Change and variation in a hyper-arid cultural landscape: A methodological approach using remote sensing timeseries (Landsat MRR and TM, 1973-1996) from the wadi vegetation of the eastern desert of Egypt
Tittel | Change and variation in a hyper-arid cultural landscape: A methodological approach using remote sensing timeseries (Landsat MRR and TM, 1973-1996) from the wadi vegetation of the eastern desert of Egypt |
Publikasjonstype | Thesis |
Utgivelseår | 1999 |
Forfattere | Andersen, GL |
Refereed Designation | Unknown |
Undertittel | Master Thesis No. 59 |
Utgiver | The University of Bergen |
Thesis Type | Master Thesis |
Thesis Number | 59 |
Sammendrag | Nine wadi localities in a hyper-arid environment have been registered in the field and studied using earth observation data. Branch height, crown – and trunk – diameter, and indicators of land-use such as present traces of browsing, lopping and charcoal production were registered for arboreal vegetation, mostly Acacia tortilis and Balanites aegyptiaca. A point mapping (GPS) was selected to optimise subsequent integration with raster data and to facilitate a detailed interpretation of change images. Field data and change images are interpreted according to two gradients, one cultural and one hydrological. Derived tree maps are overlaid referenced TM data in order to detect differences between pixels with and without vegetation. The Red band is the most consistent spectral band in its content of vegetation information. Nevertheless it is apparent that several methodological and technical factors constrain the possibilities to register vegetation in this environment of very scarce vegetation cover. Similar problems are also recognised in the change analysis which is based on the difference between Red bands of the years compared. Four different datasets are part of the analysis: 1973, 1979, 1984 (all Landsat MSS images) and 1996 (TM). Field data indicate that changes are taking place in the cultural landscape of the Eastern Desert, and the change is primarily due to processes that both in causes and consequences is associated with ‘deforestation’. Although several sources of errors introduce variations in the change images, the images do reflect the field observations. |
Forfatterens adresse | NERSC |
Attachment | Size |
---|---|
Masteroppgave Gidske L. Andersen.pdf | 15.31 MB |