Simulated rise in atmospheric CO2 concentrations during the most recent deglaciation event, resulting from the release of a Southern Ocean carbon pool (red line) and the breakdown of the permafrost and Southern Ocean carbon pools (blue line). Inclusion of the permafrost carbon pool results in a better match with CO2 reconstructions from the EPICA Dome C ice core (black diamonds), suggesting a plausible role of permafrost carbon in the deglacial CO2 rise during the Mystery Interval. The error bars are the 2σ error estimates taken from the EPICA Dome C ice core data. Graphic: Crichton, et al., 2016 / Nature Geoscience

By Andrew H. MacDougall
22 August 2016

(Nature Geoscience) – Between 17,500 and 14,500 years ago, a period sometimes referred to as the Mystery Interval1, atmospheric CO2 concentrations began their post-glacial rise from about 190 ppm in glacial times to approximately 270 ppm by the beginning of the Holocene. The rise in CO2 during the Mystery Interval is associated with large negative anomalies in the carbon isotopic composition of CO2 (refs 2,3). These anomalies suggest that a long-isolated carbon pool that was formed from a biological source was released to the atmosphere. A large pool of old 13C-depleted carbon in the Southern Ocean has been invoked as the source, but questions over the timing and magnitude of this release remain. Writing in Nature Geoscience, Crichton and colleagues5 report evidence from numerical simulations that suggest the primary source of the deglacial carbon during the Mystery Interval was instead a permafrost carbon pool.

Soil and bedrock that have temperatures below 0 °C for longer than two years are considered to be permafrost6. Permafrost soils hold an immense quantity of carbon in the form of partly decayed organic matter: carbon held in permafrost-affected soils is estimated to comprise ~35% of the total terrestrial carbon pool in the modern world7. Much of this carbon is held in permafrost soil horizons and is therefore frozen and protected from microbial decay7. Permafrost carbon — like all organic matter — has a low δ13C value and, because it can be locked in frozen soils for thousands of years, permafrost carbon typically has very little radiocarbon remaining6.

Extensive permafrost regions are thought to have existed during the Last Glacial Maximum. Under this cold climate, even though terrestrial productivity was half that of the pre-industrial period, the carbon pool housed in soils and vegetation was only ~10% smaller than that of the late Holocene8. However, the inactive fraction of the terrestrial carbon pool was about 45% larger than that of which exists today8. There is no palaeoclimate proxy to directly estimate the size of the permafrost carbon pool during the Last Glacial Maximum. Nonetheless, a large glacial permafrost carbon pool fits the criteria of a large inert carbon pool in a low primary productivity world.

Crichton and colleagues5 use an Earth system model of intermediate complexity to simulate the evolution of atmospheric CO2 concentration from the Last Glacial Maximum until the year 1850. In model simulations, the dissipation of the enhanced Southern Ocean carbon pool enlarges the atmospheric carbon pool by over 100 ppm of CO2 (red line, Fig. 1). However, the rise in CO2 concentrations occurs roughly 3,000 years after the rise observed in the ice-core record. Adding a permafrost carbon module to the Earth system model narrows the difference between the model simulation and palaeoclimate CO2 record, with simulated CO2 and δ13C closely matching the data until the onset of the Holocene (blue line, Fig. 1).

The simulations suggest a simplified storyline for the deglacial rise in atmospheric CO2. At the end of the Last Glacial Maximum, changes in Earth's orbit caused Northern Hemisphere summertime insolation to rise. These warmer summer conditions induced thaw of permafrost soils, which began to release long-sequestered carbon as CO2. The liberated permafrost carbon further warmed the climate, inducing deglaciation and further release of carbon from permafrost soils. Sea-level rise and a warming climate then triggered changes in brine formation and sinking in the Southern Ocean, which resulted in the dissipation of the glacial Southern Ocean carbon pool. However, regrowth of the terrestrial biosphere sequestered more carbon than was released from the terrestrial realm. Thus in net terms, although deglaciation was promoted by the release of permafrost carbon to the atmosphere, the ocean carbon pool was the dominant source of the glacial–interglacial rise in atmospheric CO2 concentrations. [more]

Permafrost carbon: Catalyst for deglaciation

ABSTRACT: The atmospheric concentration of CO2 increased from 190 to 280 ppm between the last glacial maximum 21,000 years ago and the pre-industrial era1, 2. This CO2 rise and its timing have been linked to changes in the Earth’s orbit, ice sheet configuration and volume, and ocean carbon storage2, 3. The ice-core record of δ13CO2 (refs 2,4) in the atmosphere can help to constrain the source of carbon, but previous modelling studies have failed to capture the evolution of δ13CO2 over this period5. Here we show that simulations of the last deglaciation that include a permafrost carbon component can reproduce the ice core records between 21,000 and 10,000 years ago. We suggest that thawing permafrost, due to increasing summer insolation in the northern hemisphere, is the main source of CO2 rise between 17,500 and 15,000 years ago, a period sometimes referred to as the Mystery Interval6. Together with a fresh water release into the North Atlantic, much of the CO2 variability associated with the Bølling-Allerod/Younger Dryas period ~15,000 to ~12,000 years ago can also be explained. In simulations of future warming we find that the permafrost carbon feedback increases global mean temperature by 10–40% relative to simulations without this feedback, with the magnitude of the increase dependent on the evolution of anthropogenic carbon emissions.

Permafrost carbon as a missing link to explain CO2 changes during the last deglaciation


  1. rpauli said...

    thank you for this.. most of the world would not have been able to see this if it was not for DES .. thank you so much. I think.  

  2. Jim said...

    I looked all over, but neither Cardiff University nor any other source wrote a press release around this paper. Which is weird, because it's so important, possibly one of the most revealing studies in climate science this year.  

  3. rpauli said...

    Good find, Jim.

    It's the curse of the costly communications... so much is unread because of the fees... Free only to academics, poor students and professors are overwhelmed and unable to read and respond.


Blog Template by Adam Every . Sponsored by Business Web Hosting Reviews