Browsing by Author "Barnes, Jarryd"

Now showing 1 - 1 of 1
  • Thumbnail Image
    Item
    Altimetry comparison, backward wave refraction and extreme wave analysis of NCEP data: Cape Point to Richards Bay
    (Stellenbosch : Stellenbosch University, 2018-12) Barnes, Jarryd; Schoonees, K.; Rautenbach, C.; Stellenbosch University. Faculty of Engineering. Dept. of Civil Engineering.
    ENGLISH ABSTRACT: Offshore wave climate along the south and east coast of South Africa has been investigated in this thesis. The design waves are needed to understand the offshore wave climate and necessary for the design of structures in this region. The wave condition data was produced by the WAVEWATCH III numerical model, i.e. NCEP data. The ultimate aim of the study was to derive extreme wave estimates off the South African coastline and assess its accuracy through comparison of various datasets. Comparisons of Satellite Altimetry data with NCEP and wave buoy data at two locations along the south coast (Cape Point and FA-Platform) was performed. Thereafter, wave conditions, obtained from 'backward' refracting nearshore data to the offshore NCEP locations, were performed in an attempt to validate NCEP data for East London and Richards Bay. Two approaches were conducted in order to backward refract, namely; Snell's Law (serving as a first estimate) and SWAN (fully describing the wave processes). The SWAN approach achieved offshore conditions by running a numerical model of various input conditions and creating a lookup table or matrix by means of an interpolation function using those general input conditions and output results. The methodology is important as neither the SWAN outputs nor the NCEP data could be regarded as 'ground truth'. An Extreme Wave Analysis was performed for the offshore NCEP data at six locations along the south and east coast of South Africa. An Exponential, Gumbel and 3-parameter Weibull distribution was conducted, where it was found that the latter was a best fit for the South African waters. An objective and conservative automated Peaks-over-Threshold (POT) value technique was used and wave heights for 1 in 1, 5, 10, 30, 50 and 100 year return periods with associated directions were obtained. Lastly, an analysis of the different distributions as well as percent exceedance graphs of the peak period per site was performed and compared.