Pages

2011-10-28

The Messinian Salinity Crisis (1/3) - Salt all over the Mediterranean


[Spanish version of this here]
[This is a science-diffusion article, an introduction to the MSC. Details on our research will appear here upon publication]

Pliny the Elder begun his Historia Naturalis describing the geography known to the Romans. His account starts inevitably at the narrow entrance to the Mediterranean Sea, echoing local myths on the origin of the two sharp mountains flanking the Gibraltar Strait:
"(...) the inhabitants have called them the Columns of Hercules; they believe that they were dug through by him; upon which the sea, which was before excluded, gained admission, and so changed the face of nature."
Pliny knew very little about the formation of the Earth: he simply lacked the knowledge we have accumulated since the 18th century to understand the processes shaping landscape. However, he understood that the Earth is not static but it changes as a result of dynamic processes. And he did know the widespread salt outcrops present along most of the Mediterranean coast: The Romans used profusely the gypsum salt crystals as window glasses, for example. Today, we can only imagine what the origin of Pliny's account was, but science is giving us strong evidence that the Mediterranean was once indeed largely desiccated and that it was later refilled in a catastrophic event, just as those native Iberian peoples believed.

In 1867, paleontologist Karl Mayer-Eymar realized that those gypsum outcrops had all a similar geological age, and named that period Messinian after the impressive salt mines near the Sicilian town of Messina. In 1954, that period of Mediterranean-wide salt deposition was named the Messinian Salinity Crisis by the italian geologist Raimondo Selli, and is now dated back to 5 and 6 million years ago, the time when the first hominids walked in Africa.
The whole mountain in this picture is made out of gypsum
salt crystals deposited during the Messinian Salinity Crisis.
The unsettling house too. The blocks are fallen due to the steep
valley excavated in the gypsum by the river that runs between the
mountain and the photographer. Near Sorbas (Almería, España)
(Photo: D.G-C). 
During the Messinian times, as today, the Mediterranean was also a deficitary area. I'm not talking about economics here, but about a hydrological deficit: the amount of water delivered by the rivers and by rainfall to the Med is not enough to compensate its water looses by evaporation. Thanks to the connection at the Strait of Gibraltar, this deficit is compensated by a net inflow of 70,000 m3/s of Atlantic water, a flow equivalent to about 40 times the Niagara Falls. The Atlantic is thus constantly refilling the Mediterranean net evaporation. This current is very well known to sailors because it easily brings your ship to a speed of several meters per second in one of the busiest crossroads in the world. Less known is the outflow current underneath, bringing deep, dense, saline water from the Med to the Atlantic. Both were widely used during WWII, by submarines aiming at crossing the strait silently (great film, Das Boot).

Present inflow and outflow from the Atlantic across the Gibraltar
Strait. The outgoing current is denser and runs underneath the inflow. 
Both were used by the submarines during World War II to sneak
in and out of the Med silently (Prenhall/Pearson).
Important: note that to increase salinity in the Med you only need to reduce the outflow, whereas in order to let it go dry, you need to stop also the inflow. For this reason, the presence of salt rocks of Messinian age all over the Med does not mean it went dry. Artificial salt pans, for example, evaporate sea water keeping the brine at a constant level, maintained by further inflow of sea water. In 1849, italian chemist J. Usiglio experimented with this process and observed that, although the sea salt contains ten times more halite than gypsum, it is gypsum that precipitates first due to its lesser solubility. This could explain the abundance of gypsum onshore the Mediterranean, in contrast with the relative scarcity of onshore halite.
Present water salinity at the surface of the ocean, showing a
saltier Mediterranean caused by the higher evaporation 
and isolation of its waters. Source: World Ocean Atlas 2005

But what happened then offshore? If the salts were related to a pan-Mediterranean event, salt should also have accumulated in the deeper parts of the sea. Only in the 1960s, studies of the reflection of seismic waves (echoes of vibrations sent from a boat) started to show ubiquitous evidence of a layer a few hundred meters below the sea floor. It was named the 'M reflector', and extended laterally up to approximately the 1500 m depth contour of the present sea, suggesting that one single process was responsible for it. Today we know that below the M reflector about 10% of the salt contained in the entire global ocean is trapped since the Messinian times.

In the 70's, 3000-m-deep drillings carried out on board the Glomar Challenger proved the presence of  salt deposits offshore Mallorca, demonstrating that the classical onshore outcrops had an equivalent in the open sea. It also found anhydrites and pebbles, suggesting that the Mediterranean consisted of a series of brackish lakes, possible remnants of a desiccated Med. But these were just weak circumstantial evidence, and both could be (and were indeed) argued.
Distribution of Messinian salts in the Mediterranean 
(Ryan, 2008)

All the research published was supporting the existence of a great evaporitic basin affecting both the shallow marine basins at the margins of the Mediterranean and the deeper inner parts of the sea. But still this does not imply a desiccation or a large drawdown of the Med, as discussed above. 


Section across the Nile in Aswan (Egypt) by Chumakov (1967), 
based on wells. It shows a valley excavated by the 
river during the MSC, now filled with later sediments.
The main evidence really supporting a desiccation and a large level fall of the Mediterranean arrived actually somewhat earlier, back in the late 1950s. The building of the Aswan Dam (1,200 km upstream from Alexandria) became disturbed by the discovery of a deep narrow gorge excavated in granite, hundreds of metres below sea level (Chumakov, 1967). This gorge has been since filled by loose sediments that made difficult the construction of the dam. It is now known that the gorge follows and deepens downstream the Nile, reaching more than 2000 m deep below Cairo. The Rhone, Ebro and Po rivers also have similar gorges buried below their present deltas, filled with post-Messinian sediment. In contrast with other canyons formed today in the Atlantic or the Pacific (linked to turbidity currents originated underwater at the continental shelf), the Messinian ones along the Mediterranean coast look like drowned river valleys (Loget et al., 2006, Urgeles et al., 2010).


Valley excavated during the Messinian salinity crisis at the
mouth of the Ebro River, as derived from recent seismic
reflection images. Scientific paper here
Today, the progress of the MSC is still a matter of debate, and the occurrence of a large drop of the Mediterranean level is not yet fully agreed. What are the arguments in favor and against this possible Mediterranean desiccation? And what were the processes responsible for the crisis? In an upcoming post I will detail on this, and in a 3rd one I will focus on the outcomes of our own research that will soon appear in Nature. [update: link; pdf here]

[Next chapter here]

References:

  • Pliny's Historia Naturalis full text.
  • More science diffusion on the MSC in this article and in David Bressan's blog.
  • The scientific paper will be linked here upon publication.
  • Loget, N., Van Den Driessche, J. On the origin of the Strait of Gibraltar. Sedim. Geol. 188–189, 341–356 (2006).
  • Chumakov, I. S. (1973), Pliocene and Pleistocene deposits of the Nile valley in Nubia and upper Egypt, Initial Rep. Deep Sea Drill. Proj., 13, 1242-1243.
  • Ryan, W. B. F., Decoding the Mediterranean salinity crisis. Sedimentology 56, 95-136 (2008). doi: 10.1111/j.1365-3091.2008.01031.x