Abstract
In closed-basin lakes, sediment porewater salinity can potentially be used as a conservative tracer to reconstruct past fluctuations in lake level. However, until now, porewater salinity profiles did not allow quantitative estimates of past lake-level changes because, in contrast to the oceans, significant salinity changes (e.g., local concentration minima and maxima) had never been observed in lacustrine sediments. Here we show that the salinity measured in the sediment pore water of Lake Van (Turkey) allows straightforward reconstruction of two major transgressions and a major regression that occurred during the last 250 ka. We observed strong changes in the vertical salinity profiles of the pore water of the uppermost 100 m of the sediments in Lake Van. As the salinity balance of Lake Van is almost at steady-state, these salinity changes indicate major lake-level changes in the past. In line with previous studies on lake terraces and with seismic and sedimentological surveys, we identify two major transgressions of up to +105 m with respect to the current lake level at about 135 ka BP and 248 ka BP starting at the onset of the two previous interglacials (MIS5e and MIS7), and a major regression of about −200 m at about 30 ka BP during the last ice age.
Highlights
Past lake-level changes with respect to the present lake level of 1645 m above sea level (a.s.l.) have been reported in studies of lake terraces on land[8,9,10,11]; in seismic[12, 13], volcanological[14], and sedimentological[15] studies; and in studies of the geochemistry of the uppermost 10 m of the sediment column[3, 16, 17]
Earlier lake-level reconstructions by geochemical means in short sediment cores suggest that a maximum regression of more than −400 m from the present lake level occurred at about 15 ka BP, with a possible desiccation period[3, 20]
To gain new insights into the hydrology of Lake Van, we present a reconstruction of the past major fluctuations in the lake volume and the corresponding lake level using the salinity measured in the pore water of core catcher samples collected at Ahlat Ridge (38.667°N, 42.670°E, water depth 357 m; black star in Fig. 1, right-hand panel) during drilling operations conducted as a part of the ICDP (International Continental Scientific Drilling Program, www.icdp-online.org) PaleoVan project[21,22,23] (Fig. 2)
Summary
Past lake-level changes with respect to the present lake level of 1645 m above sea level (a.s.l.) have been reported in studies of lake terraces on land[8,9,10,11]; in seismic[12, 13], volcanological[14], and sedimentological[15] studies; and in studies of the geochemistry of the uppermost 10 m of the sediment column[3, 16, 17]. Seismic transects across the basin of Lake Van reveal the presence of clinoforms and erosive channels, suggesting past low lake levels several hundred meters below the present lake level. The timing of the inferred regressions over the past 80 ka contrasts somewhat with the transgressions suggested by the age estimates of the terraces on land[9], limiting any final assessment of past water volumes (or lake levels), and of the major hydrological states of Lake Van. Earlier lake-level reconstructions by geochemical means in short sediment cores suggest that a maximum regression of more than −400 m from the present lake level occurred at about 15 ka BP (corresponding to a sediment depth of about 10 m, according to the age model adopted in the relevant publications), with a possible (but highly speculative) desiccation period[3, 20]. No indicators of desiccation, such as evaporites (e.g., aragonite crusts, dolomite) or iron oxides have been observed anywhere in the sedimentary sequence of Lake Van[6, 7, 15, 17] and conformable strata, in the deep basin, as seen in seismic data[6, 12] show no evidence that the lake was ever completely dry
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
