Aeolian sand transport models, widely employed by coastal scientists and managers, assume temporal and spatial homogeneity within the saltation field. This research questions that assumption by demonstrating that the saltation field is event-driven, therefore indicating that the saltation field is not temporally steady. The findings from this research may explain a portion of the conclusions from previous studies that indicated inequalities between model-estimated and field-measured aeolian sand transport. The relationship between unsteadiness in a turbulent wind field and pulses in a sand transport field was investigated on a beach near Shoalhaven Heads, New South Wales, Australia. Microphone-based saltation sensors, “miniphones,” and thermal anemometers (both instruments constructed exclusively for this field experiment) were co-located (0.02 m separation on center) and deployed between 0.01 and 0.0225 m above the bed, and sampled at 6000 Hz. Average grain size at the field site was 0.30 mm. Five runs totaling 2050 seconds of wind and saltation data were analyzed. The continuous wavelet transform, using the Morlet wavelet base, was the principle method for analyzing the wind and saltation records. The cross continuous wavelet transform was used to analyze the wind and saltation time series concurrently. Wind, saltation, and cross events were discerned by selecting wavelet power coefficients between wavelet scales of 0.4 and 3.0 seconds and with coefficients exceeding the 95% confidence interval. Average event spacing was 6.10, 6.50, and 6.73 seconds for the wind, saltation, and cross events, respectively. The average event spacing measured in this research was compared to the empirical-based model presented by Rao, Narashimha, and Narayanan (1971). The correspondence between the model and this research strongly suggests that bursting-type coherent structures were present. The durations of average wind, saltation, and cross events were 1.87, 2.10, and 1.73 seconds, respectively. Integral time scales, calculated using normalized auto correlation and power spectral density analysis, were approximately two seconds for the wind and saltation systems. The temporal coincidence of the integral time scale estimations and the event durations for the wind and saltation system strongly suggests that wind events are driving sand transport events.

Sand ramps are widespread but understudied landforms which have the potential to provide detailed palaeoenvironmental information in dryland regions. This thesis investigates the utility of sand ramps as late Quaternary palaeoenvironmental archives by addressing two research questions: (1) What are the main controls on sand ramp formation in southern Africa? (2) What does the sand ramp record tell us about late Quaternary palaeoenvironments in southern Africa? The distribution of sand ramps in southern Africa was surveyed using Google Earth™ and 75 features were identified in southern Namibia. Ten of these sand ramps, and an additional feature from South Africa, were studied in the field. Sediments and morphology were mapped and a total of 64 OSL dates, 96 sediment samples, 10 heavy mineral assemblages and OSL sensitivity data from 8 samples were examined. The distribution of sand ramps suggests formation is dependent on (1) sediment supply, (2) accommodation space, (3) persistent unidirectional wind and (4) a variable semi-arid to arid climate. Chronologies and sediment analyses indicate individual sand ramp accumulation is locally controlled by sediment supply modulated by the availability, and nature, of the accommodation space. Comparison between the Namibian and South African sand ramps suggests considerable regional variation in the factors controlling sediment supply. The Namibian sand ramps show an affinity to local ephemeral river channels. Periods of dated sand ramp activity in multiple features are interpreted as periods of increased regional fluvial activity. Significant activity occurred at ~21-12 ka (with peaks at 21-18.5 ka and 14.5-12 ka), ~8.5-7.5 ka and ~2 ka. Some activity is also indicated at 85-65 ka and 45-35 ka. These results correspond well to regional records. Overall, this study demonstrates that sand ramps can provide palaeoclimatic information on both the local and regional scale but only if a number of caveats are taken into consideration.

Le rôle des particules de poussières comporte une grande incertitude des modèles climatiques, alors qu’ils constituent une grande influence sur le bilan radiatif global et l’apport en nutriments pour divers écosystèmes. Ce projet de recherche permet d’évaluer la composition de particules de poussières en Namibie, en fonction de trois vallées constituées de rivières éphémères en région côtière. La sélection des vallées s’est basée sur leur fréquence importante d’émissivité, observée par image satellitaire. En 2017, des mesures en temps réel ont été prises ainsi que des échantillons de surface, dont l’analyse des sols et filtres s’est effectuée en laboratoire. Les propriétés des sols, c’est-à-dire la distribution de la taille, de la minéralogie et composition élémentaire, seront traitées dans ce mémoire. Le but étant d’évaluer les sources d’émission et le potentiel de transport des poussières, par une meilleure connaissance de l’interaction des sols en relation avec les particules transportées par le vent. Il a été observé que le transect de la vallée du Huab possède une taille de particule plus fine en raison de divers facteurs, dont une rivière éphémère plus active. Le transect de la vallée de Omaruru quant à lui possède une composition en minéraux plus variée, dont une petite concentration en éléments lourds. Pour le transect de la vallée du Kuiseb, en raison du site d’échantillonnage et une forte influence de la mer de sable, des particules plus près de la fraction du sable ont été identifiés en plus grande quantité par rapport aux autres sites. Il est donc suggéré que, malgré une grande concentration de particules fines pour le Huab, son processus d’émission demande une fréquence de vent et un environnement propice à l’émission des particules par suspension. Contrairement à certaines vallées telles que le Hoanib ou le Hunkab, le Kuiseb et le Omaruru semblent influencés par la désintégration des argiles lors de la saltation et le bombardement de particules plus grossières. / Dust particles are primarily known to modify radiative forcing as well as nutrient contribution of coastal regions, but account for an important part of climatic model uncertainties. This project evaluates the characteristics of dust particles in river valleys of Namibia within three valleys of the coastal region. The selection of sites was determined by their frequency of emissions observed by satellite imagery. In 2017, fieldwork measures were taken as well as surfaces samples, which were analysed for their particle size distribution, their mineralogy as well as their elemental composition. The results presented in this thesis look at potential transport emission, to better understand interactions between soil and airborne particles. It was observed that the transect of the Huab valley presents the finest distribution of particles, primarily because of its active river providing finer particles to dust sources. The transect of the Omaruru has a more varied particle size distribution relative to the Huab valley and the Kuiseb, but has some coarser particles present in small quantities. Because of its location, the transect of the Kuiseb is strongly influenced by the Namib sand sea, which is reflected in its particle size distribution. Because of existing literature about dust particles, it is possible to suggest that the Huab valley needs stronger winds to be able to be emissive as it erodes predominantly through a modified direct suspension mechanism. Saltation and bombardment seem to play a more important role in the Kuiseb valley than the other valleys along the coast, such as the Hoanib or the Hunkab, because of its important composition of sand and geomorphology proprieties.