This study presents GIS-based estimates of permafrost extent in the northern circumpolar region during the Last Glacial Maximum (LGM), based on a review of previously published maps and compilations of field evidence in the form of ice-wedge pseudomorphs and relict sand wedges. We focus on field evidence localities in areas thought to have been located along the past southern border of permafrost. We present different reconstructions of permafrost extent, with areal estimates of exposed sea shelf, ice sheets and glaciers, to assess areas of minimum, likely and maximum permafrost extents. The GIS-based mapping of these empirical reconstructions allows us to estimate the likely area of northern permafrost during the LGM as 34.5 million km(2) (which includes 4.7 million km(2) of permafrost on exposed coastal sea shelves). The minimum estimate is 32.7 million km(2) and the maximum estimate is 35.3 million km(2). The extent of LGM permafrost is estimated to have been between c. 9.1 to 11.7 million km(2) larger than its current extent on land (23.6 million km(2)). However, 2.4 million km(2) of the lost land area currently remains as subsea permafrost on the submerged coastal shelves. The LGM permafrost extent in the northern circumpolar region during the LGM was therefore about 33 percent larger than at present. The net loss of northern permafrost since the LGM is due to its disappearance in large parts of Eurasia, which is not compensated for by gains in North America in areas formerly covered by the Laurentide ice sheet.

Atmospheric concentrations of carbon dioxide increased between the Last Glacial Maximum (LGM, around 21,000 years ago) and the preindustrial era(1). It is thought that the evolution of this atmospheric carbon dioxide (and that of atmospheric methane) during the glacial-to-interglacial transition was influenced by organic carbon that was stored in permafrost during the LGM and then underwent decomposition and release following thaw(2,3). It has also been suggested that the rather erratic atmospheric delta C-13 and Delta C-14 signals seen during deglaciation(1.4) could partly be explained by the presence of a large terrestrial inert LGM carbon stock, despite the biosphere being less productive (and therefore storing less carbon)(5,6). Here we present an empirically derived estimate of the carbon stored in permafrost during the LGM by reconstructing the extent and carbon content of LGM biomes, peatland regions and deep sedimentary deposits. We find that the total estimated soil carbon stock for the LGM northern permafrost region is smaller than the estimated present-day storage (in both permafrost and non-permafrost soils) for the same region. A substantial decrease in the permafrost area from the LGM to the present day has been accompanied by a roughly 400-petagram increase in the total soil carbon stock. This increase in soil carbon suggests that permafrost carbon has made no net contribution to the atmospheric carbon pool since the LGM. However, our results also indicate potential postglacial reductions in the portion of the carbon stock that is trapped in permafrost, of around 1,000 petagrams, supporting earlier studies(7). We further find that carbon has shifted from being primarily stored in permafrost mineral soils and loess deposits during the LGM, to being roughly equally divided between peatlands, mineral soils and permafrost loess deposits today.

Vegetation is an important feature in the Earth system, providing a direct link between the biosphere and atmosphere. As such, a representative vegetation pattern is needed to accurately simulate climate. We attempt to reconstruct past and present vegetation with a data driven approach, to test if this allows us to create robust global and regional vegetation patterns. The motivation for this stems from the possibility of avoiding circular arguments when studying past time periods where vegetation is used to reconstruct climate, and climate is used to construct vegetation. By using the Random Forest machine learning tool, we train the vegetation reconstruction with available biomized pollen data of present and past conditions and are able to produce reasonable broad-scale vegetation patterns for the Pre-Industrial and the Mid-Holocene together with a few modeled climate variables. We test the methods robustness by introducing a systematic temperature bias based on existing climate model spread and compare the result with that of LPJ-GUESS, a process-based dynamic global vegetation model. Results prove that the Random Forest approach is able to produce robust patterns for periods and regions well constrained by evidence, but fails when evidence is scarce. The robustness is achieved by sacrificing a dynamic vegetation response to a changing climate.

The transfers of carbon between land, ocean and atmosphere, and their relation to temperature variability over glacial and interglacial cycles continue to intrigue the scientific community. Over the past four decades, many have focused on the role of the Southern Ocean to explain the atmospheric carbon dioxide (CO₂) patterns seen in ice core records, but recent advances also include mentions of a possible terrestrial component. We quantify important terrestrial organic soil carbon (C) stocks north of 23˚, using palaeo-data and modeled climate to reconstruct terrestrial C dynamics from the Last Glacial Maximum until present at millennial time steps. During the deglaciation, C storage declined to reach a minimum around 10 kyr BP, a trend which then turned and led to progressively higher soil C stocks during the Holocene. Net changes in mineral soil C stocks are small, even though significant geographic shifts occurred; instead, deglacial and interglacial terrestrial C stock dynamics are dominated by losses from permafrost loess, inundation of continental shelves and gains in peatlands, processes commonly overlooked in complex Earth System Models.