Methane (CH 4 ) has a large contribution to the global radiative budget and is responsible for about 0.5°C of present global warming over the period 1850 -1900(IPCC, 2021. Methane has a relatively short perturbation lifetime (12.4 years (Balcombe et al., 2018)) and high global warming potential (28-36 times that of CO 2 over a 100year period (IPCC, 2021)). As such, a decline in CH 4 emissions will rapidly reduce global CH 4 concentrations and mitigate the impact of climate change at decadal time scales (United Nations Environment Programme & Climate & Clean Air Coalition, 2021). However, any efforts to target CH 4 emissions reductions require a thorough understanding of the dominant CH 4 sources and sinks and their temporal and regional distribution and trends.Methane is produced in three ways-pyrogenically, thermogenically, or biogenically-from both anthropogenic and natural processes. Pyrogenic sources of CH 4 include biofuel combustion (e.g., wood burning for heating and cooking) and biomass burning (e.g., wildfires and peat fires). All pyrogenic sources produce CH 4 from the incomplete combustion of organic matter. Thermogenic CH 4 is produced from the breakdown of organic matter buried deep within the Earth's crust at high pressure and temperature. Although geological CH 4 is released naturally into the atmosphere through gas seeps, most is released through activities related to the exploration, mining, and transport of fossil fuels (Hmiel et al., 2020;Janssens-Maenhout et al., 2019;Petrenko et al., 2017). The majority of biogenic CH 4 is produced in anaerobic environments by the microbial mediated breakdown of organic matter. These environments include wetlands, inland waters, marine sediments, ruminants such as cattle, rice paddies, manure management and wastewater and landfill systems. Small quantities of CH4 are also produced from the aerobic bacterial metabolization of methylated compounds (e.g., Florez-Leiva et al., 2013) and even photochemically (Li et al., 2020). Counter-balancing these CH 4 sources are three chemically driven atmospheric sinks of CH 4 . The first two reactions with tropospheric OH radicals and tropospheric atomic chlorine account for ~88% (476 -677 Tg CH 4 yr −1 ) and ~2% (1-35 Tg CH 4 yr −1 ) of the total sink, respectively, with a third stratospheric sink (e.g., reaction with O('D), Cl and OH in the stratosphere) accounting for a further ~5% (12-37 Tg CH 4 yr −1 ) (Saunois et al., 2020). However, due to their highly reactive nature, the key reactants are inherently difficult to quantify, driving a significant level of uncertainty in the spatial and temporal distribution of atmospheric sink estimates (Zhao et al., 2019). Many fundamental aspects of the spatial distribution of OH are currently unresolved, for example, estimates of the interhemispheric gradient can vary from 0.85 to 1.4 (NH/SH) depending on the methodology

Highlights d Northern peatlands remain a CO 2 sink of $0.1 Pg C year À1 until 2300 under RCP2.6 d Northern peatlands become a CO 2 source of $0.2 Pg C year À1 by 2300 under RCP8.5 d CH 4 emissions from northern peatlands will increase 5-fold by 2300 under RCP8.5 d Modeling of peatland resilience, vegetation, and peat quality changes should be improved

No abstract

No abstract

Process-based land surface models are important tools for estimating global wetland methane (CH 4 ) emissions and projecting their behavior across space and time. So far there are no performance assessments of model responses to drivers at multiple time scales. In this study, we apply wavelet analysis to identify the dominant time scales contributing to model uncertainty in the frequency domain. We evaluate seven wetland models at 23 eddy covariance tower sites. Our study first characterizes site-level