What causes the ice ages in the late Pliocene and Pleistocene?

Abstract

For more than a century, two families of ice age theories have been proposed: insolation based theories proposed by Adhemar, and atmospheric CO2 ones proposed by Tyndall. The major technique advance of deep-sea sediment drilling, as well as new interpretations of stable oxygen isotopic composition of the deep-sea fossil foraminifera, which is regarded as a proxy index of continental ice volume and sea water temperature, established the now well recognized glacial-interglacial variations in climate over the late Pliocene and Pleistocene, e.g. ~3 Ma. A landmark progress came from Hays et al. (1976) who unambiguously demonstrated that the change in insolation induced by the Earth’s axial tilt (obliquity), the wobble of this tilt (precession), and the degree of circularity of the Earth’s orbit around the sun (eccentricity) brings on ice ages (glaciations) every hundred thousand years or so during the Pleistocene. Many subsequent observations, from the deep-sea to continental records, confirmed the co-variation between insolation and the alternation of glacial ice age phases, and warmer interglacial phases, at least in terms of frequency domains. However, these findings cannot explain the so called ~100 ka conundrum, where the size of 100 ka insolation forcing is relatively small yet the apparent ice sheet response is large, as well as the bi-polar symmetry in climate changes at precessional frequency of ~23 ka, despite the fact that the precession cycle drives insolation changes in antiphase between the two hemispheres. As such, other factors may be considered in the driving of Northern Hemisphere Glaciation (NHG) global glacial-interglacial variations. Several hypotheses have been put forward, such as: (1) the gradual decreasing CO2 drove global cooling through to a threshold value where insolation changes may force the ice age cycles; (2) the effect of local insolation that modulated the East Antarctica Ice Sheet after the gradual CO2 decline that drove its growth to maximal extent; (3) the Southern Ocean “marine biological pump” changes forced by the ocean water stratification and current changes that modulated the atmospheric CO2 thus triggered ice ages; (4) the clearing of regolith under the North American ice sheets which modulate ice sheet thickness and extent changes which allow the ice sheets to grow further south and be influenced more by precessional cycles, as well as strengthened the fresh bedrock weathering and drawdown more atmospheric CO2; and (5) because ice mass balance depends on whether the temperature is above or below the freezing point, a physically more relevant parameter to measure insolation forcing should be the insolation integrated over a given threshold that allows for ice melting, and thus triggering the glaciations. Unfortunately, we cannot determine which one is right or several of them acting together so far. The most distinct features of the ice age climate are both the onset of the NHG at ~2.7 Ma and the transition of ~41 ka based glacial-interglacial cycles to cycles of a ~100 ka dominated frequency at ~1.0 Ma, namely the Early Middle Pleistocene Transition. We suggest that, both insolation and CO2 changes together forced the climate of the ice ages: the stepwise cooling that occurred at ~2.7 Ma and ~1.0 Ma may be linked with gradual atmospheric CO2 reduction, during which the atmospheric CO2 levels fell to a threshold value that triggered high-latitude (both poles) ice expansion. Under a new cooling state since ~2.7 Ma, the dominant climatic cycle at ~41 ka was forced by obliquity (orbital tilt), while the ~23 ka precession cycle was cancelled by opposing effects at both poles. On the other hand, we propose that the ~100 ka cycles that became dominant following approximately 1 Ma were forced by a CO2 based “marine biological pump” change in the Southern Ocean, or is related to stochastic behavior of the ice sheets. The ~400 ka eccentricity cycle, which is a stronger absolute influence on insolation is an outstanding rhythm in the late Cenozoic, yet is not as strong in climate records as the precessional and obliquity cycles during the late Pliocene and Pleistocene, and thus needs further examination. In summary, we suggest ice ages are caused by the integrated forcing of insolation, atmospheric CO2, ocean conveyer and ice sheet feedback etc., in which the insolation forcing is primary, a threshold value of the other facts was attained, the ice age happened. However, our hypothesis is very tentative, more investigation is still needed on what causes the ice ages.