Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.613454
Title: Explorative coastal oceanographic visual analytics : oceans of data
Author: George, Richard Llwyd Simon Frost
Awarding Body: Prifysgol Bangor University
Current Institution: Bangor University
Date of Award: 2013
Availability of Full Text:
Access through EThOS:
Access through Institution:
Abstract:
The widely acknowledged challenge to data analysis and understanding, resulting from the exponential increase in volumes of data generated by increasingly complex modelling and sampling systems, is a problem experienced by many researchers, including ocean scientists. The thesis explores a visualization and visual analytics solution for predictive studies of coastal shelf and estuarine modelled, hydrodynamics undertaken to understand sea level rise, as a contribution to wider climate change studies, and to underpin coastal zone planning, flood prevention and extreme event management. But these studies are complex and require numerous simulations of estuarine hydrodynamics, generating extremely large datasets of multi-field data. This type of data is acknowledged as difficult to visualize and analyse, as its numerous attributes present significant computational challenges, and ideally require a wide range of approaches to provide the necessary insight. These challenges are not easily overcome with the current visualization and analysis methodologies employed by coastal shelf hydrodynamic researchers, who use several software systems to generate graphs, each taking considerable time to operate, thus it is difficult to explore different scenarios and explore the data interactively and visually. The thesis, therefore, develops novel visualization and visual analytics techniques to help researchers overcome the limitations of existing methods (for example in understanding key tidal components); analyse data in a timely manner and explore different scenarios. There were a number of challenges to this: the size of the data, resulting in lengthy computing time, also many data values becoming plotted on one pixel (overplotting).
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.613454  DOI: Not available
Share: