West African sea level variability under a changing climate –what can we learn from the observational period?

Abstract

This study focuses on mean sea-level variability at the West African coast in the observational period (1993–2013) and its offshore waters, investigating its decadal variability, long-term trends and the large-scale climate patterns that are connected to its variability. To achieve this objective, statistically analyses is performed on several available data sets: sea-level data from tide gauges (Takoradi, Tema and Forcados), satellite altimetry (combined TOPEX/Poseidon, Jason-1 and Jason-2/OSTM), gridded sea-level reconstruction (Church et al., J Clim 17(13):2609–2625, 2004), meteorological reanalysis (NCEP), a high-resolution ocean model simulation driven by this meteorological reanalysis, and, observational data sets (The Hadley Centre Global Sea Ice and Sea Surface Temperature (HadISST1), and the Atlantic Multi-decadal Oscillation (AMO) index). Ghana is the only country along the West African coast with two relatively long sea-level records available (Takoradi and Tema), but with data quality concerns (Woodworth et. al., Afr J Mar Sci 29(3):321–330, 2007). Attempts are made to combine these two records, which cover different but overlapping periods, to construct a regional sea-level curve for Ghana (1929–1981) that may be regionally representative. A physical connection is identified between the AMO, sea-surface temperature and sea level in the Gulf of Guinea and mean sea-level trends and variability of the West African coast. It has been found that a stronger AMO is connected with higher mean sea-level in the Tropical Atlantic and in particular also at the Gulf of Guinea sea-level. This connection may explain the multidecadal variability of sea-level there, and in particular the negative trends between 1955 and 1975 and the positive trends thereafter. In addition, warmer sea surface temperatures in the Gulf of Guinea are also connected with higher sea-level, although a simple estimation based on reasonable assumptions of the thermal expansion of the water column is not sufficient to explain the connection between sea-surface-temperature and sea-level. More detailed modelling studies will be needed to explain this link. Although this study provides useful information for adaption strategies in Ghana, the research is unable to provide sea-level information between the years 1981 and 1993 because of lack of data.