Modelling Coupled Oscillations of Volcanic CO2 Emissions and Glacial Cycles
Following the mid-Pleistocene transition, the dominant period of glacial cycles changed from 40 ka to ~100 ka. It is broadly accepted that the 40 ka glacial cycles were driven by cyclical changes in obliquity. However, this forcing does not explain the ~100 ka glacial cycles. Mechanisms have been proposed for ~100 ka cycles, but none are universally accepted. Any proposed mechanism for ~100 ka glacial cycles must give the Earth's climate system a memory of tens-of-thousands-of-years. This timescale is difficult to achieve for surface processes, however it is possible for the solid Earth. Recent work suggests volcanic CO2 emissions change in response to glacial cycles and that there could be a ~50 ka delay in that response. Such a lagged response could drive glacial cycles from 40 ka cycles to an integer multiple of the forcing period. Under what conditions could the climate system admit such a response? We investigate this using a reduced-complexity model to establish the threshold volcanic variations required to induce 100 ka glacial cycles. This threshold is compared to the probable range of volcanic activity over glacial cycles. Our model is composed of three component models for energy balance, ice sheet growth and atmospheric CO2. The model is driven by insolation alone, with other components varying according to a system of coupled, differential equations. The model replicates modern climate conditions and seasonality, CO2-doubling experiments, and - when forced with ice core CO2 - the late-Pleistocene sea level record. We obtain a switch from 40 ka to ~100 ka cycles as the mid-ocean ridge CO2 response to glacial cycles is increased. These ~100 ka cycles are phase-locked to obliquity, lasting 80 or 120 ka. The mid-ocean ridge response required is, at best, just within a 95% confidence interval of MOR CO2 emissions.