Abstract:
Marine Isotope Stage 11 (MIS 11) is known to be the closest analogue to the Holocene (MIS 1) in terms of Earth’s orbital configuration during the last 400,000 years. This has often been used as an argument to investigate climate trends of MIS 11, reconstructed from natural geological archives in order to evaluate current and future climate developments. However, the orbital configuration during MIS 11 was not exactly identical to that of MIS 1 and the comparability of climate trends during both interglacials is currently hotly debated. In this study, the first high resolution palaeoclimatic investigation of MIS 11 is presented here for the Eastern Mediterranean Sea, known to yield an especially sensitive climate archive. High-resolution, continuous multi-proxy palaeoclimatic data have been generated from two sediment cores of this region (ODP Site 964: 36°16’N, 17°45’E, 3658 m; GeoTü-SL96 32°46’N, 19°12’E, 1399 m). The records have been tied to an absolute time scale using an age model based on stable oxygen isotopes, planktonic faunal abundance events and sapropel formations. As a proxy of surface water conditions, assemblage compositions of planktonic foraminifera were determined in 404 samples and supplemented with alkenone unsaturation ratios in one of the cores. The new MIS 11 data indicate that MIS 11 sapropel formation and onset of interglacial conditions in both cores correlated with the second insolation peak after Termination V. In contrast, the MIS 1 sapropel S1 coincides with the first insolation peak after Termination I. In contrast, both cores show a significant sapropel layer in MIS 1, indicating a different depth for the oxycline during MIS 11. In addition, the size of the oxygen isotopic peak at ODP Site 964 is larger than in the Holocene at the same site, indicating enhanced monsoonal activity and/or Black Sea discharge during MIS 11, despite a weaker insolation forcing.
The alignment of MIS 11 sapropel to the second insolation maximum of MIS 11 reveals an apparently delayed response of the pelagic system to deglaciation. Benthic foraminifera concentrations and Ba/Ca ratios in the sediment furthermore suggest high productivity for the pre-sapropel interval. The most likely scenario to explain the apparent delayed response, in stark contrast with global trends and pollen evidence from the Mediterranean region, is high winter productivity during relatively cool and wet interglacial conditions in early MIS 11, fuelled by enhanced influx of terrigenous material, leading to the persistence of glacial planktonic foraminifera assemblages and alkenones representing the cold season signal until the development of the MIS 11 sapropel. Within the later part of MIS 11, planktonic foraminifera assemblage compositions show three conspicuous phases. The assemblages occurring during these phases resemble faunas known from later sapropels. As they coincide with insolation maxima and low Ba/Ca and Fe/Al values in the sediment, they appear to be controlled by orbitally driven maxima in seasonality and stratification with low productivity and thus no sapropel formation. In addition, the planktonic foraminifera assemblages throughout late MIS 11 remain dominated by the warm-water indicator G. ruber, reaching an absolute maximum in late MIS 11 and the first half of MIS 10 (~ 80 %). This pattern can be explained by unusually warm temperatures during the late phase of MIS 11, which is consistent with alkenone data, combined with extremely low productivity, as indicated by benthic foraminifera concentrations. During sample preparation, anomalies in shape and size of G. ruber were identified, which led to the definition of four different morphotypes within this species, in accordance with literature. Abundances of the morphotypes significantly change between glacials and interglacials, and in three of the morphotypes of G. ruber, significant offsets in stable isotope composition were found. Since the isotope shifts among the three G. ruber morphotypes are systematic and often exceed 1 ‰, their understanding is essential for the interpretation of all G. ruber – based proxy records for palaeoceanographic reconstructions. The apparent delayed reaction of MIS 11 pelagic ecosystem in the Mediterranean region to global trends, resulting in a sapropel formation coinciding with the second insolation peak of the interglacial period, makes it difficult to align MIS 11 and MIS 1 with respect to the deglaciation. The degree of orbital analogy between MIS 11 and MIS 1 was clearly insufficient to force analogous climatic trends during the two interglacials that could be used to predict the future development of the Holocene without human impact. The alignment of the two interglacial periods based on sapropel formation as proposed here identifies a significant asymmetry between the two interglacials and highlights the anomalous length of MIS 11 as not necessarily representative of the course of the current interglacial.