Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.655377
Title: Spatially-explicit modelling of habitat permeability for mammalian wildlife
Author: Watkins, Angela
ISNI:       0000 0004 5364 1143
Awarding Body: University of Southampton
Current Institution: University of Southampton
Date of Award: 2015
Availability of Full Text:
Access through EThOS:
Full text unavailable from EThOS. Please try the link below.
Access through Institution:
Abstract:
Least-cost and agent-based models present alternative approaches to modelling animal movements, at the population level and individual level respectively. This study introduces and tests a novel integration of adapted least-cost methods into agent-based simulations, in order to analyse connectivity and habitat preferences in two species of mammal. An initial proof-of-concept study built a set of empirically validated least-cost models of European hedgehogs (Erinaceus europaeus) into a simple agent based model. Agents most closely simulated natural behaviours of dispersing hedgehogs when their movements accounted for temporally-dependent habitat preferences in addition to least-cost pathways informed by the connectivity map. The fitness of these agents increased in more highly fragmented landscapes, in contrast to agents that took least-cost pathways without time-specific preferences. The integration of functional connectivity with individual behaviour combined the advantages of both modelling techniques. Quantitative analysis of the individual-level consequences of moving within different landscape scenarios provides a unique way of applying model outcomes to direct conservation action. A second conceptual study applied integrative methods to the construction of an agent-based simulation scaled for jaguars (Panthera onca) occupying fragmented landscape in Belize. This simulation tested alternative configurations of a wildlife corridor currently under development in Central Belize as part of the intercontinental Mesoamerican Biological Corridor. Six alternative corridor configurations and three control conditions differed substantially in their effectiveness at mixing agents across the environment, despite relatively little difference in individual welfare. Best estimates of jaguar movement behaviours suggested that a set of five narrow corridors may out perform one wide corridor of the same overall area. The first two studies set the framework for developing a detailed simulation of jaguar behaviour and population dynamics in a mixed forest and farmland landscape in the south of Belize. This more complex model drew on empirical data on resident jaguars in the region to simulate typical movement, feeding, reproduction and mortality events within a stable natural population. An overview of the construction and application of the model precedes detailed descriptions of its calibration, sensitivity analysis and validation with empirical data. Agents located inside protected forest reserves exhibited higher fitness, expressed in higher fecundity and lower energy- and habitat-related mortality, than agents located outside these reserves. Model validation showed similar patterns to field data in landscape utilisation and the spatial distribution of individuals. This approach to spatial modelling of population dynamics can provide novel insights into effective conservation strategies for large carnivores. Application of the model to the fragmented central corridor region of Belize sets the context for real-world conservation planning. Under current conditions, simulated jaguars formed a small but stable population with various levels of immigration. Implementation of wildlife corridors showed the largest tracts of physically connected reserves increased connectivity between spatially disconnected habitat patches but also increased vulnerability to environmental degradation.
Supervisor: Doncaster, Charles ; Noble, Jason Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.655377  DOI: Not available
Keywords: QH301 Biology
Share: