Use this URL to cite or link to this record in EThOS:
Title: Modelling tropical tree species distributions for improved conservation assessments of dipterocarps in Malaysia and the Philippines
Author: Amaludin, Nazahatul Anis
ISNI:       0000 0004 2739 7722
Awarding Body: University of Aberdeen
Current Institution: University of Aberdeen
Date of Award: 2012
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
Southeast Asian forests are globally important for their high species richness and endemicity of plant species. However, the region has suffered from high rates of deforestation, which potentially reduces the availability of habitat for tree species and other components of biodiversity. As a result many species may be at risk of extinction, but the extent of this problem has not been addressed systematically. In this thesis I conduct conservation assessments for multiple tree species in the Dipterocarpaceae growing in Sabah and the Philippines to determine the extent of extinction risk arising from loss of primary forest cover. I used the ecological niche modelling algorithm MAXENT to predict the extent of the historic species distribution and then overlaid contemporary land use maps to estimate the amount of habitat loss for 15 dipterocarps in Sabah and 27 species in the Philippines. Conservation assessments were made by reference to the thresholds of habitat loss required to assign the various categories of threat used in the IUCN Red List. The specific objectives of this thesis were (a) to reconstruct the temporal pattern of loss of primary forest cover in Sabah over the last 25 years using published vegetation maps derived from aerial photography and satellite images; (b) to determine whether habitat loss and conservation assessments estimated using ecological niche models are sensitive to the resolution of data on the distribution of soils within the sampling window; (c) to determine whether estimates of habitat loss and conservation assessments are sensitive to decisions over land use classification from satellite images; and (d) to compare conservation assessments based on ecological niche models to assessments derived from reduction in the extent of occurrence (EOO) and the area of occupancy (AOO), which are the traditional techniques recommended by the IUCN. The ultimate aim of the thesis was to refine the methodology for conducting rapid species conservation assessments for endangered tropical tree species. The rate of loss of primary forest cover in Sabah between the mid 1980s and 2010 was estimated at 0.6 % yr-1. This value obscures substantial variation between consecutive intervals over this period, which ranged from a minimum of 0.2 % yr-1 to a maximum of 8.4 % yr-1. The estimated impact of this loss of primary forest on the amount of habitat available for species of Dipterocarpaceae was dependent on the method used for predicting species distributions. Ecological niche models (ENMs) that incorporated a high resolution of soil information predicted a lower mean area within the historic distribution and a lower amount of remaining habitat area than ENMs developed using either low resolution or no soil data. Percentage habitat loss (mean 71- 75 %) did not differ in response to the spatial resolution of soil data in ENMs. Similarly, estimated amounts and percentages of remaining habitat were strongly dependent on whether secondary vegetation was considered to be suitable habitat for dipterocarps. In general, conservation assessments derived from ENMs were more conservative (i.e. predicted a higher degree of threat) than those developed using the traditional methods based on computing the extent of occurrence and area of occupancy. Ecological niche modelling represents a powerful new tool for implementing rapid conservation assessments of tropical forest trees, but this thesis has identified caveats that must be considered before they can be routinely applied. In particular, many species are under-collected or the distribution of samples is biased, and significant up-scaling of collection effort is required before reliable ENMs can be obtained.
Supervisor: Not available Sponsor: Ministry of Higher Education, Malaysian Government
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available
Keywords: Dipterocarpaceae