Use this URL to cite or link to this record in EThOS:
Title: Modelling and prediction of seabed morphodynamics using a multi-dimensional statistical method with forcing
Author: Bakare, A.-M.
Awarding Body: University College London (University of London)
Current Institution: University College London (University of London)
Date of Award: 2011
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
This research identifies and validates a data-driven statistical method for morphological and morphodynamic modelling and prediction. The method, called the spatial regression model has been modified to account for a single forcing factor. In the model, spatial behaviour across a domain is accounted for as surface functions; changes between successive surfaces then explain the spatial and temporal evolution, which is used to calculate a prediction. In the model extension, the covariability between the time series of the morphology and forcing are assessed to derive an estimate that scales future forcing. The model is applied to idealised morphological scenarios and two study sites, which are the nearshore zone of Poole Bay, governed by wave and anthropogenic factors, and the Great Yarmouth sandbank system, governed by tidal, wave and storm conditions. The idealised scenarios results show that the model identifies morphological evolution. For scenarios with temporal periodicity, stochastic effects and noise, it generated predictions with Brier Skill Scores (BSS) of ≥0.97. Accounting for forcing improves the prediction for scenarios with complex behaviour. At the Great Yarmouth site, a BSS of 0.65 is obtained using the morphological characteristics only. In using the wind time series as a proxy for the forcing, the skill score decreases. For the Poole Bay site, a 0.64 BSS is obtained assessing the morphological characteristics only. Accounting for the wave time series as the forcing improves the score compared to that without forcing for the same time period. In assuming no change occurs across the domain, the model improves on results for the scenarios but generates larger errors at the real sites. Model sensitivity is dictated by the characteristics and complexity of the morphological behaviour and the spatial and temporal resolution of the associated time series datasets, which in turn influences the prediction accuracy. The spatial regression model can be a useful tool for morphodynamic modelling applications at large scales, where accounting for external forcing conditions can improve prediction results, given certain behaviour and data properties.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available