Diplomarbeit

Spatio-temporal Biodiversity Patterns in Semiarid Savanna Ecosystems of the Kaokoveld, Northwest Namibia

Sibylle Hassler (12/2022-12/2006)

Betreuer: Anke Jentsch, Carl Beierkuhnlein

Abstract:

Fourteen years after the biodiversity convention in Rio de Janeiro scientific authors have reached consensus that preventing biodiversity loss is critical for continued ecosystem functioning. Methods for monitoring biodiversity are needed – remote sensing is providing some techniques, but there is much room for improvement in field sampling and data analysis. This study presents the application of two current data analysis methods for evaluating beta diversity of vegetation, namely the HETerogeneity concept and the unsupervised artificial neural network approach of self-organising maps. The methods are based on a dataset taken on a hexagonal sampling design suitable for grid-based biodiversity assessment on various scales. They include estimates of biodiversity change in space and time.

The study area is situated in the Kaokoveld in northwest Namibia, in semiarid savanna ecosystems typical of many savanna regions in Africa. These are subject to the conflicting interests of conservationists, wanting to protect the Kaokoveld biodiversity hotspot, and of local people, needing to utilise the resources of these savannas to ensure their livelihood. This conflict accentuates the need for appropriate management practices to make sustainable use and phytodiversity conservation possible, based on sound scientific knowledge of savanna patterns and their variability.

Although several methodological inaccuracies still need to be eliminated and tests of the HETerogeneity concept and self-organising maps in other ecosystems will help to establish their place among traditional analysis methods they were successfully applied for vegetation pattern analysis in the present study. Main findings were the importance of patternforming elements like tree islands and washes. These are crucial for ecosystem functioning in these areas and were seen to react sensitively to changes, for instance in moisture conditions, between dry and wet season. The marked intraannual variability in vegetation responding to shifting water and nutrient availability indicates the need for spatio-temporal biodiversity monitoring.