engleski

Sea level change during MIS 5a recorded in submerged speleothems from the Eastern Adriatic Sea (Croatia)

INTRODUCTION In order to reconstruct relative sea-level changes on the Eastern Adriatic coast, 16 speleothems were collected from submerged caves and pits along the Croatian coast. Their growth provides an unambiguous indicator of former low sea levels and can be used to constrain past sea level elevation and timing. Two of them, taken from the U Vode Pit on the Krk Island, apparently recorded Late Pleistocene transgression known as marine isotope stage (MIS) 5a that occurred at ~ 80 ka. Similar research, for example, has been undertaken on submerged speleothems from the Tyrrhenian Sea (Bard et al., 2002), from Bahamas (Richards et al., 1994, Lundberg & Ford, 1994), from Balearic Islands (Fornós et al., 2002) and from the Eastern Adriatic Sea (Vrhovec et al., 2001, Surić et al., 2005). The Adriatic Sea is a semi-enclosed epicontinental basin situated between the Apennines and Dinaric mountain ranges, characterized by a relatively shallow northern part (0 to -120 m) with low gradient (0.02°). Such morphology makes it very sensitive to the eustatic changes, so after the Last Glacial Maximum (LGM), postglacial sea- level rise generated eight-fold widening of the shelf area (Correggiari et al., 1996 ; Cattaneo et al., 2003). Much of submarine Eastern Adriatic and Croatia is karstic, developed mostly in Cretaceous and Palaeogene limestones. During the global sea-level low stand associated with the LGM (-121 ± ; ; 5 m, Fairbanks, 1989), much of the coastal submarine karst (caves, pits with speleothem deposits) would have been above sea level. More than one hundred submarine caves and pits have been discovered and most of them have been explored, but comparing with adjacent coast and islands, their density is probably several times higher. Currently, the deepest speleothems found within them are those from the submarine of Ist Island from the depth of 60 m (Vrhovec et al., 2001). Krk Island is the second largest Croatian island and situated in the northern part of the Adriatic Sea (Fig. 1). Two stalagmites from the depths of 14.5 m (K-14) and 18.8 m (K-18) were collected from the U Vode Pit situated in the southern part of the island. Pit is formed in the Upper Cretaceous limestone and it is not completely submarine feature. The entrance of the pit is on the coast, at the elevation of 5.5 m, and the bottom is at -24 m. Relatively thin bordering bedrock allows free circulation between the pit and open sea, as well as settling of marine organisms (serpulids) within the pit. Only upper 0.5 m is filled with brackish water. METHODS AND RESULTS Stalagmites K-14 (26 cm long) and K-18 (15 cm) were taken from their growth position and elevation recorded with digital depth gauge. Each sample was cut along the growth axis and polished to enable insight into the growth layers. Sub-samples (230-415 mg) for MC-ICPMS U-Th measurements were drilled from the polished surface to constrain the timing and duration of what appeared to be hiatuses in deposition – recognised by changes in crystallography, morphology and colour of speleothem calcite and a thin red sediment layer (arrows in Fig 2 and 3). After several days of being exposed on the air, a white substance also appeared on the discontinuities that contain the red layer. The nature of this material was analysed using X-ray diffraction measurements directly on speleothems. U-Th: The age range of sub- samples is ~94 to 54 ka (Fig 2 and 3). Uranium concentrations are low (30 - 70 ng g-1) and errors are typically 1-6% (2 ), depending on the extent of detrital contamination. Combining the data for both samples, three phases of growth are observed >90 to 80 ka, a short period at ~80 ka, and ~65 to ~54 ka. We consider the initiation of the extended hiatus in K-18 (and perhaps also that in K-14) to be related to the sea level high event correlated with marine isotope stage 5a. In addition to changes in the morphology, crystallography and colour of calcite between the growth phases, initial 234U/238U also shows abrupt shifts indicating substantial differences between the hydrological regime for each phase of growth. XRD: Qualitative X-ray diffraction measurements were done by PANalytical X'pert Pro thete- theta diffractometer equipped with multilayer parabolic monochromator using CuKα radiation. Measurements were performed directly on speleothems. On both speleothems, in the region marked by red line, halite and gypsum peaks were recorded in addition to calcite. No evidence for gypsum was observed for in situ analysis of the speleothem growth just above and below the hiatuses, where only small peaks for aragonite and halite were found in addition to calcite. DISCUSSION AND CONCLUSION According to the results obtained by radiometric measurements and X-ray diffraction, we can assume that deposition of speleothems K-14 and K-18 had been more or less continuous in subaerial conditions from 94 to 80 ka and 65 to 54 ka (and perhaps another short period of growth at ~ 80 ka. Sea level must have been lower that -14 to -18 m at these times. Minerals recorded within hiatuses (calcite, gypsum, halite) are common cave minerals, but this assemblage also coincides with the suite of minerals that precipitate due to the evaporation of seawater (Ca-carbonates, gypsum, anhydrite, halite, K-Mg chlorides, arranged from less to the most soluble) (Seibold & Berger, 1996). Halite could, presumably, crystallize from the sea water that penetrated through the porous parts of speleothems due to its high concentration in the sea water, but it is unlikely that gypsum precipitated from the same water. It is more likely that it has been precipitated in situ at a time of exposure to marine conditions in sea regression period. In accordance with obtained U-Th ages, this sea-level oscillation can be attributed to MIS 5a interstadial that was marked with two distinct sea-level highstand at ~84 and ~77 ka (Potter & Lambeck, 2003). Growth in the speleothem K-18 ceased from 79 to 65 ka, while the shallower speleothem, at -14 m, experienced shorter period of submerging ~80 ka, so we can presume that the highest MIS 5a relative sea level in this region was slightly above -14 m.

submerged speleothems ; U-Th dating ; sea-level change ; MIS 5a ; Adriatic Sea ; Croatia

nije evidentirano