Cancellation of the precessional cycle in δ18O records during the Early
Pleistocene
Abstract
Deep-sea δ18O records show a pronounced difference in Milankovitch
periodicity between the Early and Late Pleistocene. δ18O is interpreted
as a proxy for ice sheet volume and temperature, which led to the
conclusion that glacial-interglacial cycles considerably changed their
rhythm during the mid-Pleistocene. This transition is referred to as the
mid-Pleistocene Transition (MPT). Specifically, the precessional
component of the Milankovitch cycles is absent in Early Pleistocene δ18O
records, despite its continuous presence in solar insolation forcing to
the ice sheets. Climate feedbacks involving (sea) ice, geological
processes and carbon and nutrient cycling have been proposed as causes
of this marked change. We however show that the absence of an Early
Pleistocene precession signal in deep-sea δ18O records could be the
result of destructive interference of the precessional cycle in the
interior ocean. Such cancellation is caused by the anti-phasing of the
precessional cycle between the North Atlantic and Southern Ocean
deep-water sources (see Figure). We explore the potential for
cancellation in the transient setup of the Total Matrix Intercomparison
model for a wide range of source signal strengths. Our results show that
cancellation can cause the absence of the precessional signal due to
cancellation in the interior South-Atlantic, Indian and Pacific basins.
Cancellation is especially widespread for a relative end-member
contribution typical for the Early Pleistocene. We therefore conclude
that the precessional component is likely incompletely archived in Early
Pleistocene δ18O records, and appears as an actual change in
Milankovitch periodicity across the MPT. Proxies not susceptible to
cancellation of precession (such as those currently retrieved across the
MPT from Antarctica) would be able to verify to what extent deep-sea
δ18O correctly represents Pleistocene climate.