Human activities account for roughly two-thirds of the methane entering the atmosphere each year, with natural processes making up the remaining third. That approximate split, however, masks enormous complexity on both sides. Natural methane emissions are dominated by wetlands but also include termites, thawing permafrost, ocean seeps, and even living plants. Human-caused emissions come from fossil fuel extraction, livestock, rice farming, landfills, and wastewater systems. The balance between these sources is not static, and recent record-breaking growth in atmospheric methane has researchers scrutinizing both categories more closely than ever.
Why Atmospheric Methane Is Growing So Fast
Methane concentrations in the atmosphere have been climbing sharply since around 2007, after a period of relative stability. Between 2007 and 2013, global methane grew at about 5.7 parts per billion per year, and that pace spiked to 12.5 parts per billion in 2014 alone.1Global Biogeochemical Cycles. Rising atmospheric methane: 2007–2014 growth and isotopic shift The growth rates in 2020 and 2021 then broke all records since systematic measurements began in 1983. Researchers estimate that global methane emissions jumped by about 27 teragrams in 2020 compared to 2019, reaching roughly 611 teragrams per year, and stayed elevated at around 605 teragrams in 2021.2Atmospheric Chemistry and Physics. Methane emissions are predominantly responsible for record-breaking atmospheric methane growth rates in 2020 and 2021
The puzzle is figuring out which sources are driving that surge. Teasing apart the contributions of wetlands responding to warmer and wetter conditions, expanding fossil fuel operations, and growing livestock herds is one of the thorniest problems in atmospheric science. That challenge is what makes the natural-versus-human distinction so important and so difficult to pin down precisely.
Wetlands Dominate the Natural Budget
Wetlands are the single largest natural source of methane, responsible for median emissions of around 164 teragrams per year, which is roughly a third of all methane released globally.3PubMed. Methane emissions from wetlands: biogeochemical, microbial, and modeling perspectives from local to global scales The basic mechanism is straightforward: waterlogged soils are low in oxygen, and microorganisms called methanogens thrive in those oxygen-deprived conditions, breaking down organic matter and releasing methane as a byproduct. The gas then escapes to the atmosphere through the water, through plant stems, or as bubbles.
Tropical wetlands alone account for a large share of the total. One global assessment estimated tropical wetland emissions at about 94 teragrams per year, though the range of uncertainty is wide, from 56 to 158 teragrams, because fluxes vary enormously even within a single wetland type.4Global Biogeochemical Cycles. Assessing Methane Emissions From Tropical Wetlands: Uncertainties From Natural Variability and Drivers at the Global Scale Inland deep-water wetlands and shallow-water wetlands each contribute substantially, while coastal wetlands and human-made wetlands add smaller amounts. Temperature plays a dominant role in controlling how much methane wetlands produce. Both soil temperature and air temperature show the strongest effects on wetland methane fluxes, with warmer conditions generally driving higher emissions.5PubMed Central. Unraveling Spatially Diverse and Interactive Regulatory Mechanisms of Wetland Methane Fluxes to Improve Emission Estimation This creates a worrying feedback loop: as global temperatures rise, wetlands emit more methane, which in turn warms the planet further.
Termites, Geological Seeps, and Plants
Beyond wetlands, several other natural processes contribute methane to the atmosphere. Termites are one of the more surprising sources. These insects harbor methane-producing microbes in their guts, and their sheer global biomass means the emissions add up. Recent estimates put termite methane at around 15 teragrams per year, which accounts for roughly 4% of emissions from natural sources when wetlands are excluded.6PubMed. The challenge of estimating global termite methane emissions However, that number comes with significant caveats. A large fraction of the methane produced inside termite mounds gets consumed by other microbes before ever reaching the atmosphere, perhaps 20 to 80% depending on the species and mound structure.7Scientific Reports. Global termite methane emissions have been affected by climate and land-use changes Estimates based on isolated termites in laboratory settings probably overstate how much actually escapes into the air.
Geological sources also contribute. Methane seeps from the ocean floor, where gas migrates upward through faults, mud volcanoes, and other structures. These seeps are often associated with gas hydrates, ice-like deposits of methane trapped under high pressure in marine sediments.8Geofluids. A Geophysical Review of the Seabed Methane Seepage Features and Their Relationship with Gas Hydrate Systems When sea levels drop, reduced water pressure can destabilize these hydrates and release trapped gas, a process that has played out repeatedly over geological time.9Scientific Reports. A record of seafloor methane seepage across the last 150 million years
Living plants are another source that surprised researchers when it was first documented. Terrestrial plants can emit methane under normal oxygen-rich conditions through a process that is still not fully understood.10PubMed. Methane emissions from terrestrial plants under aerobic conditions High ultraviolet stress can also cause plant material to break down in ways that release methane.11PubMed Central. Emission of methane from plants The global magnitude of this plant source remains uncertain, but it adds another layer of complexity to the natural budget.
Thawing Permafrost and Thermokarst Lakes
Among natural sources, thawing permafrost gets the most attention for its future potential. As Arctic and alpine permafrost melts, ancient frozen organic matter becomes available for microbial decomposition, producing methane in the process. Thermokarst lakes, which form when ice-rich ground collapses, are particularly potent emitters. One airborne survey across Alaska and northwestern Canada found that methane hotspots occupied less than 0.01% of the northern permafrost land area but contributed an estimated 1.1 teragrams of methane per year, roughly 4% of the entire pan-Arctic wetland budget.12Global Biogeochemical Cycles. Characterizing Methane Emission Hotspots From Thawing Permafrost These hotspots can be extraordinarily concentrated: at one site in interior Alaska, extreme daily methane fluxes from a single hotspot adjacent to a thermokarst lake were more than twenty times the average from the surrounding area.
On the Tibetan Plateau, thermokarst lakes cover just 0.2% of the permafrost region but emit enough methane annually to offset about 6% of the carbon absorbed by surrounding grasslands.13Nature Communications. Methane emissions from thermokarst lakes must emphasize the ice-melting impact on the Tibetan Plateau A substantial portion of that methane, about 17% of the annual total, escapes during ice-melting periods, a phase that had previously been underestimated. With further warming, these lakes are projected to expand, potentially increasing their methane output by 1.1 to 1.2 times by 2100.
Making matters worse, the natural process that might slow these emissions, anaerobic oxidation of methane in lake sediments, appears to be negligible in thermokarst settings. In interior Alaska thermokarst lakes, researchers found that the rate of anaerobic methane oxidation was two orders of magnitude lower than the rate of methane production, amounting to less than 2% of what methanogens were producing.14Limnology and Oceanography. Anaerobic oxidation of methane does not attenuate methane emissions from thermokarst lakes In other words, there is virtually no natural brake on methane escaping from these sediments as permafrost continues to thaw.
Fossil Fuel Operations
On the human side of the ledger, fossil fuel extraction and distribution is one of the largest methane sources. The Global Fuel Exploitation Inventory estimated 2020 methane emissions of about 23 teragrams from oil, 20 from gas, and 31 from coal operations worldwide.15Earth System Science Data. Using new geospatial data and 2020 fossil fuel methane emissions for the Global Fuel Exploitation Inventory (GFEI) v3 That totals roughly 74 teragrams per year from fossil fuels alone. These emissions come from leaking pipelines, venting equipment, incomplete combustion in compressor engines, and routine operations across the supply chain.16PubMed. Capturing fugitive methane emissions from natural gas compressor buildings
A persistent problem with fossil fuel methane is that official inventories consistently undercount the real emissions. A meta-analysis of studies published since 2014 found that actual measured emissions from oil and gas operations averaged about 2.5 times what inventory-based estimates predicted.17Earth’s Future. Top‐Down Versus Bottom‐Up Atmospheric Emission Estimates From Oil and Natural Gas Operations In 86% of studies, real-world measurements exceeded the official numbers. Part of the gap comes from the fact that inventory methods and atmospheric measurements capture different timescales. Emissions from oil and gas equipment can spike dramatically during certain operations and at certain times of day, so a measurement taken during peak afternoon hours will look very different from an annual average derived from equipment counts.18PubMed Central. Temporal variability largely explains top-down/bottom-up difference in methane emission estimates from a natural gas production region Still, the consistent direction of the discrepancy, with real measurements nearly always higher, suggests that inventories genuinely miss a lot of leakage.
Livestock and Rice Paddies
The livestock sector is the single largest human-caused methane emitter.19PubMed Central. Strategies to Mitigate Enteric Methane Emissions from Ruminant Animals Cattle, sheep, goats, and other ruminants produce methane in their digestive systems through a process called enteric fermentation. Methanogens living in the rumen convert part of the feed into methane, which the animal mostly exhales or belches. A typical dairy cow in the Netherlands produces around 110 to 130 kilograms of methane per year.20Animal Feed Science and Technology. A model of enteric fermentation in dairy cows to estimate methane emission for the Dutch National Inventory Report using the IPCC Tier 3 approach Multiply that across the world’s roughly one billion cattle, plus hundreds of millions of other ruminants, and the emissions are enormous.
Rice paddies are another major agricultural source. Flooded rice fields create the same kind of oxygen-starved conditions found in natural wetlands, and methanogens thrive in the waterlogged soil. Most of the methane produced in paddy soils gets oxidized before reaching the surface, but the 10 to 20% that escapes represents a globally significant flux.21Communications Earth & Environment. Global methane emissions from rice paddies are now increasingly quantifiable One of the most promising strategies for reducing rice paddy emissions is alternate wetting and drying, where fields are periodically drained rather than kept continuously flooded. Field trials have shown this approach can cut methane emissions by 66 to 70% without reducing rice yields.22PLOS ONE. Influence of rice varieties, organic manure and water management on greenhouse gas emissions from paddy rice soils
Wildfires as a Growing Methane Source
Wildfires sit in an awkward category between natural and human-caused emissions. Fires have always been part of Earth’s ecosystems, but climate change and land management are altering fire regimes in ways that amplify their impact. Globally, biomass burning releases an average of about 19 teragrams of methane each year, with nearly half originating from Africa.23Journal of Geophysical Research: Atmospheres. Methane Emissions From Wildfires: Trends and Anomalies During extreme wildfire events, emissions can surge enough to cancel out years of emission reductions from other human sources. The uncertainty around wildfire methane is also substantial: in boreal forests, different emissions inventories disagree by a factor of nearly three for the same region.24Remote Sensing. Methane Emissions in Boreal Forest Fire Regions: Assessment of Five Biomass-Burning Emission Inventories Based on Carbon Sensing Satellites
How Scientists Tell the Sources Apart
Distinguishing natural methane from human-caused methane in the atmosphere relies heavily on isotopic fingerprinting. Methane molecules contain carbon and hydrogen atoms that come in different isotopic forms, and the ratios of these isotopes vary depending on how the methane was produced. Biological sources like wetlands and livestock tend to produce methane that is lighter in carbon-13 compared to methane from fossil fuel extraction or geological seeps, which tends to be heavier.
The shift to more negative carbon-13 values observed in atmospheric methane since 2007 has been one of the key clues suggesting that biological sources, whether wetlands, agriculture, or both, are driving recent growth.1Global Biogeochemical Cycles. Rising atmospheric methane: 2007–2014 growth and isotopic shift But isotopic attribution is not as clean as it sounds. Hydrogen isotope ratios can sometimes distinguish sources that carbon isotopes cannot, such as telling apart landfill methane from cattle feedlot methane in a region where both are present.25Geophysical Research Letters. Using stable isotopes of hydrogen to quantify biogenic and thermogenic atmospheric methane sources: A case study from the Colorado Front Range
Recent experimental work has also complicated the traditional isotopic categories. Biological methane can span a much wider range of carbon-13 values than previously assumed, depending on how much carbon is available to the methanogens and what environmental conditions they face.26Geochemical Perspectives Letters. The stable carbon isotope fractionation of methanogenesis products at complete carbon consumption When carbon is scarce, the isotopic fractionation between the methane and its source material can nearly disappear, making biological methane look like it came from a geological source. This blurring of isotopic boundaries is one reason the global methane budget remains hard to close precisely.
How Methane Leaves the Atmosphere
Methane does not accumulate forever. Its atmospheric lifetime is roughly 9 to 12 years, far shorter than carbon dioxide, which persists for centuries. The primary removal mechanism is reaction with hydroxyl radicals in the lower atmosphere, which break methane down through a series of chemical steps.27Greenhouse Gases: Science and Technology. Atmospheric removal of methane by enhancing the natural hydroxyl radical sink This hydroxyl-radical pathway accounts for the vast majority of methane destruction.
Soil bacteria known as methanotrophs provide a secondary sink. These microbes consume atmospheric methane as it diffuses into well-aerated soils, removing an estimated 20 to 60 teragrams per year.28PubMed Central. Characterization of methanotrophic bacterial populations in soils showing atmospheric methane uptake This soil sink is sensitive to land use. Grassland soils are important methane consumers, but intensive management with heavy nitrogen inputs and high grazing densities reduces their capacity.29Soil Biology and Biochemistry. Methane sink function of grassland soil microbiomes – negative effects of intensive management persist three years after land-use extensification The damage is not quickly reversible: even three years after switching to less intensive management, the methane-consuming capacity of these soils had not recovered. Methanotrophs in soil represent the only biological sink for atmospheric methane,30Geoscientific Model Development. Soil Methanotrophy Model (MeMo v1.0): a process-based model to quantify global uptake of atmospheric methane by soil making land-use decisions surprisingly relevant to the methane budget.
What 650,000 Years of Ice Cores Show
For perspective on how unusual current methane levels are, ice cores provide an unbroken record stretching back hundreds of thousands of years. Analysis of the EPICA Dome C core from Antarctica shows that atmospheric methane never exceeded about 773 parts per billion during any of the warm interglacial periods over the past 650,000 years.31PubMed. Atmospheric methane and nitrous oxide of the Late Pleistocene from Antarctic ice cores During older, cooler interglacials before 420,000 years ago, peaks were even lower, around 600 parts per billion. Today’s atmospheric methane concentration is roughly 1,920 parts per billion, more than double the highest level seen in any interglacial on record.
Dual isotope records from Antarctic ice cores show that tropical wetlands and seasonally flooded plains were the primary drivers of natural methane variations throughout these glacial cycles.32PubMed Central. Glacial/interglacial wetland, biomass burning, and geologic methane emissions constrained by dual stable isotopic CH(4) ice core records In other words, the natural system has always seen methane go up and down with climate, driven largely by how wet and warm the tropics were. But the magnitude of today’s concentrations is unprecedented in the ice core record, and the excess is almost entirely attributable to human activity layered on top of the natural cycle.
Why Cutting Methane Matters for Near-Term Warming
Because methane is a potent greenhouse gas but breaks down relatively quickly, reducing emissions can yield climate benefits on a timeline measured in decades rather than centuries. One analysis found that pursuing all readily available methane mitigation measures could slow the rate of near-term warming by around 30% and avoid about a quarter of a degree of additional warming by mid-century.33Environmental Research Letters. Acting rapidly to deploy readily available methane mitigation measures by sector can immediately slow global warming Over the longer term, the cumulative benefit could exceed half a degree by the end of the century.
That said, the expected benefits of methane reduction have been debated. Some modeling work has produced more modest projections, estimating that even maximally feasible reductions in methane and black carbon phased in from 2015 to 2035 would reduce temperatures by only about 0.16 degrees Celsius by 2050, with a wide uncertainty range.34PubMed Central. Near-term climate mitigation by short-lived forcers The disagreement partly reflects different assumptions about how aerosol pollution reductions (which happen alongside methane cuts in clean-energy transitions) might offset some of the cooling benefit. When those interactions are accounted for, methane reductions still produce net cooling, about 0.15 degrees by mid-century and 0.50 degrees by end of century, because the methane benefit more than compensates for the warming effect of losing reflective aerosol pollution.35Environmental Research Letters. Significant climate benefits from near-term climate forcer mitigation in spite of aerosol reductions
The practical appeal of methane mitigation lies in where the low-hanging fruit sits. Plugging leaks in natural gas infrastructure, capturing landfill gas, altering water management in rice paddies, and modifying livestock feed are all technically feasible right now. Natural sources like wetlands and permafrost, by contrast, are far harder to control and may accelerate as the planet warms, which makes reducing controllable human emissions all the more urgent.