Productivity cycles in Lake Hovsgol, NW Mongolia, during the last 1 Ma and the age model of the HDP-04 drill core record

Repetitive patterns in the records of total organic carbon (TOC), total nitrogen (TN) and δ13Corg observed in the Lake Hovsgol sediment section from HDP-04 drill core reflect past changes in productivity of Lake Hovsgol and in the isotopic composition of the lake's carbon pool. Lake Hovsgol productivity proxy signals are interpreted to represent the response of the Hovsgol lacustrine system to glacial–interglacial cycles of the Pleistocene. This interpretation is supported by the apparent orbitally-forced pattern in the TOC, TN and δ13Corg records of the past 250 ka in the BDP-96-2 drill core from neighboring Lake Baikal.The intervals with independent age control, such as the radiocarbon-dated last glacial–interglacial transition and the paleomagnetic reversals, make it evident that productivity proxy signals are reliable indicators of past cold-to-warm and warm-to-cold climate transitions, as seen from the agreement with the pattern of global climate change in marine δ18O records. The Brunhes/Matuyama reversal during the MIS 19 interglacial coincides with a distinct peak of TOC and TN in the Hovsgol record, similar to the signal during the Holocene interglacial. By contrast, the upper Jaramillo reversal in the Lake Hovsgol record occurs in a diatom-free calcareous interval characterized by minima in TOC, TN and by a ‘glacial’-type range of δ13Corg values. In both Lake Baikal and Lake Hovsgol records, peaks in TOC and TN contents help distinguishing past interglacials and interstadials, and isotopically-heaviest δ13Corg values help identify past glacial intervals.An age model for the HDP-04 drill core section is proposed based on recognizing the repetitive patterns in Lake Hovsgol productivity and lithologic records as regional paleoclimate cycles of middle to late Pleistocene. Absolute dates and diatom biostratigraphic correlation ties to the Lake Baikal record are used as key controls. In the proposed age model, the interval 81–24 m in the HDP-04 sediment section below the major unconformity is correlated to MIS 27 through late MIS 13, whereas the upper 24 m of the HDP-04 section is suggested to have recovered the sedimentary record of late MIS 7 to MIS 1.