Impacts of ENSO on coastal South African sea surface temperatures

Master Thesis


Permanent link to this Item
Journal Title
Link to Journal
Journal ISSN
Volume Title
The impact of El Niño Southern Oscillation (ENSO) on the Southern African inland climate is well documented and provides skill in the seasonal forecast of rainfall but little is known of the impact of ENSO on the ocean surrounding South Africa. The aim of this study is to assess the impact of ENSO on sea surface temperatures around the coast of South Africa and to calculate SST trends around the coast. I start by updating the study of Rouault et al (2010) on the very topic with an additional 10 years of data and two additional newer datasets which allow sampling closer to the coast where wind-driven upwelling is more active. The new highresolution ERA 5 reanalyzed climate dataset is also used to look at the atmospheric forcing of sea surface temperatures by ENSO. As in Rouault et al. (2010), I study five similar threedegree-long coastal regions around South Africa, namely: West Coast, South Coast, Port Elizabeth/Port Alfred, Transkei, Kwazulu-Natal and a larger offshore Agulhas Current area domain. Three SST datasets are evaluated in this study: the 1 ̊x1 ̊Optimal Interpolation sea surface temperature (OI SST) used by Rouault et al (2010), the 0.25 ̊x 0.25 ̊ Optimal Interpolation SST and the 4 km x 4 km Advanced Very High-Resolution Radiometer (AVHRR) Pathfinder SST version 5.3. The 0.25 ̊x 0.25 ̊OI SST resolvesSST anomalies better in these coastal regions as compared to 1 ̊x1 ̊ OISST. The difference in results among the three products concerning trends and correlation with ENSO is a cause for concern. The 4 km x 4 km AVHRR Pathfinder allows for SST to be extracted even closer to the coast but missing values are numerous and hamper the use of this dataset for ENSO composites and trend analyses. Results show a significant positive correlation with El Niño in summer at the monthly scale, reaching a maximum correlation of 0.45 at 3 months lag. Correlation is the highest in late summer. There is a negative correlation in the Agulhas Current area, opposite to those with ENSO and West Coast. The impact of ENSO on the coast of South Africa, West Coast and South Coast is due to change in surface wind speed with weaker upwelling favorable during El Niño leading to warmer than normal coastal water SST and stronger than normal Southeasterly winds during la Niña leading to cooler than normal coastal water. The wind perturbation is part of largescale basin-wide perturbations in the tropical Atlantic, in the South Atlantic high-pressure atmospheric system and in the westerly wind pattern of the midlatitude to the south. Non-ENSO related impact can be as important as ENSO related SST perturbation and is also linked to large scale perturbations in the South Atlantic. There is no relation between the strength of ENSO and the strength of the perturbation, and some ENSO events do not lead to the expected canonical warming or cooling. The large-scale SST perturbations seem to be caused by anomalous surface turbulent flux of latent and sensible heat and abnormal wind speed and direction. This study opens the possibility of seasonal forecasting of SST in the South Benguela upwelling system because of the positive lag correlation between ENSO and SST with ENSO leading SST.