context of global conservation, and to propose evidence-based solutions. The quantitative-triage methodology I apply seeks to ensure that area-based conservation efforts achieve maximal positive impact for irreplaceable and increasingly vulnerable biodiversity.
In Chapter 2, I assess the prevalence and magnitude of the spillover effects of PAs in a global systematic review of the literature. The first step in this review was to identify the most robust methods for measuring spillovers from PA establishment. Subsequent extraction and collation of all relevant data from the peer-reviewed literature provided generalised inferences on the existence, prevalence, magnitude, and direction of recorded spillover effects. I found that, in general, PAs have reduced deforestation rates effectively within their boundaries, without displacing deforestation pressures to nearby unprotected areas (i.e., without causing leakage, the problematic type of spillover). However, given the limitations of the available data, further confirmation of these results was required.
In Chapter 3, I use counterfactual methods identified in the systematic review to measure the target and non-target effects of recently established protected areas on deforestation rates within a global hotspot of habitat conversion: the Brazilian Amazon. A key methodological step for measuring the impacts of PA establishment is the estimation of a counterfactual a baseline rate of deforestation expected in a given landscape, from which deviations from the baseline in the PA and in its spillover zone were determined. The results from this regional study from the neotropics were aligned with those of the global systematic review, and I concluded that leakage effects from PA establishment were not of sufficient magnitude to warrant direct mitigation efforts, compared to known problems of impermanence, location bias, and non-representativeness.
In Chapter 4, I address the problems that arise when local priorities drive PA expansion decisions at the expense of globally coordinated conservation efforts. I demonstrate a data-driven solution using a case study of Tasmanian forests using a transparent decision-support tool. I evaluated the potential allocation of land available for protection based on four objectives – representation of biogeographic units, connectivity, threat mitigation, and cost efficiency – each weighted equally. As a contrast to this ‘objective-neutral’ scenario, I also devised a jurisdictionally informed plan for tenure allocation, which prioritises features according to the origin of the target to which they contribute, where global conservation goals are valued over continental objectives over local concerns. Representational gap analyses and this jurisdictionally informed planning strategy identified the northeast of Tasmania as a target region for global and continental conservation goals.
In Chapter 5, I investigate the extent to which broad-scale design criteria for protected-areas (e.g., coverage of bioregions, network connectivity, threat mitigation, cost efficiency) reflect ecological priorities identified by fine-scale data: in this case, community-composition and focal species’ relative abundance data from a year of continual camera-trap monitoring for mammals and ground-dwelling birds collected from 201 sites. The focal area for this investigation was the northeast Tasmanian target region identified in Chapter 4. Machine-learning-assisted image classification to the species level enabled efficient calculation of distribution, diversity and relative-abundance metrics. Protectable sites were then ranked by these metrics and compared with other rankings generated using coarser proxies for biodiversity, validating the latter and grounding the global prioritisation of this region in the local ecology.
Collectively, the systematic review and the case studies within the complex socioecological contexts of Brazil and Tasmania yielded important contributions to the ‘crisis discipline’ of conservation science. This body of research revealed that known, entrenched problems are more important threats to effective area-based conservation than the widely perceived problem of PA leakage. Further, modern monitoring technology, computing power, machine- and statistical-learning methods and spatial decision support tools can be leveraged to achieve globally coordinated spatial prioritisation without losing sight of the ecological communities using the habitats being protected.
- PhD Thesis
Department/SchoolSchool of Natural Sciences
PublisherUniversity of Tasmania