Use this URL to cite or link to this record in EThOS:
Title: Habitat suitablity modelling in the New Forest National Park
Author: Douglas, Sarah Jane
ISNI:       0000 0004 2686 2389
Awarding Body: Bournemouth University
Current Institution: Bournemouth University
Date of Award: 2009
Availability of Full Text:
Access from EThOS:
Access from Institution:
The New Forest National Park is a unique semi-natural landscape which contains many species and habitats which are rare and/or threatened. In order to effectively aid in the conservation of these species, particularly in the face of climate change, there is a requirement to know their habitat requirements and distributions within the New Forest. However, due to limited resources there are gaps in knowledge about this for many of these species. Habitat suitability modelling was carried out to suggest unsurveyed sites of potentially suitable habitat (and consequently higher likelihood of species occurrence) for selected species of high conservation concern (Chamaemelum nobile, Galium constrictum, Gladiolus illyricus, Hipparchia semele, Nemobius sylvestris, Pilularia globulifera, Plebejus argus and Poronia punctata). The performance of several modelling approaches was compared. Of the models based on the use of GIS spatial data, an approach requiring only species presence data (Ecological Niche Factor Analysis (ENFA)) was compared to approaches additionally requiring absence or pseudo-absence data (Generalised Linear Models (GLMs) and Generalised Additive Models (GAMs)). An additional approach that did not require GIS data, Bayesian Belief Network (BBN) modelling, was also used to incorporate finer-scale variables not available in GIS format. This relatively new approach to habitat suitability modelling was also used to predict the potential impact of climate change on the suitability of the habitats for the selected species. The evaluation results showed that the presence-absence GLM and GAM models out-performed the presence-only ENFA method, and that the use of pseudo-absences and automated stepwise variable selection proved effective for developing these models. Species with specialist habitat requirements tended to be modelled more accurately than more generalist species. The BBN models also achieved high evaluation values, and were particularly valuable in being able to provide a quantitative assessment of the potential impact of climate change on the selected species. Habitat suitability modelling at the scale of an individual predicted area of the size of the New Forest has so far been rare, as have predictions of climate change on specific species at this scale. However, the results of this research show that this can be a valuable approach to aid in management and conservation of species and their habitats in protected areas.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available
Keywords: Geography and Environmental Studies ; Biology and Botany