Overexplaining or underexplaining methane's role in climate change.

Methane lies at the nexus of climate and air quality, being both a major anthropogenic greenhouse gas—causing about one-half of the warming of carbon dioxide—and a precursor of tropospheric ozone pollution. Over the industrial era, atmospheric methane abundances rose from about 720 parts per billion (ppb) (10 mole fraction) to over 1,850 ppb today. Humans have driven this change largely through agriculture, waste, and fossil fuel emissions. The community’s regular review of the science of atmospheric methane via the Intergovernmental Panel on Climate Change (IPCC) Assessment Reports [First Assessment Report (FAR), 1990 (1); Second Assessment Report (SAR), 1995 (2); Third Assessment Report (TAR), 2001 (3); Fourth Assessment Report (AR4), 2007 (4); and Fifth Assessment Report (AR5), 2013 (5)] has maintained both a scientific interest and political urgency as nations seek to mitigate near-term climate change and keep the overall warming to less than 2 °C (6–9). The recent history (Fig. 1), based on ref. 10, shows a complex overall growth with different rates and even a pause from 2000 to 2006. The many conflicting reports of this recent variability (11–23) suggest that it remains unexplained, or perhaps overexplained. Past work has separately used measurements ofmethane, its isotopes, and related gases to interpret the methane history. Two new publications (24, 25) combine these complementary data into a consistent Bayesianmodeling framework and use advanced statistical methods to match all observations simultaneously subject to the prior constraints. Notably, they advance our understanding of what could have caused the variability. The similarities and differences of optimal solutions that emerge from both studies teach us about the information contained in present observations, as well as their limits. Both Rigby et al. and Turner et al. suggest that oxidation of methane by tropospheric OH increased from the 1990s through the 2000s and that this loss process was mainly responsible for the brief plateau in global mean methane. In contrast, most past work pointed to stable emissions as the cause of the plateau. The new papers differ, however, on what explains the steady rise since 2007: Rigby et al. (24) find a high likelihood that methane emissions rose while OH fell; Turner et al. (25) suggest, counterintuitively, that methane emissions decreased but OH decreased even more. Nevertheless, both papers agree that the observational constraints can be accommodated with differing, nearly optimal solutions, including constant OH levels, so there is no contradiction between their conclusions. Although there are many emission and climate processes that can change the amount of tropospheric OH and hence the rate of methane loss, neither model NOAA ESRL GMD April 2017 see Dlugokencky et al., 1994 1850