Show simple item record

dc.contributor.advisorMudd, Simon Marius
dc.contributor.advisorMudd, Simon Marius
dc.contributor.authorMuir, Freya Margaret Ella
dc.date.accessioned2019-02-25T15:10:38Z
dc.date.available2019-02-25T15:10:38Z
dc.date.issued29/11/2018
dc.identifier.urihttp://hdl.handle.net/1842/35485
dc.description.abstractWith an increase in the probability of more severe precipitation events predicted by multiple climate change models, the necessity to accurately and efficiently identify a large coverage of areas at risk of flood damage is imperative. The increasing coverage and availability of very high resolution topographic data has aided the rapid development of high quality floodplain delineation and flood susceptibility mapping. However, the importance of the spatial variability of precipitation data applied in realistic hydrodynamic modelling of increased hurricane activity remains largely unexplored. This study applies a novel set of algorithms to a 372km2 area of topographic data derived from Light Detection and Ranging in Southwest Mexico, to investigate the influence cyclonic precipitation morphology has on floodwater flow predictions. An automatic floodplain delineation algorithm is run to objectively identify flood-susceptible catchments, before a cellular automaton model is applied to simulate inundation resulting from Hurricane Max, a Category 1 hurricane that occurred in September 2017. Using gridded Global Precipitation Measurement rainfall rate data and a Sentinel-2B multispectral satellite image captured immediately following the hurricane event, this study investigates to what extent remotely sensed data can be applied successfully to emulate a flood event and calibrate hydrodynamic flow model parameters to predict future inundation. It was found that simulated floodwater runoff from spatially variable incoming rainfall has a higher spatial intersection with the automatically classified water body pixels from the multispectral image. Additionally, the cellular automaton model simulated a higher peak discharge and greater extent of shallow floodwater depths in the river response to spatially variable incoming rainfall scenario compared to uniform rainfall. The results highlight the need for future consideration of precipitation morphology in flood modelling, and support the use of high abstraction models of Earth’s surface processes and fluvial hazard mitigation in data-sparse regions.en
dc.language.isoenen
dc.publisherThe University of Edinburghen
dc.subjectflooden
dc.subjectfloodwateren
dc.subjecthazard mappingen
dc.subjectsatellite imageen
dc.subjectSentinelen
dc.subjectcellular automatonen
dc.subjectspatially variable rainfallen
dc.subjectuniform rainfallen
dc.subjectLSDTopoToolsen
dc.subjectHAIL-CAESARen
dc.subjectMSc Geographical Information Scienceen
dc.subjectGISen
dc.titleSimplistic hydrodynamic modelling and multispectral image classification: predicting floodwater response to cyclonic precipitation in Southern Mexicoen
dc.typeThesis or Dissertationen
dc.type.qualificationlevelMastersen
dc.type.qualificationnameMSc Master of Scienceen
dcterms.accessRightsRestricted Accessen_US


Files in this item

This item appears in the following Collection(s)

Show simple item record