Exchange through the Bab el Mandab

Abstract

During the last 10 years there have been a number of important developments in the understanding of the exchange flow in the Bab el Mandab. New and improved estimates of the fluxes of mass, heat and salt through the Strait have been made directly from oceanographic measurements and indirectly from air–sea flux data. Hydraulic theory has been applied to explain the seasonal variations in the exchange flow, and in particular the three-layer summer circulation. The effects of wind stress, friction, and seasonal sea-level variability on hydraulic control are addressed in this paper. Analyses of oceanographic data have failed to provide direct evidence of hydraulic control, but they have highlighted the importance of tidal scale variability and differences between real continuously stratified fluids and idealised layer flows often used in hydraulic theory. Other studies have investigated the coupling between the flow in the Strait and the circulation within the Red Sea.

Introduction

The Red Sea and the Mediterranean Sea are both almost enclosed, each having just one connection to the global ocean: the Strait of Gibraltar for the Mediterranean, and the Bab el Mandab for the Red Sea.¹ Both seas have large rates of evaporation that drive inverse-estuarine circulations resulting in high-salinity outflows from their respective straits. The Mediterranean outflow is thought to have a significant impact upon the circulation of the North Atlantic. It makes a significant contribution to the heat and salt budgets of the Atlantic (Reid, 1994), and diapycnal mixing in the outflow drives circulation in the Azores current (Jia, 2000). Similarly, Red Sea outflow water has a significant impact upon the thermohaline circulation of the Indian Ocean (Schott and McCreary, 2001; Beal et al., 2000). The main pathway of the Red Sea outflow water is at intermediate depths along the western boundary of the Indian Ocean; in the Somali undercurrent through the Mozambique channel and into the Aghulas current (Beal et al., 2000). Some of it may be exported to the South Atlantic (Valentine et al., 1993; Beal et al., 2000). The properties of the outflows are dependent not just upon the air–sea fluxes and other processes within the enclosed seas, but also upon the physical processes within the straits (Stommel and Farmer, 1953; Bryden and Stommel, 1984). Hydraulic control is often assumed to be the dominant process determining the exchange flows in the Strait of Gibraltar (e.g., the review by Bryden et al., 1994) and the Bab el Mandab (Smeed, 2000).

Despite many similarities between the Bab el Mandab and the Strait of Gibraltar, there are some striking differences. Most notable of these is the seasonal variability of the flow in the Bab el Mandab, a consequence of the prevailing monsoon climate. During the winter season (October–May) the flow can be characterised as a two-layer exchange: warmer, fresher water flows into the Red Sea in the surface layer and cooler, saltier water flows into the Gulf of Aden in the lower layer (Fig. 1A). A similar two-layer exchange is observed at Gibraltar. Whilst there is some seasonal variation in the outflow from the Mediterranean (Lafuente et al., 2002), the basic pattern of a two-layer exchange persists throughout the year. The summer regime in the Bab el Mandab is very different: the high-salinity outflow is almost arrested, the flow in the surface layer is reversed, and intermediate water from the Gulf of Aden flows in to the Red Sea between the two out-flowing layers (Fig. 1B).

Another difference between the Red Sea and Mediterranean Sea is that many more observations have been made in the Strait of Gibraltar than in the Bab el Mandab. Perhaps because of the lack of observation there have been fewer modelling and theoretical studies of the Bab el Mandab too. However, the first comprehensive set of observations to span a complete annual cycle were made in the Bab el Mandab from June 1995 to November 1996 by Murray and Johns (1997). Stimulated by these observations a number of studies have ensued including modelling and theoretical work, as well as analyses of Murray and Johns’ (1997) data. Now that many of these studies have been published it is timely to review the progress that has been made. As we shall see, recent studies of the Bab el Mandab have contributed much to our understanding of, and posed new questions about, exchange flows in general.

The principal questions addressed in recent studies of the Bab el Mandab are: what are the fluxes of mass, heat and salt through the Strait; what are the mechanisms responsible for the seasonal variation in the exchange flow; is the exchange flow hydraulically controlled; and, how are processes within the Red Sea coupled to the flow in the Bab el Mandab?

These questions are worthy of study not only for their intrinsic value, but also because of their bearing on broader questions. In particular the timescale for the circulation of the Red Sea is short compared to that of the global ocean; the Red Sea is thus relatively more sensitive to changes in climatic variables (e.g., sea level or the strength of the monsoon). A proper understanding of the exchange in the Bab el Mandab is of primary importance in understanding the variability of the Red Sea from annual to millennial timescales. For example, Siddall et al. (2003) have used a model of the exchange through the strait to infer changes in the depth of the Hannish Sill, and hence global sea level, during the last 470,000 years from oxygen isotope measurements from Red Sea cores.