Skip to main content
Kent Academic Repository

Informing tropical mammal conservation in human-modified landscapes using remote technologies and hierarchical modelling

Deere, Nicolas J. (2018) Informing tropical mammal conservation in human-modified landscapes using remote technologies and hierarchical modelling. Doctor of Philosophy (PhD) thesis, University of Kent,. (KAR id:73048)

Language: English
Download this file
[thumbnail of 217Deere_NJ_2018_Tropical_Mammal_Conservation_In_Human-Modified_Landscapes.pdf]


The aggressive expansion of anthropogenic activities is placing increasing pressure on biodiversity, particularly in tropical regions. Here, conservation efforts are hindered by poor understanding of species ecology and the failure of policy instruments to account for multiple stressors of land-use change. While protected areas are central to conservation strategies, there is a general consensus that the future of tropical biodiversity will be determined by how well modified landscapes are managed. In this thesis I advance our understanding of biodiversity persistence in modified tropical landscapes to inform emerging incentive-based policy mechanisms and supply-chain initiatives. Capitalising on recent advances in remote-sensing and hierarchical occupancy modelling, I provide a spatial appraisal of biodiversity in a modified landscape in Sabah, Malaysian Borneo. Fieldwork was conducted at the Stability of Altered Forest Ecosystems (SAFE) project, a large-scale landscape modification experiment, comprising a degradation gradient of old growth forest, selectively logged forest, remnant forest patches and oil palm plantations. The assessment focused on camera-trapping of tropical mammals, as they are sensitive to anthropogenic stressors, occupy key trophic positions, and prioritised in conservation. In Chapter 2 I link mammal occupancy data to airborne multispectral remote-sensing information to show how the conservation value of modified landscapes is dictated by the intensity of the underlying land-use. Logged forests retained appreciable levels of mammal diversity, and oil palm areas were largely devoid of forest specialists and threatened taxa. Moreover, many mammal species disproportionately occupied forested areas that retained old growth structural characteristics. The most influential structural measures accounted for vertical and horizontal components in environmental space, which cannot currently be derived from conventional satellite data. Using a novel application of ecological threshold analysis, I demonstrate how multispectral data and multi-scale occupancy models can help identify conservation and restoration areas in degraded forests. In Chapter 3 I assess the potential for carbon-orientated policy mechanisms (High Carbon Stock, HCS, Approach and REDD+) to prioritise high carbon areas with corresponding biodiversity value in highly modified landscapes. The areas of highest carbon value prioritised via HCS supported comparable species diversity to old growth forest. However, the strength, nature and extent of the biodiversity co-benefit was dependent on how carbon was characterised, the spatial resolution of carbon data, and the species considered. In Chapter 4 I further scrutinised HCS protocols to evaluate how well they delineated high priority forest patches that safeguard species most vulnerable to land-use change (i.e. IUCN threatened species). The minimum core area required to define a high priority patch (100 ha) supported only 35% of the mammal community. In fact the core area criterion would need to increase to 3,199 ha in order to sustain intact mammal assemblages, and an order of magnitude higher if hunting pressure was considered. These findings underline the importance of integrating secondary disturbance impacts into spatial conservation planning. Provided landscape interventions are directed to where they will have the greatest impact, they can be financially sustaining and garner local support for conservation. To this end I provide recommendations to guide policy implementation in modified tropical landscapes to support holistic conservation strategies.

Item Type: Thesis (Doctor of Philosophy (PhD))
Thesis advisor: Struebig, Matthew
Thesis advisor: Davies, Zoe
Uncontrolled keywords: Camera-trapping, hierarchical modelling, human-modified landscapes, land-use change, mammals, oil palm, selective logging, Southeast Asia.
Divisions: Divisions > Division of Human and Social Sciences > School of Anthropology and Conservation
Funders: Organisations -1 not found.
SWORD Depositor: System Moodle
Depositing User: System Moodle
Date Deposited: 18 Mar 2019 10:10 UTC
Last Modified: 08 Dec 2022 23:12 UTC
Resource URI: (The current URI for this page, for reference purposes)

University of Kent Author Information

  • Depositors only (login required):

Total unique views for this document in KAR since July 2020. For more details click on the image.