The original 1970s models of the connection between the Mediterranean Sea and the Black Sea considered that the bedrock sill through the Bosporus Strait, over which water could flow, was shallow. While the outlet was active, the Black Sea followed the rises and falls of global sea-level. Whenever global sea level fell below the Bosporus outlet, it was assumed that the Black Sea stabilized at the outlet level, as any water loss was replenished by in-draining rivers. During the 1980s, researchers found shorelines below the present Black Sea at about -80 m and suggested that a deeper Bosporus sill would account for this and still allow Black Sea levels to synchronize with global sea levels. Complications arose. First, if the sill were -80 m deep, then the Black Sea would have reconnected to the Mediterranean
Sea about 11,000 years ago, but there is no evidence of saltwater in the Black Sea until 8,400 years ago. Second, other ancient submerged shorelines were discovered, some as deep as -155 m. Subsequent fieldwork revealed an entire submerged landscape with buried channels, buried estuaries, buried coastal lagoons, 'marine' terraces, and coastal dunes and barrier islands.
Compelling, though debated, evidence suggested a catastrophic saltwater flood in the Black Sea 8,400 years ago that rapidly drowned the early Holocene landscape (Ryan et al. 1997; Ryan and Pitman 1999; Ryan et al. 2003). The evidence comes from seabed mapping, very high-resolution seismic-reflection profiles, and seabed sampling using cores. It indicates that, over the last 2 million years, the Black Sea was mainly a giant freshwater lake, lying up to 100 m below its outlet, which suffered occasional saltwater incursions through the narrow outlet of the Bosporus when global sea levels stood higher. When the Mediterranean lay below Bosporus sill, the Black Sea (technically a lake under those conditions), like its neighbouring Caspian Sea, operated in two modes. Under cold climates, it was an expanded lake; under warm climates, it was shrunken lake. Therefore, during glacial stages of the Quaternary, the expanded lake of the Black Sea spilt into the Marmara Sea and thence into the Mediterranean. During warm stages, once a large volume of meltwa-ter had been discharged, the lake shrank to the outer shelf, and sometimes even beyond the outer shelf. Now, if the Black Sea lake fell while the Mediterranean rose, then it would be possible for a catastrophic flood to occur once the ocean reached the Bosporus sill. Such a flood would submerge more than 100,000 km2 of its exposed continental shelf. It is possible that this permanent drowning of a huge terrestrial landscape accelerated the dispersal of early Neolithic foragers and farmers into the interior of Europe at that time (Ryan and Pitman 1999).
It seems irrefutable that a saltwater inundation submerged an extensive ancient landscape. The crucial question is how fast the flooding occurred. A careful picture of events during postglacial times suggests two transgressions, the younger of which was rapid (Figure 5.1). About 15,000 years ago, the Caspian and Black Sea were brimful with glacial meltwater (Figure 5.1(a)). The Caspian overflowed into the Black Sea, and the Black Sea overflowed into the Mediterranean. From about 13,400 to 11,000 years ago, evaporation under an arid climate reduced the water level to the older shoreline, with wave action and exposure creating an unconformity in the sediments (Figure 5.1(b)). By 11,000 years ago, the regression reached lowstand of -105 m (Figure 5.1(c)). For the next 1,000 years, which corresponds to the Younger Dryas climate reversal, cooling tipped the precipitation-evaporation balance in favour of water surplus in the Caspian and Black Seas. The Caspian filled to its spillway and discharged Stephanodiscus astrea diatoms into the Black Sea. The Black Sea lake rose, draping a layer of Dreissena (a freshwater mollusc) coquina up to the -30 m isobath (Figure 5.1(d)). During this time, the global ocean lay 20 m or more below the Black Sea spillway. Evidence suggests that the onset of the Younger Dryas was rapid, occur-
Figure 5.1 Schematic of the development of the Black Sea sedimentary sequence deposited over the last glacial-interglacial cycle. (a) Delta formation at freshwater highstand (spillpoint?). (b) Forced regression with falling Black Sea level. (c) Bevelling and channel cutting, with formation of the lower shoreline deposit at around -105 m. Shallow water erosion features form seaward of the shoreline. (d) Onlap of Unit L4 (Dreissena coquina) and fill of the mid-shelf channels. (e) Sea level fall to winnow Dreissena coquina on the mid-shelf (and seaward of the shoreline by shallow-water processes) and formation of the upper shoreline deposit at around -80 m. (f) Highstand marine deposition. Source: After Major (2002).
ring in less than one hundred years, and the rise of the Black Sea lake was gradual enough for coastal onlap to occur. From 10,000 to 8,500 years ago, corresponding to the B0lling-Aller0d, renewed aridity and evaporation shrank the Black Sea down to its upper shoreline, with the exposed Dreissena coquina (a coarse-grained, porous, friable variety of clastic limestone made up chiefly of fragments of shells) being eroded into shelly sand and gravel (Figure 5.1(e)). Coastal dunes formed at this time, the sand coming from beaches, the coquina, and dry riverbeds. Around 8,400 years ago, the Mediterranean connected with the Black Sea, triggering the terminal transgression that led to the modern 'snapshot' depicted in Figure 5.1(f), in which water level rose from -95 m to -30 m. Sediments laid down between 8,400 and 7,100 years ago show a transition from brackish to marine conditions. This younger transgression occurred swiftly, with coastal dunes dated at 8,500 years old drowned by a century later. The fauna in sediment cores from -50 to -90 m reveal an early change to saline conditions. Three chief arguments favour this being a catastrophic flood (Ryan et al. 2003, 547). First, the coastal dunes are in excellent preservation. Second, there is an absence of onlap in the brackish to marine mud drape above unconformity 1b. Third, pollen spectra suggest that the Black Sea's climate was arid until after the initial change to salty conditions.
Criticisms of the catastrophic terminal flooding of the Black Sea rest on three arguments (Aksu et al. 2002a, 2002b; Hiscott et al. 2002). First, a sapropel deposit in the Marmara Sea would probably have needed a lens of freshwater provided by outflow from the Black Sea to form. According to Ryan et al.'s (2003) observations, the Black Sea does seem to have flowed out during the Younger Dryas, which would correspond to the initiation of the Marmara Sea sapropel some 10,600 year ago. Second, a subaqueous delta lies in the Marmara Sea south of the Bosporus Strait, presumably built from sediments borne from the Bosporus valley during persistent Black Sea outflow (Hiscott et al. 2002). However, the dating of this delta complex is not yet conclusive and may not invalidate the rapid flood hypothesis. It simply shows that delta building was still active 10,200 years ago during the Younger Dryas outflow (Ryan et al. 2003). Third, some researchers claim, on the basis of extrapolated core dates, that lowstand shelf-edge deltas in the Black Sea were flooded by 12,000-11,000 years ago (Aksu et al. 2002a). This extrapolation, based on sedimentation rates in the marine cover is speculative but does not fit with the good evidence that the Black Sea shoreline between 10,000 to 8,500 years ago sat well below the Bosporus sill, which would have not have permitted any outflow at that time. Persistent Holocene outflow would require a deep sill, but a deep sill would not allow the transgression across the Black Sea shelf during the Younger Dryas to the -30 m isobath.
Was this article helpful?