Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.637277
Title: Identifying and mapping tropical vegetation and its phenology using satellite imagery : case studies from Peruvian and Brazilian Amazonia
Author: Hill, R. A.
Awarding Body: University of Wales Swansea
Current Institution: Swansea University
Date of Award: 1997
Availability of Full Text:
Access through EThOS:
Abstract:
This thesis investigates satellite remote sensing as a tool (i) to identify and map, and (ii) to study the phenology of ecologically-defined primary forest types, disturbance areas and regenerative phases in Amazonia. This is achieved using Landsat TM and NOAA AVHRR imagery, in the form of single-date and multi-temporal data sets, for five field locations (one in Peru and four in Brazil). It is shown possible to identify (and therefore map) various groups of tropical vegetation types in single-date 3-band (visible, near and middle infrared) TM imagery. Their contrasting spectral reflectance characteristics are considered to relate to different shadowing patterns resulting from canopy structure (especially density and roughness). Thus, ecologically distinct vegetation types can only be discriminated and mapped by per-pixel digital classification procedures in optical imagery if they vary in parallel to structural differences. The combined use of near and thermal infrared spectral data improves forest type discrimination, as the latter relates not only to canopy structure but also to environmental conditions (such as soil and atmospheric moisture availability/demand). Increased vegetation spectral separability is also demonstrated by the use of two-date as opposed to single-date imagery, and by the combined use of visual and digital image analysis techniques (enabling the incorporation of contextual information). For multi-data AVHRR imagery, the spectral characteristics of Amazonian vegetation types relate to both the structural characteristics and phenological patterns of the vegetation and to the viewing geometry of the sensor. By studying two-date AVHRR imagery of near identical viewing geometry, vegetation phenological response to climatic seasonality across eastern Amazonia is detected. Furthermore, by analysing the differential response of selected vegetation types at three locations in near and thermal infrared wavebands, it is possible to infer the underlying physiological mechanism behind the vegetation phenological activity.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.637277  DOI: Not available
Share: