Assessment of Paleo-Landscape Features using Advanced Remote Sensing Techniques, Modelling and GIS Methods in the Lake Manyara Basin, Northern Tanzania

DSpace Repositorium (Manakin basiert)


Dateien:

Zitierfähiger Link (URI): http://hdl.handle.net/10900/68445
http://nbn-resolving.de/urn:nbn:de:bsz:21-dspace-684454
http://dx.doi.org/10.15496/publikation-9864
Dokumentart: Dissertation
Erscheinungsdatum: 2016
Sprache: Englisch
Fakultät: 7 Mathematisch-Naturwissenschaftliche Fakultät
7 Mathematisch-Naturwissenschaftliche Fakultät
Fachbereich: Geographie, Geoökologie, Geowissenschaft
Gutachter: Hochschild, Volker (Prof. Dr.)
Tag der mündl. Prüfung: 2016-02-08
DDC-Klassifikation: 550 - Geowissenschaften
Schlagworte: Geoinformationssystem , Fernerkundung , Tansania , Landschaftsanalyse
Freie Schlagwörter: Manyara
Landschaftsrekonstruktion
multispektral
SAR
Paläolandschaft
Paleo-Landscape
multispectral
landscape reconstruction
Lizenz: http://tobias-lib.uni-tuebingen.de/doku/lic_ohne_pod.php?la=de http://tobias-lib.uni-tuebingen.de/doku/lic_ohne_pod.php?la=en
Zur Langanzeige

Abstract:

In researching the evolution of hominids, the East African Rift System acts as a vital region. The rift valleys enabled some of the most sensational hominid findings to date. Various hypotheses have been developed in the last decades, which try to explain the influence of changes in paleo-climate, paleo-landscape and paleo-environment on hominin evolution in the Quaternary. Additionally, the sediments and the morphology of the East African Rift System provide excellent terrestrial archives for paleo-environmental reconstruction. Lake Manyara is located in an endorheic basin in the eastern arm of the East African Rift System in northern Tanzania. The surroundings of the Lake Manyara are in the focus of paleontological and archaeological investigations. For instance, two hominin bearing sites were found within the catchment of the Makuyuni River, as well as artefacts and fossils are periodically uncovered. The study area, which is located east of the present-day lake, provides an insight into relevant geological and geomorphological drivers of paleo-landscape evolution of the whole region. This thesis aims at contributing to the understanding of landscape evolution in the Lake Manyara region. Compared to other regions in the East African rift system, few landscape evolution studies took place for the Lake Manyara basin. As such, an integrative scientific investigation of the spatial situation of paleo-landscape features and of paleo-lake level fluctuations is missing. The proposed study utilizes state-of-the-art remote sensing based research methods in evaluating the landscape, and in concluding from present-day landforms and processes, how the landscape developed during the Pleistocene and Holocene. In striving to accomplish this goal, this cumulative dissertation comprises eight central research questions, which are introduced in a conceptual framework. The research questions have been considered in seven scientific publications, which describe the applied methodologies and results in detail. The framework of the thesis provides a coherent and detailed interpretation and discussion of the scientific findings. The research questions and outcomes of the analyses are listed below. Key drivers of landscape development in the East African Rift System are tectonic and tectonically induced processes. Drainage network, stream longitudinal profiles and basin analysis based on topographic analyses, as well as lineaments extracted from remote sensing images, were successfully used as methods in identifying tectonic activity and related features in rift areas. The application of a gully erosion model suggests that the gully channel systems in the study area are relatively stable and that they had developed prior to the last significant lake regression. The paleo-landscape and the paleo-environment are closely connected to lake level changes of the paleo-Lake Manyara. Hence, a key question concerns the extent of the Manyara Beds, which are lacustrine deposits that indicate the maximum extent of the paleo-Lake Manyara. A combined analysis, utilizing ASTER multispectral indices and topographic parameters from a digital elevation model, led to the spatial delineation of lacustrine sediments. Their extent indicates a relation to lacustrine sediments in the southern part of the basin, and reveals lacustrine / palustrine deposits further east. A methodological comparison of Support Vector Machines and Boosted Regression Trees, which served as classification methods to identify the lacustrine sediments, exhibited high accuracies for both approaches, with minor advantages for Support Vector Machines. Closely related to the previous research question is the question on the spatial distribution of surface substrates. By incorporating a WorldView-2 scene and Synthetic Aperture Radar data to the previously mentioned datasets, it was possible to distinguish between nine topsoil and lithological target classes in the study area. The surface substrates indicate the underlying lithologies, sediments and soils, as well as soil formation processes. Between the village of Makuyuni and the present-day Lake Manyara, paleo-shorelines and terraces were formed by various paleo-lake levels. Questions arise, at which elevation these features occur and what is the maximum elevation, which was reached. ALOS PALSAR and TerraSAR-X backscatter intensity information provided the possibility of an area-wide mapping of those morphological features. Some radiometric dates exist for stromatolites from a distinct paleo-shoreline level, which support the interpretation of the lake fluctuations. The paleo-shoreline, which was identified with the highest elevation, coincides with the elevation of the lowest possible outlet of the closed Manyara basin. It can be assumed that the paleo-Lake Manyara over-spilled into the neighboring Engaruka and Natron-Magadi basins. The question of the location of sites with a high probability of artefact and/or fossil presence is important for future archaeological and paleontological research. ASTER remote sensing data and topographic indices contributed likewise to the predictive modelling of probabilities of archaeological and paleontological sites in the study area. Generally, paleontological sites are found on a higher elevation, compared to Stone Age sites. In addition, fossil sites seem to be related to stable paleo-landscape features according to this study’s findings. The results of this dissertation provide new insights in the landscape development of the Lake Manyara basin. The scientific findings contribute to the understanding of the landscape evolution for the study area, as well as for the neighboring basins in the East African Rift System. The applied geospatial methodologies can be transferred to other study areas with similar research needs.

Das Dokument erscheint in: