Search

In this thesis, I present analysis and interpretation of the multiwavelength spectra and variability of Active Galactic Nuclei (AGN). The most luminous sustained sources in the Universe, these powerful objects are consistent with being the result of gas accretion on to central galactic supermassive black holes. Due to their compact sizes, the inner regions of AGN cannot be spatially resolved by conventional means, so we must instead use spectroscopy and temporal monitoring to determine their composition and structure. I undertake a number of studies of the spectral energy distributions (SEDs) of AGN, using data from infrared-X-ray bands and employing a range of numerical models. Results from SED modelling of 11 moderate redshift (1.5 < z < 2.2) AGN are presented, in which there is a selection bias towards nuclei with cooler accretion discs. I find that the peak of the SED is sampled by our data for 5/11 objects, thereby breaking several of the model degeneracies that affected previous studies. This results in stronger constraints on the physical processes at work in these AGN, and provides a powerful tool with which I examine and discuss the relationships between the various radiating components, including those of the emission line regions, dusty torus and host galaxy. I then explore the nature of four 'hypervariable' AGN, for which the origin of their extreme variability is currently unknown. Through an investigation of their SEDs, I find that either an accretion rate change, or gravitational microlensing by a star in a foreground galaxy, are energetically consistent with the data. The new insights provided by this work lead me to suggest several worthwhile routes for the future development of research in these areas. With the next generation of telescopes, satellites and surveys on the horizon, it will be possible to build on my results, to further our understanding of AGN.

This thesis explores the inherent temporality embedded within the complex stratigraphic sequence of the ‘tell’ site of Çatalhöyük, an important Anatolian Neolithic settlement situated upon the Konya Plain, South-Central Turkey. Recently the Çatalhöyük Research Project has digitized all of its single context excavation data, fully integrating their digital archive within an intra-site GIS, as an aid to analysis and interpretation. This process of digitisation excludes the Harris matrix, which, despite being integral to the recording system, and the main source of relative temporal data for the development of the site, remains an analogue mode of analysis. This research digitally visualises the stratigraphic sequence, both dynamically and intuitively (moving beyond conventional archaeological methods of phasing and periodisation), utilising the temporal capabilities of ArcGIS 10 to generate robust and dynamic intra-site spatiotemporal models. By focusing upon two case studies as a ‘proof-of-method’ (a ‘typical’ sequence of two fully excavated superjacent buildings – Buildings 65 and 56, and one unusually large and well preserved burnt building – Building 77), the experimental appending of stratigraphically-based temporal data onto the spatial component of an excavation dataset within a GIS, and subsequent analysis of associated material culture within its spatiotemporal context, has proved an innovative way to articulate and visualise the site’s space through time. This represents a transparent, repeatable and critical approach to post-excavation analysis, using current computing technologies. Focusing upon integrated spatiotemporal analysis of excavation data and associated material culture within these models also facilitates greater understanding of the relationship between space and time in archaeology within the data structure of primary recording in archaeological excavations. The resultant spatiotemporal animations combine this data as a new type of ‘visual narrative’ that may help illustrate the social meaning of these structures, potentially telling the bigger story of the site within its wider context of the Anatolian Neolithic.

Humans have an innate ability to communicate visually; the earliest forms of communication were cave drawings, and children can communicate visual descriptions of scenes through drawings well before they can write. Drawings and sketches offer an intuitive and efficient means for communicating visual concepts. Today, society faces a deluge of digital visual content driven by a surge in the generation of video on social media and the online availability of video archives. Mobile devices are emerging as the dominant platform for consuming this content, with Cisco predicting that by 2018 over 80% of mobile traffic will be video. Sketch offers a familiar and expressive modality for interacting with video on the touch-screens commonly present on such devices. This thesis contributes several new algorithms for searching and manipulating video using free-hand sketches. We propose the Visual Narrative (VN); a storyboarded sequence of one or more actions in the form of sketch that collectively describe an event. We show that VNs can be used to both efficiently search video repositories, and to synthesise video clips. First, we describe a sketch based video retrieval (SBVR) system that fuses multiple modalities (shape, colour, semantics, and motion) in order to find relevant video clips. An efficient multi-modal video descriptor is proposed enabling the search of hundreds of videos in milliseconds. This contrasts with prior SBVR that lacks an efficient index representation, and take minutes or hours to search similar datasets. This contribution not only makes SBVR practical at interactive speeds, but also enables user-refinement of results through relevance feedback to resolve sketch ambiguity, including the relative priority of the different VN modalities. Second, we present the first algorithm for sketch based pose retrieval. A pictographic representation (stick-men) is used to specify a desired human pose within the VN, and similar poses found within a video dataset. We use archival dance performance footage from the UK National Resource Centre for Dance (UK-NRCD), containing diverse examples of human pose. We investigate appropriate descriptors for sketch and video, and propose a novel manifold learning technique for mapping between the two descriptor spaces and so performing sketched pose retrieval. We show that domain adaptation can be applied to boost the performance of this system through a novel piece-wise feature-space warping technique. Third, we present a graph representation for VNs comprising multiple actions. We focus on the extension of our pose retrieval system to a sequence of poses interspersed with actions (e.g. jump, twirl). We show that our graph representation can be used for multiple applications: 1) to retrieve sequences of video comprising multiple actions; 2) to navigate in pictorial form, the retrieved video sequences; 3) to synthesise new video sequences by retrieving and concatenating video fragments from archival footage.

Modified bases in RNA molecules are common, with over one-hundred individual modifications reported to date. In transfer RNA (tRNA), these modifications help hold the molecule in its proper conformation, allow the aminoacyl-synthetases to attach the amino acid to the correct tRNA for transfer to the ribosome, and in the correct matching of the codon-anticodon pair at the ribosome. Two positions on the tRNA molecule are significant in this last action; positions 34 and 37. Position 34 is the first base in the anticodon, and pairs with the last base in the codon during translation; modifications at this point allow the anticodon to pair with multiple codons. Modifications at position 37 help stabilise the anticodon loop of the tRNA, and help the efficiency of the translation. The threonylcarbamyoladenosine (t6A) modification at position 37 is required for the efficient translation of ANN codons. At least four enzymes, TsaBCDE, are involved in the formation of this modification, and are essential for bacterial viability. In this study, the interactions between the TsaB, TsaC, TsaD and TsaE have been investigated. Protein complex formation has been identified both in-vitro and in-vivo between TsaB and TsaD. The proteins interact in a 1:1 ratio, and the results indicate that this interaction stabilises the TsaD enzyme. The interaction appears to be unchanged over a physiological range of temperatures. No stable complex was identified between TsaC or TsaE with any of the other enzymes, however calorimetry suggests that there is an interaction between TsaE and both TsaB and TsaD. TsaC is believed to synthesise the threonylcarbamoyladenylate intermediate from threonine, bicarbonate and ATP; the activity of the enzyme complex (as measured by ATPase activity) in a media containing these compounds is increased in the presence of TsaC. If the cell is prevented from maintaining a source of t6A, it mounts a number of transcriptional responses. The transcription of certain amino acid biosynthetic genes is increased, corresponding to the tRNAs requiring the t6A modification, and in the transport and conversion of sulphates. This reflects the cellular response to starvation of an amino acid; the inability to incorporate the amino acids into the polypeptide chain due to the hypomodified tRNA is interpreted as the absence of that particular amino acid. Other transcriptional effects can be interpreted as a consequence of translational stalling where the t6A modification is required, leading to alterations in transcriptional promotion. TsaB and TsaD have also been shown to bind a small number of compounds in a high-throughput fragment-based drug discovery assay. Some of these have been shown to potentially affect the ATPase activity of the enzymes.

Landscape in Roman literature is manifest with symbolic potential: in particular, Vergil and Ovid respond to ideologically loaded representations of abundance in nature that signal the dawn of the Augustan golden age. Vergil's Eclogues foreground a locus amoenus landscape which articulates both the hopes of the new age as well as the political upheaval that accompanied the new political regime; Ovid uses the same topography in order to suggest the arbitrary and capricious use of power within a deceptively idyllic landscape. Moreover, for Latin poets, depictions of landscape are themselves sites for poetic reflection as evidenced by the discussion of landscape ecphrases in Horace's Ars Poetica. My thesis focuses upon the depiction and refiguration of the locus amoenus landscape in the post-Augustan epics of the first century AD: Lucan's Bellum Civile, Valerius Flaccus' Argonautica, Statius' Thebaid and Silius Italicus' Punica. Landscape in these poems retains the moral, political and metapoetic force evident in the Augustan archetypes. However, I suggest that Lucan's Neronian Bellum Civile fundamentally refigures the landscapes of Latin epic poetry, inscribing the locus amoenus with the nefas of civil war in such a manner that it redefines the perception of landscape in the succeeding Flavian poets. Lucan perverts the landscape, making the locus horridus, a landscape of horror, fear and disgust, the predominant landscape of Latin epic; consequently, the poems of Valerius, Statius and Silius engage with Lucan's refiguration of landscape as a means of expressing the horror of civil war. In the first part of my thesis I examine archetypal landscapes, including those of the Augustan poets and Lucan's Bellum Civile. Taking an approach which engages with literary reception theory and the concept of the â horizon of expectationâ as a framework within which literary topographies can be understood as articulating a response to the thematics of civil war, in the second part of my thesis I demonstrate the manner in which landscapes represent a coherent and paradigmatic response to Lucan's imposition of his civil war narrative within the literary landscape of Roman literature.

The aim of this dissertation was to investigate the potential for building ground energy systems (BGESs) in the UK by analysing key country specific attributes. More recently these systems have been presented in the UK as having the ability to reduce carbon emissions in comparison to conventional heating and cooling plant. Due to the essential transition to a low carbon economy, this was the main driver and motivation to conduct the research in this field. An initial distinction was made between closed and open loop systems. The former system type utilises a network of pipes installed in the ground to abstract and reject heat via a circulating fluid. Using existing well technology, open loop systems abstract and discharge groundwater whereby the heat transfer between the ground and building effectively occurs within the building via a heat exchanger. Both technology approaches are relevant in the UK where the temperate climate provides a thermodynamic advantage for heat rejection and abstraction from a building, i.e. to aid cooling and heating respectively. A reversible heat pump unit is commonly used in non-domestic installations to allow the prevailing temperature from the ground side system to be reduced or increased accordingly. The research project was split into three main themes. Firstly, a technology specific spatial review of the geology, hydrogeology and climatic effects throughout the UK was conducted using a GIS software package. Secondly, a design phase analysis concentrated on the impact of climate change by applying different weather files to thermal models, different carbon factor projections for electricity, part load sizing, energy prices and the marginal cost of carbon abatement over the lifetime of the system. Finally, an operational phase analysis concentrated on the actual performance of four existing systems monitored over a 12 month period. Four different case study buildings were used for the design phase analysis, and a different set of four building for the operational phase analysis. Using referenced values for the thermal properties of the geology, the resulting heat exchanger length for vertical closed loop systems is primarily led by the respective heating and cooling loads. For the same building type this varied significantly in the three different UK climate types. Following this the thermal conductivity and starting borehole temperature was also shown to make significant differences. For horizontal closed loop systems the heat exchanger length is again led by the building location. The lead ground related parameter is the saturation of the ground as this is likely to vary considerably from location to location. A high thermal conductivity is not necessarily beneficial as this can increase the alignment with the prevailing air temperature and reduce the thermodynamic advantage. There is limited potential for open loop systems due to availability of productive aquifers. The yield analysis showed very large ranges in yield and hence the peak heating and cooling capacity from a single well. This was true when analysing particular aquifer zones and also when comparing sub regions of different aquifer types. The percentage change in peak heating and cooling capacity and annual demand due to climate change was very significant for all the four buildings analysed. The use of dynamic carbon dioxide projections in comparison to static factors currently stated in government publications also highlighted significant variations in projected carbon dioxide savings for a BGES over a 20 year period. The bivalent energy and economic assessment underlined the difference in marginal utilisation and cost effectiveness by varying the size of the BGES relative to the peak load. As the utilisation reduced the cost effectiveness also reduced. These results were translated through to huge variances in the marginal cost of carbon abatement. The operational monitoring results showed that existing BGESs are not performing according to efficiencies quoted by manufacturers. This sometimes led to marginal or negative carbon reduction and operational cost savings versus conventional plant. The poor performance was linked to the commissioned flow rates through the heat pump, buffer tank sizing, the heating distribution installation, maintenance and servicing frequency, insufficient owner knowledge and part load efficiency of heat pump plant.

Student evaluation of teaching (SET) has become a central feature of university teachers' performance assessment and Hong Kong is no exception. However, a number of writers have called into question its accuracy as a performance measurement tool. Regardless of the accuracy question, research indicates that teacher behavioural characteristics are amongst the most significant factors affecting student ratings of teacher performance. One aspect of university teacher behaviour that has received little attention is classroom leadership style.;In this study, a classroom leadership instrument was developed for use in a Hong Kong university context. This instrument was based on the transformational-transactional leadership model because there is evidence that aspects of the model have potential for enhancing student evaluation of teaching scores. Additionally, the transformational style has been associated in the literature with a number of benefits that have prima-facie relevance to the university classroom. The literature has also indicated a possible gender effect in that any enhancement to teaching scores might be especially pronounced in the case of female students.;The central outcome of the study was the development of an instrument that was capable of the valid and reliable measurement of classroom leadership style in a Hong Kong university context. A key finding was that employment, by university teachers, of the transformational and active transactional dimensions of classroom leadership style were significantly and positively associated with student perception of desirable classroom leadership outcomes. In this study, no gender effect was detected. Furthermore, there was an indication that adoption of the transformational and active transactional style in the classroom could have a substantial positive effect on teacher SET scores although this finding was not conclusive due to sample size limitations. Finally, the findings lent support to the view that the transformational-transactional leadership notion is robust across cultures.

The design of biomechanical devices is often difficult because it is not feasible to test them within human or animal patients. Initial designs may be tested using mechanical rigs or finite element models to simulate the complex loading conditions within biological environments. Recent advances in computer technology have allowed computer simulations to become more detailed. However, educated assumptions must be made to simplify the conditions that will effect the device. Highly accurate results can be obtained but it is essential that these results are verified using a mechanical test rig. This will identify if human error or inaccurate assumptions have occurred. Mechanical test rigs were designed to simulate the forces acting within four different biomechanical systems. This involved simulating the forces applied to the lumbar spine, the greater trochanter of the hip, and to the shin; for the testing of inter-vertebral cages, hip protectors and floor coverings, and shin pads respectively. Load cells were used to measure the resultant impact forces applied to the hip and shin. The time-domain data were recorded for each impact and from these data it was possible to calculate the maximum impact force and the total energy absorbed by the greater trochanter. Mechanical testing of intervertebral cages involved pressure sensitive paper to measure resultant pressure trends at the interface between the vertebral end plate and the cages. This instantly revealed pressure trends and computer programs were written to determine the contact area, calibrate pressure values, and to allow the improved visualisation of results. Results from the mechanical test rigs are also suitable for validation of finite element analysis studies. New information has been generated to improve understanding about the pressure trends under intervertebral cages and the key design features have been identified. The hip test rig was used in the design of a novel hip protector, to determine the role of floor coverings in an impact, and to compare numerous hip protectors using typical impact conditions. Results show that a combination of energy absorbent flooring with a hip protector is the best environment for a reduced risk of a hip fracturing. Design recommendations were made regarding the material composition and the thickness of the materials used. This allowed the novel hip protector to be designed as small as possible but still sufficiently reduces the impact force from a person falling onto their hip. The hip test rig was modified to simulate the shin and was used to compare several commercially available shin guards. New information was generated to identify which design features are best to reduce impact forces. Further developments have been made to the first version of the hip impact test rig. The latest version of this hip test rig is currently being used as a benchmark for all hip protectors and is a likely contender as the basis of a standard testing system for a European Standard.

Aggregates provide a distinct niche within the water column with elevated levels of nutrients compared to the surrounding water. Differences have been observed between aggregate and planktonic communities of ammonia oxidising bacteria within the environment, but such differences have not been explained. Ammonia oxidising bacteria are important due to their role in the first rate-limiting step of nitrification in which nitrogen is converted from its most reduced to its most oxidised form. The aim of this study was to determine mechanisms responsible for aggregate colonisation using a model system. To achieve this aim, a model aggregate system was developed in which colonisation of aggregates by ammonia oxidising bacteria under different conditions could be determined. Candidate ammonia oxidiser strains, Nitrosomonas europaea and Nitrosospira tenuis, were used to demonstrate the effect of the sampling method on observed distribution of AOB within the model system. The degree of bias of the b-AMO and CTO PCR primer sets was assessed and indicated that the predominance of Nitrosospira species in many environmental surveys was not due to primer bias. The composition of particulate communities of AOB determined the extent of colonisation by other species of AOB. The immediate colonisation of aggregates by N. tenuis suggested that differences in particulate and planktonic AOB communities were not solely due to differences in substrate availability and activity of differing AOB under such conditions, although the ability of attached communities of AOB to respond more quickly when substrate becomes available was again illustrated. This study provides evidence of the differences between particle-attached and free-living AOB and indicates how such differences can be related to the ecology of these organisms in natural environments.

GroEL is a heat shock protein, involved in the folding of denatured proteins under stressed conditions. Flagellin is a structured protein of flagella, and can undergo phase variation so that at any one time, synthesis of flagellin occurs from either of two genes, fliC or fliB. Porins, of which OmpC is an example, allow passage of nutrients and small molecules through the bacterial cell outer membrane. All of these proteins have previously been described as being immunogenic. In the work described here, antibodies specific to each of these antigens were detected in both innately susceptible and resistant mice after infection with an attenuated strain of S. typhimurium, and after subsequent challenge with the virulent organism, suggesting that they may each be involved in protective immunity. The immunogenicity of a vaccine consisting of these four recombinant Salmonella antigens was assessed in the mouse model. GroEL- and flagellin-specific antibodies were detected in innately susceptible mice after immunisation together with the adjuvant DDA. Moreover, analysis of IgG isotypes showed increased titres of both IgG1 and IgG2a, the latter indicating that cell-mediated immunity had been generated. However, challenge of immunised mice with virulent S. typhimurium demonstrated that the protective efficacy of the vaccine was of only low-level. Immunisation of innately susceptible and resistant mice using a tetravalent DNA vaccine expressing all four antigens resulted in increased antibody against GroEL, FliC and FlijB in both mouse strains. As with the subunit vaccine, IgG isotype analysis showed increased titres of both IgG1 and IgG2 indicting that both Type 2 and Type 1 helper T-cell responses had been elicited. However, considerable variation was observed within immunised mouse groups and so the protective efficacy of the vaccine was not determined.