Assessing Intra-seasonal land surface change and long term trends in the Succulent Karoo biome using coarse resolution  satellite data and interpolated rainfall surfaces

Roberts, Jonathan Wesley

Assessing Intra-seasonal land surface change and long term trends in the Succulent Karoo biome using coarse resolution satellite data and interpolated rainfall surfaces

Thesis / Dissertation

2005

Publisher

University of Cape Town

Department

Department of Environmental and Geographical Science

Faculty

Faculty of Science

Abstract

The Succulent Karoo is a biodiversity hot spot situated along the west coast of southern Africa. While it is predominantly recognized as a west coast vegetation type its borders stretch as far east as Steytlerville in the Little Karoo. The area contains the largest number of endemic succulent species in the world and harbors nearly 10 percent of the total number of succulent species worldwide. Furthermore, spring mass-flowering events draw thousands of tourists to the region, providing welcome input to the local economies. The floral diversity is however under threat from various environmental forces. These forces include inappropriate land use practices resulting in Land Degradation and the ever-present threat of Climate Change. With the pressures that are presently being placed on the biome and the threat of future changes to the climatic regime, it is necessary to develop methodologies to monitor vegetation changes within the biome. These methodologies should be able to discern between climate driven and land use driven changes that may denude the rich biological diversity of the biome. Monitoring regional changes in vegetation necessitates the use of coarse resolution satellite data in conjunction with statistically derived rainfall surfaces. For the purposes of this study intraseasonal land surface change is monitored using the Normalized Difference Vegetation Index derived from the National Oceanic and Atmospheric Administration (NOAA), Advanced Very High Resolution Radiometer (AVHRR). Rainfall surfaces are taken from Hewitson and Crane (2005). Ancillary temperature data taken from NCEP reanalysis is also used. Three analysis methodologies are used to answer questions relating to the relationship between rainfall and vegetation growth. Firstly, a Spatio Temporal Correlation filter (STCf) is used to investigate the spatial and temporal nature of seasonal vegetation changes (Where is seasonal vegetation change homogenous?). Secondly Principal Component Analysis is used to monitor seasonal vegetation change (What is the seasonal nature of NOVI in the Succulent Karoo?) and finally the relationship between precipitation and vegetation growth is classified using a Self Organising Map ( When and where is the relationship between rainfall and vegetation growth strongest?). Results from the three analysis methodologies are combined to determine zones of potential critical change. Within each of these zones drivers of vegetation change are identified and discussed. Most notably, regression analysis conducted on the output of the Self Organising Map indicates that temperature plays a more important role in vegetation growth than precipitation. This result indicates that the biome may be more vulnerable to the impacts of Global Warming than previously anticipated. Results also indicate that future studies of vegetation change using satellite imagery should be cognisant of the potential effects of soil background reflectance. In conclusion this thesis shows that much of the Succulent Karoo biome is under threat from both inappropriate land use and Climate Change, which may potentially lead to degradation of the biodiversity and rangelands that support the local population.