Pleistocene cyanobacterial mounds in the Perachora Peninsula, Gulf of Corinth, Greece: structure and applications to interpreting sea-level history and terrace sequences in an unstable tectonic setting

Large Pleistocene cyanobacterial mounds in the Perachora Peninsula, Gulf of Corinth, Greece, occur on a cut terrace and are interpreted to have formed in freshwaters on the basis of the principal constructing element, a freshwater cyanobacterium Rivularia haematites [Richter et al. (1979) Neues Jahrb. Geol. Paläontol. 159, 14–40], which pre-dates all other biota in and on the mounds. However, because of corals dated at ca. 125 ka in marine debris engulfing the mounds, the cut terrace has also been considered to be the erosional component of the Tyrrhenian Stage 5 highstand, an episode when the Gulf of Corinth was flooded by marine water from the Mediterranean Sea via the Rio Strait. This contradiction is addressed by consideration of mound ecology. Mounds are up to 10 m high, and several tens of metres wide, with simple dome to bulbous morphology, commonly coalesced with neighbouring mounds. Mounds consist of accreted layers of a branched structure of R. haematites, and encrusted by coralline alga Lithophyllum pustulatum, which can tolerate brackish water. No examples of the alga encrusted by Rivularia were found. The remaining space is filled with clotted microbial structure, and layers of micrite containing peloids, shell fragments and planktonic foraminifera. The mounds formed submerged in water, indicated by their accreted structure and rounded morphology. The terrace and mounds are bounded by steeply dipping coastal faults, with several hundred metres of water depth on neighbouring subsided hangingwall blocks. Thus the mound field grew on a triangular promontory protruding into the eastern end of the gulf, surrounded on two sides by deep water. The mounds are here interpreted to have grown when the Gulf of Corinth was a fresh to brackish lake, when Mediterranean sea level was below the contemporaneous depth of the threshold at the Rio Strait. Therefore the cut terrace on which the mounds occur cannot be the Tyrrhenian highstand. Relative sea-level rise permitted seawater flooding, to allow, firstly, colonisation by brackish water-tolerant L. pustulatum, followed by fully marine Tyrrhenian highstand biota, thus indicating that the mounds grew in a glacial phase, probably Stage 6, prior to marine Tyrrhenian deposits. This work has important implications for the relative dating of cut terraces and the deposits on them. Therefore: (1) not all cut terraces in coastal settings are necessarily the result of sea-level rise; and (2) terrace deposits are not necessarily always genetically related to the cut surface beneath them.