Daniel Moeketsi

The South African Global Navigation Satellite System (GNSS) network of dual frequency receivers provide an opportunity to investigate solar cycle effects on ionospheric Total Electron Content (TEC) over the South Africa region by taking advantage of the dispersive nature of the ionospheric medium. For this task, the global University of New Brunswick Ionospheric Modelling Technique (UNB-IMT) was adopted, modified and applied to compute TEC using data from the southern African GNSS Network. TEC values were compared with CODE International GNSS services TEC predictions and Ionosonde-derived TEC (ITEC) measurements to test and validate the UNB-IMT results over South Africa. It was found that the variation trends of GTEC and ITEC over all stations are in good agreement and show pronounced seasonal variations with high TEC values around equinoxes for a year near solar maximum and less pronounced around solar minimum. Signature TEC depletions and enhanced spikes were prevalently evident a...

The development of regional ionospheric Total Electron Content (TEC) models has contributed to understanding the behaviour of the ionosphere and the coupling of the ionosphere to space weather activities on both local and global scales. In the past several decades, the International Global Navigation Satellite Systems (IGS) Service networks of dual frequency receiver data have been applied to construct global and regional models of ionospheric TEC. These models were mainly developed in the Northern Hemisphere where there is a dense network of ground based GNSS receivers for regional data coverage. Such efforts have been historically rare over the African region, and have only recently been initiated. This study reports on the investigation of the effect of mid-latitude geomagnetic storms on TEC over South Africa for selected epochs of Solar Cycles 23 and 24. The MAGIC package was used to estimate TEC over South Africa during Post Solar Maximum, Solar Minimum, and Post Solar Minimum epochs. It is found that TEC is largely determined by the diurnal cycle of solar forcing and subsequent relaxation, but effects due to storms can be determined.

ABSTRACT Thesis (M.Sc.)--North-West University, Potchefstroom Campus, 2004. A three-dimensional (3D) steady-state electron modulation model based on Parker (1965) transport equation is applied to study the modelling of – 7 MeV galactic and Jovian electrons in the inner heliosphere. The latter is produced within Jupiter's magnetosphere which is situated at - 5 AU in the ecliptic plane. The heliospheric propagation of these particles is mainly described by the heliospheric diffusion tensor. Some elements of the tensor, such as the diffusion coefficient in the azimuthal direction, which were neglected in the previous two-dimensional modulation studies are investigated to account for the three-dimensional transport of Jovian electrons. Different anisotropic solar wind speed profiles that could represent solar minimum conditions were modelled and their effects were illustrated by computing the distribution of 7 MeV Jovian electrons in the equatorial regions. In particular, the electron intensity time-profile along the Ulysses spacecraft trajectory was calculated for these speed profiles and compared to the 3-10 MeV electron flux observed by the Kiel Electron Telescope (KET) on board the Ulysses spacecraft from launch (1990) up to end of its first out-of-ecliptic orbit (2000). It was found that the model solution computed with the solar wind profile previously assumed for typical solar minimum conditions produced good compatibility with observations up to 1998. After 1998 all model solutions deviated completely from the observations. In this study, as a further attempt to model KET observations more realistically, a new relation is established between the latitudinal dependence of the solar wind speed and the perpendicular polar diffusion. Based on this relation, a transition of an average solar wind speed from solar minimum conditions to intermediate solar activity and to solar maximum conditions was modelled based on the assumption of the time-evolution of large polar coronal holes and were correlated to different scenarios of the enhancement of perpendicular polar diffusion. Effects of these scenarios were illustrated, as a series of steady-state solutions, on the computed 7 MeV Jovian and galactic electrons in comparison with the 3-10 MeV electron observed by the KET instrument from the period 1998 up to the end of 2003. Subsequent effects of these scenarios were also shown on electron modulation in general. It was found that this approach improved modelling of the post-1998 discrepancy between the model and KET observations but it also suggested the need for a time-dependent 3D electron modulation model to describe modulation during moderate to extreme solar maximum conditions.

The Centre for High Performance Computing (CHPC) in Cape Town (under the auspices of the CSIR's Meraka Institute) was launched in May 2007 and is a national centre initiated by the Department of Science and Technology. The first phase of computational procurement consists of an IBM e1350 Linux cluster with 160 nodes. It has successfully passed its initial benchmark requirements of 2.5 Teraflops. The facility is the largest computational science laboratory in southern Africa and provides a unique platform for high performance computing research in various fields.

This paper presents a case study on the scalability of several versions of the molecular dynamics code (DL_POLY) performed on South Africa‘s Centre for High Performance Computing e1350 IBM Linux cluster, Sun system and Lengau supercomputers. Within this study different problem sizes were designed and the same chosen systems were employed in order to test the performance of DL_POLY using weak and strong scalability. It was found that the speed-up results for the small systems were better than large systems on both Ethernet and Infiniband network. However, simulations of large systems in DL_POLY performed well using Infiniband network on Lengau cluster as compared to e1350 and Sun supercomputer.