Use this URL to cite or link to this record in EThOS:
Title: Habitat mapping and biodiversity monitoring using earth observation and GIS
Author: Thackrah, G.
Awarding Body: University of Wales Swansea
Current Institution: Swansea University
Date of Award: 2001
Availability of Full Text:
Access from EThOS:
Areas of high conservation value in the UK often have a legal requirement for some form of regular monitoring of their status. This thesis investigates the use of methods in Earth Observation to achieve this both reliably and faster than land based approaches. This is done through the use of a case study in south Wales, Kenfig National Nature Reserve. The site is a candidate Special Area for Conservation under the European Union's Habitats Directive. This is due to the presence of the fen orchid, Liparis loeselii, and petalwort, Petalophyllum ralfsii. Use is made of data from two principal airborne multispectral sensors, the Compact Airborne Spectrographic Imager (CASI) and the Airborne Thematic Mapper (ATM) as well as a Light Detection and Ranging (LIDAR) instrument. Simple methods of EO data processing are used to classify these data and produce accurate maps of habitat in an area of high conservation value. Habitats with similar spectral properties are hard to classify in this way. Methods of data fusion and information aggregation were thus applied to assist in the identification of habitats inaccurately identified with the simpler methods of classification. These are shown to produce more robust results. Furthermore, ancillary data, in the form of elevation information, may also be used to refine the classification process again. Using a method of information aggregation that allows constituent sources to be differentially weighted in light of a priori knowledge regarding classifier accuracy, a robust classification is produced. It is concluded that the data fusion approaches allow more robust classification products to be supplied to the end users of the data, the managers of sites of conservation interest such as Kenfig NNR. The availability of such approaches in commonly available computing environments is currently limited; it is therefore suggested that they be made more widely available by suppliers of GIS technology to conservation organisations.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available