Atmospheric methane (CH 4 ) is a potent greenhouse gas, and its mole fraction has more than doubled since the preindustrial era 1 . Fossil fuel extraction and use are among the largest anthropogenic sources of CH 4 emissions, but the precise magnitude of these contributions is a subject of debate 2 , 3 . Carbon-14 in CH 4 ( 14 CH 4 ) can be used to distinguish between fossil ( 14 C-free) CH 4 emissions and contemporaneous biogenic sources; however, poorly constrained direct 14 CH 4 emissions from nuclear reactors have complicated this approach since the middle of the 20th century 4 , 5 . Moreover, the partitioning of total fossil CH 4 emissions (presently 172 to 195 teragrams CH 4 per year) 2 , 3 between anthropogenic and natural geological sources (such as seeps and mud volcanoes) is under debate; emission inventories suggest that the latter account for about 40 to 60 teragrams CH 4 per year 6 , 7 . Geological emissions were less than 15.4 teragrams CH 4 per year at the end of the Pleistocene, about 11,600 years ago 8 , but that period is an imperfect analogue for present-day emissions owing to the large terrestrial ice sheet cover, lower sea level and extensive permafrost. Here we use preindustrial-era ice core 14 CH 4 measurements to show that natural geological CH 4 emissions to the atmosphere were about 1.6 teragrams CH 4 per year, with a maximum of 5.4 teragrams CH 4 per year (95 per cent confidence limit)—an order of magnitude lower than the currently used estimates. This result indicates that anthropogenic fossil CH 4 emissions are underestimated by about 38 to 58 teragrams CH 4 per year, or about 25 to 40 per cent of recent estimates. Our record highlights the human impact on the atmosphere and climate, provides a firm target for inventories of the global CH 4 budget, and will help to inform strategies for targeted emission reductions 9 , 10 .

Permafrost and methane hydrates are large, climate-sensitive old carbon reservoirs that have the potential to emit large quantities of methane, a potent greenhouse gas, as the Earth continues to warm. We present ice core isotopic measurements of methane (Δ14C, δ13C, and δD) from the last deglaciation, which is a partial analog for modern warming. Our results show that methane emissions from old carbon reservoirs in response to deglacial warming were small (<19 teragrams of methane per year, 95% confidence interval) and argue against similar methane emissions in response to future warming. Our results also indicate that methane emissions from biomass burning in the pre-Industrial Holocene were 22 to 56 teragrams of methane per year (95% confidence interval), which is comparable to today.

An amendment to this paper has been published and can be accessed via a link at the top of the paper.

The Last Interglacial (129–116 thousand years ago (ka)) represents one of the warmest climate intervals of the past 800,000 years and the most recent time when sea level was metres higher than today. However, the timing and magnitude of the peak warmth varies between reconstructions, and the relative importance of individual sources that contribute to the elevated sea level (mass gain versus seawater expansion) during the Last Interglacial remains uncertain. Here we present the first mean ocean temperature record for this interval from noble gas measurements in ice cores and constrain the thermal expansion contribution to sea level. Mean ocean temperature reached its maximum value of 1.1 ± 0.3 °C warmer-than-modern values at the end of the penultimate deglaciation at 129 ka, which resulted in 0.7 ± 0.3 m of thermosteric sea-level rise relative to present level. However, this maximum in ocean heat content was a transient feature; mean ocean temperature decreased in the first several thousand years of the interglacial and achieved a stable, comparable-to-modern value by ~127 ka. The synchroneity of the peak in mean ocean temperature with proxy records of abrupt transitions in the oceanic and atmospheric circulation suggests that the mean ocean temperature maximum is related to the accumulation of heat in the ocean interior during the preceding period of reduced overturning circulation. Rapid oceanic and atmospheric circulation shifts led to a transient peak in the mean temperature of the ocean at the start of the Last Interglacial, according to noble gas isotope records from an Antarctic ice core.

Ice core measurements of the concentration and stable isotopic composition of atmospheric nitrous oxide (N2O) 74,000-59,000 years ago constrain marine and terrestrial emissions. The data include two major Dansgaard-Oeschger (D-O) events and the N2O decrease during global cooling at the Marine Isotope Stage (MIS) 5a-4 transition. The N2O increase associated with D-O 19 (similar to 73-71.5 ka) was driven by equal contributions from marine and terrestrial emissions. The N2O decrease during the transition into MIS 4 (similar to 71.5-67.5 ka) was caused by gradual reductions of similar magnitude in both marine and terrestrial sources. A 50 ppb increase in N2O concentration at the end of MIS 4 was caused by gradual increases in marine and terrestrial emissions between similar to 64 and 61 ka, followed by an abrupt increase in marine emissions at the onset of D-O 16/17 (59.5 ka). This suggests that the importance of marine versus terrestrial emissions in controlling millennial-scale N2O fluctuations varied in time. Plain Language Summary Nitrous oxide is a powerful greenhouse gas that is produced naturally in soils and oceans. An important unresolved question is the extent to which anthropogenic warming will stimulate additional emissions from these sources, further adding to the warming. Past variations in the abundance of nitrous oxide have been observed using ice core reconstructions, but the reasons for the variations are not well understood. Nitrous oxide produced in soils is isotopically distinct from nitrous oxide produced in oceans. New measurements of the isotopes of atmospheric nitrous oxide provide constraints on how marine and terrestrial sources must have changed, driving fluctuations in nitrous oxide concentration during two intervals of rapid warming and a prolonged period of global cooling. The reconstructed changes in nitrous oxide sources provide insights into relationships between marine and terrestrial ecosystems and climate.