Late Holocene climates of the Near East deduced from Dead Sea level variations and modern regional winter rainfall

The Dead Sea is a terminal lake of one of the largest hydrological systems in the Levant and may thus be viewed as a large rain gauge for the region. Variations of its level are indicative of the climate variations in the region. Here, we present the decadal- to centennial-resolution Holocene lake-level curve of the Dead Sea. Then we determine the regional hydroclimatology that affected level variations. To achieve this goal we compare modern natural lake-level variations and instrumental rainfall records and quantify the hydrology relative to lake-level rise, fall, or stability. To quantify that relationship under natural conditions, rainfall data pre-dating the artificial Dead Sea level drop since the 1960s are used. In this respect, Jerusalem station offers the longest uninterrupted pre-1960s rainfall record and Jerusalem rains serve as an adequate proxy for the Dead Sea headwaters rainfall. Principal component analysis indicates that temporal variations of annual precipitation in all stations in Israel north of the current 200 mm yr⁻¹ average isohyet during 1940–1990 are largely synchronous and in phase (70% of the total variance explained by PC1). This station also represents well northern Jordan and the area all the way to Beirut, Lebanon, especially during extreme drought and wet spells. We (a) determine the modern, and propose the past regional hydrology and Eastern Mediterranean (EM) climatology that affected the severity and length of droughts/wet spells associated with multiyear episodes of Dead Sea level falls/rises and (b) determine that EM cyclone tracks were different in average number and latitude in wet and dry years in Jerusalem. The mean composite sea level pressure and 500-mb height anomalies indicate that the potential causes for wet and dry episodes span the entire EM and are rooted in the larger-scale northern hemisphere atmospheric circulation. We also identified remarkably close association (within radiocarbon resolution) between climatic changes in the Levant, reflected by level changes, and culture shifts in this region.     

U-Th dating of Lake Lisan (late Pleistocene dead sea) aragonite and implications for glacial east Mediterranean climate change

U-series dating can be an effective means to obtain accurate and precise ages on Quaternary carbonates. However, most samples require a correction for U and Th in admixed detritus. This complication is often addressed through generation of U-Th isochrons, requiring analyses of several coeval samples. In addition, presence of water-derived (hydrogenous) Th in the carbonate can cause inaccuracies in isochron ages.This study reports a high-resolution U-series chronology of sediments deposited by Lake Lisan, the last glacial precursor of the Dead Sea. The strategy employed combines multiple measurements from a few stratigraphic heights and fewer analyses from many heights in a single described and measured section. The resulting chronology is based on ages at 22 heights in a ∼40-m-thick section covering the interval of ∼70-14 calendar ka BP. The effects of admixed detritus are evaluated using trace elements. Nearly pure aragonite samples, indicated by very low abundances of insoluble elements such as Nb and Zr, were found to contain hydrogenous Th, which causes the uncorrected U-²³⁰Th age of a modern sample to be ∼2.5 ka. Nevertheless, accurate ages have been obtained by correcting for the detrital and aqueous interferences. The resulting ages are in stratigraphic order, and their accuracy is evidenced by consistency of Lisan Formation U-series and ¹⁴C ages with the coral-based calendar-radiocarbon age calibration.The U-Th ages provide a context to unravel the limnological history of Lake Lisan. Boundaries between the Lower, Middle, and Upper stratigraphic units correspond to the MIS 4/3 and 3/2 transitions, respectively. During MIS 2 and 4 the lake generally showed a stable two-layer configuration and a positive fresh-water balance, reflected by deposition of laminated aragonite-detritus. Dry intervals during MIS 2 and 4 are indicated by thick gypsum layers and an inferred depositional hiatus, which are temporally associated with Heinrich events H1 at ∼17 ka and H6 at ∼65 ka, respectively. During MIS 3 the lake level was unstable with intermittent dry periods indicated by abundant clastic layers and a significant hiatus between ∼43-49 ka. Clastic layers are associated with Dansgaard-Oeschger events during MIS 3, and indicate lake level declines during abrupt Northern Hemisphere warmings. Overall, the climate of the Eastern Mediterranean region shows a strong linkage to the Northern Hemisphere climate, with increasing lake size and stability during cold periods, and fluctuations and dessication during warmings and Heinrich events.     

Sea-rain-lake relation in the Last Glacial East Mediterranean revealed by δO-δC in Lake Lisan aragonites

We investigated the Sea-Rain-Lake relation during the Last Glacial-Holocene in the East Mediterranean region by comparing the δ¹⁸O and δ¹³C records of authigenic aragonite deposited in Lake Lisan, the Dead Sea, Mediterranean foraminifera, and speleothems. The Lisan Formation data display long- and short-term variations of δ¹⁸O, representing steady-state conditions of the lake (e.g., 5.6‰ ± 0.5‰ and 4.5‰ ± 1‰ in the Upper and Lower Members of the Lisan Formation, respectively), and short-term excursions reflecting large floods and droughts. The long-term (steady-state) δ¹⁸O values of the Lisan aragonites show similarity to the corresponding time-equivalent records of the Eastern Mediterranean foraminifera and Judea Mountain speleothems: The Last Glacial deposits are in all of them 2‰-3‰ heavier than the Holocene ones. We interpret this similarity as reflecting the significance of the source effect on the long-term behavior of isotopic reservoirs: Speleothem δ¹⁸O is strongly influenced by the marine reservoir that contributes its vapor to rain formation; the lake δ¹⁸O is dominated by the composition of the inflowing water. Short-term variations in the isotopic composition of rainfall are dominated by the amount effect and the temperature and those of the Lake’s upper water mass by the lake’s water balance.δ¹³C values are more variable than δ¹⁸O in the same Lisan sequences (e.g., δ¹³C in the Lower Member is 1.0‰ ± 1.7‰, whereas δ¹⁸O is 4.6‰ ± 0.7‰) and are 1‰ to 1.5‰ higher in the Upper Member than in the Lower and Middle Members of the Lisan Formation. These variations reflect significant increase in primary productivity of the lake and algal bloom activity. It appears that the hypersaline-saline lakes were not as “dead” as the Dead Sea is and that algal activity had an important impact upon the geochemistry of Lake Lisan.The δ¹⁸O data combined with independent geochemical and limnologic information (e.g., level fluctuations) indicate that Lisan time was characterized by high precipitation-high lake stands-high atmospheric humidity, whereas the Holocene Dead Sea shows the opposite behavior. This paleoclimatic reconstruction is consistent with independent evidence for significantly wetter conditions in the East Mediterranean region during the Last Glacial period.     

Flow dynamics,sedimentation and erosion of glacial lake outburst floods along the Middle Pleistocene Scandinavian Ice Sheet (northern central Europe)

During the Middle Pleistocene late Saalian glaciation of northern central Europe numerous pro‐glacial lakes formed along the southwestern margin of the Scandinavian Ice Sheet. Little is known about the drainage history of these lakes, the pathways of glacial lake outburst floods and their impacts on erosion, sedimentation and landscape evolution. This study investigated the impact of the late Saalian Weser and Münsterland Lake (Germany) outburst floods. In particular, we reconstructed the routing and flow dynamics of the lake outburst flood and analysed the flood related sediments. We employed one‐dimensional hydraulic modelling to calculate glacial lake outburst flood hydrographs. We modelled the flow pathway and local flow conditions along the pathway based on the boundary conditions of two different hydrographs and two different ice‐margin positions. The modelling results were compared with geomorphological and sedimentological field data in order to estimate the magnitude and impact of the flood on erosion and sedimentation. Two major lake drainage events are reconstructed for the study area, during which approximately 90–50 km³ of water was released. Modelling results indicate that the lake outburst floods created a high‐energy flood wave with a height of 35–50 m in confined valley areas that rapidly spread out into the Lower Rhine Embayment eventually flowing into the North Sea basin. The sedimentary record of the outburst floods comprises poorly sorted coarse‐grained gravel bars, long‐wavelength bedforms and sandy bedforms deposited by supercritical and subcritical flows. Some parts of the sandy flood deposits are rich in reworked mammoth bones or mammoth and horse teeth, pointing to reworking of older fluvial sediments, hydraulic concentration and subsequent re‐sedimentation of vertebrate remains. These deposits are preserved in sheltered areas or at high elevations, well above the influence of postglacial fluvial erosion. The flood‐related erosional features include up to 80‐m‐deep scour pools, alluvial channels and streamlined hills.