No abrupt changes in CF4 emissions by granite weathering and erosion over the Mid-Pleistocene Transition

The climate evolution over the Pleistocene (last 2.6 Myr) is characterised by a sequence of cold glacials that are interrupted by warmer interglacial phases. These glacial-interglacial cycles are expressed in changes in global ice volume, ocean temperature, and other parameters that serve as proxies to infer land and ocean processes or to provide information about radiative forcing changes.

The Mid-Pleistocene Transition (MPT; 1.2-0.9 Myr) is characterised by the transition from 41 kyr cycles to about 100 kyr cycles and likely a trend toward colder glacial periods. The Mid Brunhes (MB) marks a climate transition at about 450 kyr after which the interglacial intensity (e.g. temperatures and greenhouse gas concentrations) increased. Consolidated explanations of the causes of both the MPT and the MB are lacking. Internal climate dynamics or feedback mechanisms are required since frequencies and the power of orbital parameters did not systematically change over the last 1.2 Myr.

For the MPT, two main classes of explanations have been put forward to account for the advent of the 100-kyr cycles. One argues that the ice sheets grew larger because the glacial climate became colder, driven by a long-term decline in glacial CO₂ (GHG forcing). The other argues that NH ice sheets survived the next potential termination because land-surface properties (e.g. regolith, routing of ice sheets) made them less sensitive to meltdown.  

Here, we present a record of CF₄ measured on ice core samples from the EPICA and Beyond EPICA ice cores, as well as some even older samples from the Alan Hills blue ice area that may shed additional light on the potential reasons for the MB and MPT. CF₄ is a natural gas with a very long lifetime (on the order of 100 kyr) occurring predominantly in granitic rocks and other acidic plutonites and is released during chemical weathering and erosion of these rocks. Granites are globally distributed but have a bias towards high northern latitudes (Laurentide region, Scandinavia). Accordingly, some connection of CF₄ release to ice sheet extent is to be expected. Over the Pleistocene, the surfaces of these high northern areas have been intensively eroded, and we would expect that a removal of this regolith layer would have left a sizable imprint in our CF₄ record. However, our CF₄ record shows a rather gradual decline from 2.6 Myr BP toward its local minimum reached shortly before the MB. In contrast, our CF₄ record is consistent with the view that the CF₄ release is most sensitive to chemical weathering and thus to temperature and the hydrologic regime. While we observe moderate changes in the average long-term CF₄ emission flux (several glacial-interglacial cycles), the dominant variability is between glacials and interglacials, with the interglacials exhibiting about 40% higher emissions than colder times. The post-MB increase in CF₄ is thus most easily understood through increased weathering during the warmer post-MB interglacials.

Our work provides new constraints on the regolith hypothesis for the MPT, and the first record of Pleistocene granite weathering/erosion trends.