Relative Contribution of External Sources of Mean Sea-level Variations at Port Sudan, Red Sea

• Published in: Estuarine, coastal and shelf science Vol. 42; no. 1; pp. 19 - 30
• Main Authors: Sultan, S.A.R., Ahmad, F., Nassar, D.
• Format: Journal Article
• Language: English
• Published: London Elsevier Ltd 01.01.1996; Elsevier
• Subjects: Brackish; Earth, ocean, space; Exact sciences and technology; External geophysics; external sources; Marine; Physics of the oceans; Port Sudan; Red Sea; relative contribution; sea level; Surface waves, tides and sea level. Seiches; Red Sea; Africa; Sudan; ISW, Red Sea; Red Sea; Port Sudan; sea level; relative contribution; external sources; Amplitude; Statistical analysis; Atmospheric pressure; Annual variation; Variability; Wind effect; Evaporation; Sea level; Spectral analysis; Spectrum; Maximum; Atmospheric temperature; Air temperature
• ISSN: 0272-7714; 1096-0015
• DOI: 10.1006/ecss.1996.0002

Analysis of daily mean sea level over a 6-year period (1986–91) reveals, for the first time, the presence of a semi-annual component in addition to the annual component. Their amplitudes are 6·6 and 11 cm respectively. Variances explained by the two cycles vary significantly over the study period and average 16 and 44% respectively. The annual changes are induced by the prevailing wind regime while the semi-annual variation is associated with the evaporation rate. The spectrum of evaporation shows that the semi-annual cycle is the dominant variant in the moisture flux. Water temperature and wind speed exhibit slight semi-annual variations too. On the other hand, spectral analysis of atmospheric pressure and air temperature shows that their variability is mainly due to the annual cycle. The maximum hydrostatic response of sea level is only about 15% and lags 2 days behind with atmospheric pressure. The cross-spectrum between daily residuals of sea level and those of atmospheric pressure shows that the hydrostatic response is frequency-dependent with a maximum response occurring at 0·125 cycles day −1. An autoregressive first order model [AR(1)] has been calibrated to fit the daily residuals which explains 67% of the variance.