Seminar 8th February 2011 noon Building 46, room 2005
Modelling mammalian movements in fragmented landscapes: exploring and integrating least-cost models and agent-based simulations
Angela Watkins
School of Biological Sciences and Institute of Complex Systems Simulation
- Categories
- Agents, ArcGIS, Complex Systems, Ecology, Geographic Information Systems
- Submitter
- Angela Watkins
Habitat loss and fragmentation are serious threats to biodiversity and understanding a landscape in terms of its functional connectivity is a key requirement of current conservation work. The level of connectivity in a landscape has strong effects on the ability of a species to disperse and this is a key mechanism in determining the potential of a species to persist in the landscape. Least-cost modelling provides some capacity to evaluate the impact of landscape composition on species’ dispersal dynamics and involves constructing a model of functional landscape connectivity. Validating proposed least-cost models is an important step but has rarely been achieved due to difficulties in collecting dispersal data. Agent-based simulation modelling provides an alternative approach to modelling ecological systems by simulating the behaviour and interaction of individual entities. This study re-implements previous work introducing a new approach to analysing least cost models, and attempts to validate proposed models with empirical data on hedgehog dispersal trajectories. Agent-based simulations of hedgehog behaviour extend this work and investigate the impact of changing fragmentation on the fitness of dispersing individuals. More ecologically realistic individuals show an increase in fitness in more fragmented landscapes, in contrast to the decrease in fitness exhibited by individuals that move purely in least-cost pathways. This study highlights the importance of understanding individual-level movement decisions and supports agent-based modelling as a valid choice for exploring ecological systems.