FEEDBACKS IN THE EARTH SYSTEM: PAST CHANGES IN DUST AND IRON FERTILIZATION OF THE OCEAN

Results of recent open ocean iron enrichment experiments, together with numerical models of the ocean carbon cycle all appear to be consistent with the “iron hypothesis” – increased dust supply to the Southern Ocean during glacial periods enhancing biological productivity and driving atmospheric CO₂ lower. The entrainment of this dust from the land surface is facilitated by arid soil conditions and sparse vegetation cover, while its transport to the open ocean becomes more efficient under a less vigorous hydrological cycle. That climate should exert a strong control on dust supply is consistent with information contained within ice cores, which show greatly enhanced dust deposition during cold dry glacial periods. Taking these two different linkages in the Earth system together, if changes in dust flux affect atmospheric CO₂ and with it global climate, and dust fluxes are in turn responsive to climate, a ‘feedback’ in the Earth system can be identified. The sign of this feedback is positive overall, and so will act as a destabilizing influence on the climate system.

Analysis of the dust-CO₂-climate feedback suggests that its impact is not uniform in climate space, but exhibits a maximum effect during mid-to-full glacial conditions when it can account for about one third of the temperature variability recorded in Antarctic ice cores. Because it appears primarily to be an intra-glacial phenomenon, it may help explain the reasons for the difference between ‘41 ka’ and ‘100 ka’ climatic regimes – the ‘41 ka’ glacial-interglacial cycles of the earlier Quaternary being characterized by the absence of a ‘full’ glacial state. However, the importance of this feedback is highly dependent not only on the sensitivity of atmospheric CO₂ to changes in dust supply (a question for the ocean carbon cycle modeling community), but also on the nature of the apparently highly nonlinear response of dust deposition to changes in global climate. Elucidating the details of the controls on dust sources and atmospheric transport as the Earth descends into a deep glacial state may therefore be a prerequisite to understanding Quaternary atmospheric CO₂ and climate variability.