No description available.

No description available.

No description available.

This thesis documents a series of studies performed on the lower atmosphere over the region of the Ross Ice Shelf, Antarctica, and its surroundings. In particular, much of the thesis focuses on the area in the vicinity of Ross Island, a mountainous protrusion in the far north-west of the permanent floating ice shelf. Weather in both the smaller and larger regions is naturally complex and generated by a range of localised and larger scale interactions. In order to better understand the meteorology of the Ross Ice Shelf, including Ross Island, we produce a synoptic climatology of the region based on surface wind output provided by the ERA Interim reanalysis. Output is taken from 1979 to 2011 and thus represents a much longer time scale than covered by previous studies of Ross Ice Shelf winds. The climatology is generated through a clustering routine based on the widely-used $k$-means technique. The results of the routine are discussed and we find that the reanalysis is capable of representing the previously reported features of the region. Cluster composites are also shown to be coherent between reanalysis output and data collected by in situ monitoring devices. We confirm that the Ross Ice Shelf Air Stream (RAS), a jet of fast-moving air that propagates from the Siple Coast across the ice shelf, is a robust feature of the climatology of the region and we find that it has a large impact on the surface temperature. The analysis is continued with reference to two widely studied modes of internal variability, the Southern Annular Mode (SAM) and El Nino-Southern Oscillation (ENSO), which are known to affect local conditions in the Ross Sea region via modulation of the Amundsen-Bellingshausen Sea low. Reanalysis output and results from the clustering routine allow us to examine the impacts of these modes upon the Ross Ice Shelf and Ross Sea in unprecedented detail. Further, we are able to tie changes in the mean pattern to variability within and between particular clusters, allowing us to ascertain the dominant synoptic patterns in forcing the mean variability. The impact on surface temperatures for both modes is found to be high and significant, which we explain with reference to changes in circulation patterns. We further use the results of the clustering algorithm to explore the climatology of the region surrounding Ross Island. By producing composites of local in situ records based on the clustering technique described previously, we are able to generate a climatology of the region that is not hampered by gaps in the observational record. We find that the climate of Ross Island is sensitive to RAS events, due to the ability of strong flows to dramatically increase the temperature. At Scott Base, on the southern tip of the Hut Point Peninsula, the temperature is found to be particularly sensitive to these events. McMurdo Station, which is located less than 3 km away, is observed to be much less sensitive, due to the modulation of synoptic flows by localised topographic influences. Particularly salient is the difference in temperature trends between these two locations, which we show to be statistically significant in the annual and seasonal means from 1979 onwards. By applying a novel temperature reconstruction technique based on the output of the clustering routine, we are able to assess the contribution of changes in circulation to temperature trends at these two locations. We conclude that a large amount of the change in temperature at Scott Base can be explained through circulatory variability over the Ross Ice Shelf. However, the trend at McMurdo Station can not be explained using this technique and may be the result of extremely localised forcing. Data availability in Antarctica is widely known to be low, due to the relative sparsity of observations and ongoing problems with data collection due to extremely inhospitable conditions and challenging logistical considerations. The lack of data at the mesoscale has hampered the understanding of localised processes in the Antarctic atmosphere that may be important for forecasting. Through the development and deployment of a distributed system of atmospheric sensors called

No description available.

No description available.

No description available.

The use of relatively small nonerodible elements (gravel) in stabilising a mobile surface was investigated. The aim of this study is to determine the different factors that affect the sand erosion due to the presence of small immobile rough elements. A regular array of spherical glass marbles was distributed on a nine metre length in the working section of a 12 metres long boundary layer wind tunnel. A bed of dune sand was spread over the marbles filling the gaps and covering the marbles. Four different shear velocities were used with two different marble spacings. The experimental results showed that the sand erosion and the sand flux along an isolated sand bed containing both mobile and immobile elements is neither uniform nor steady. The marbles start to be exposed at a certain position along the wind tunnel (the position of the maximum flux gradient and thus the maximum erosion). This is followed by a gradual exposure of the rows of marbles along the wind tunnel downwind of the initial exposure. This process continues until all the marbles are exposed downwind. The marbles are also exposed upwind of the initial exposure but this propagates at much lower rate and some marbles remain covered with sand at the end of the experiment. The sand flux measured at the downwind end of the working section remains constant for some time. When the marbles start to expose at some distance upwind, it starts to decrease rapidly. This indicates that the effect of the nonerodible elements on sand flux is not a local effect. A numerical model was developed to simulate the mechanism of the sand erosion observed in the wind tunnel experiments. The model helps in determining the different factors that influence the sand flux and sand erosion due to the exposure of the nonerodible elements. The model succeeded in simulating the erosion process and in providing a qualitative prediction of the sand level and sand flux distribution along the sand bed.

Meteorologists have spent decades attempting to predict the weather over extended periods of time. Complex models of up to several million variables can only produce reliable predictions of up to four days. By representing the atmosphere in a multi-dimensional 'phase space', we hope to find preferred areas of this space where the weather will persist. Using a simple simulation model we applied 9 clustering methods, some of which are new, to the simulated data. These methods represent 3 different levels of interactions between the user and the method. While developing new cluster methods, we also developed an outlier method which is shown to be better than 16 current multivariate outlier methods, based on a real dataset. The results of the simulation studies indicate that the more interaction between the user and the method, the better the outcome. Next we adapted the usual Ward's, and Caussinus and Ruiz's clustering methods to take time into consideration. This created 6 new time constraint clustering methods which we applied to simulated data from a new time dependent simulated model. Consistent patterns were found and the results also indicate that if we apply the usual Ward's clustering method on suspected time dependent data then we would achieve the best outcome only 35% of the time, at most. Finally we looked at ways of sieving transient observations from cluster groups and highlighting significant transitions by applying several techniques to a meteorological dataset.

No description available.