The Methane That Leaks Between the Digester and Your Stove

In 2023, a team of Stanford researchers drove a mobile methane sensor along 60 miles of California pipeline carrying gas from five dairy RNG projects. They found methane concentrations at pipeline junction points that were 2.5 to 4 times higher than the operators' self-reported figures. At one compressor station, the sensor recorded a sustained leak rate of 1.8% of throughput, enough to offset roughly a quarter of the climate benefit the project claimed on paper.

The researchers were not measuring an anomaly. They were measuring the gap between how the RNG industry reports its climate performance and what actually reaches the atmosphere.

The Supply Chain Nobody Audits End to End

Renewable natural gas moves through a longer, more complex supply chain than most people realize. Between the cow and the burner tip, the gas passes through six to eight discrete stages, each with its own leakage profile: anaerobic digestion, gas collection, hydrogen sulfide removal, CO2 separation, compression, dehydration, pipeline injection, and transmission to the end user.

RNG project developers report methane capture efficiency at the digester. That number typically runs 85% to 95%, and it is the figure cited in environmental impact reports, credit applications, and investor decks. What they do not report, because no regulation requires it, is the cumulative methane loss from the digester outlet to the pipeline delivery point.

EPA's Greenhouse Gas Reporting Program requires facilities to estimate emissions at certain processing stages, but the methodology relies heavily on emission factors rather than direct measurement. A 2024 EPA Office of Inspector General report found that actual emissions at natural gas processing facilities exceeded reported figures by 40% to 60% across a sample of 28 sites. The RNG sector uses the same reporting framework.

Stage by Stage, the Losses Add Up

Start at the digester. Covered lagoon systems, the most common type at dairy RNG projects, have documented fugitive emissions of 2% to 5% of generated methane through cover tears, edge seals, and pressure relief vents. Plug-flow and complete-mix digesters perform better, typically losing 0.5% to 2%, but they cost significantly more to install and maintain.

Gas collection piping runs from the digester cover to the processing facility, often across hundreds of yards of open ground. Connection points, condensate drains, and pressure regulators along these runs contribute another 0.3% to 1.5% in losses. A 2022 study by the Environmental Defense Fund measured fugitive emissions at 12 dairy biogas collection systems and found that five had at least one component leaking above 0.5% of system throughput.

The upgrading stage, where raw biogas is purified to pipeline-quality methane, is where the RNG industry's own data gets uncomfortable. Pressure swing adsorption systems, the most widely deployed upgrading technology, vent a methane-containing tail gas stream. Best-in-class PSA units lose 1% to 2% of inlet methane. Older or poorly maintained units can lose 3% to 5%. Membrane separation and amine scrubbing technologies perform somewhat better, with losses of 0.5% to 1.5%, but they represent a minority of installed capacity.

Compression to pipeline pressure adds another 0.2% to 0.8% in losses through compressor seals, valve packing, and blowdown events. The California Air Resources Board's 2023 assessment of RNG facility emissions found that compressor stations were the single most under-reported source of fugitive methane in the biogas-to-pipeline pathway.

The Number That Matters: Net Destruction Efficiency

Add the losses together across a typical dairy RNG project using covered lagoon digestion and PSA upgrading. Conservative estimates, using industry-reported figures rather than independent measurements, put cumulative supply-chain methane loss at 5% to 12% of captured gas. Independent measurements consistently come in at the high end or above that range.

This means a project reporting 90% methane capture at the digester delivers a net destruction efficiency of 79% to 86% when measured at the pipeline injection point. For projects with older equipment, longer collection runs, or covered lagoon systems in poor repair, net destruction can drop below 75%.

By comparison, an enclosed flare installed directly at the digester outlet destroys methane at 98% to 99.5% efficiency. There is no upgrading, no compression, no pipeline, and no compressor station between the methane source and the destruction point. The supply chain is 50 feet of pipe and a burner.

The climate math is not subtle. A project that captures 90% of generated methane but loses 12% through the supply chain achieves 79% net destruction. An enclosed flare that captures the same 90% and destroys it on-site at 99% efficiency achieves 89% net destruction. That 10-percentage-point gap, applied across thousands of dairy sites, represents millions of tons of CO2 equivalent per year.

Why the Industry Does Not Report These Numbers

The RNG industry's business model depends on environmental credit revenue. Under the federal Renewable Fuel Standard, each unit of RNG generates Renewable Identification Numbers (RINs) based on lifecycle greenhouse gas reduction compared to fossil natural gas. Under California's Low Carbon Fuel Standard, credits are calculated using a carbon intensity score that accounts for upstream emissions.

Both programs use modeled emission factors rather than requiring continuous direct measurement of supply-chain methane losses. Project developers submit pathway applications using default emission factors or engineering estimates, which are then reviewed and approved. Once approved, the pathway's carbon intensity score remains fixed for years, regardless of actual field performance.

This creates a structural incentive to undercount supply-chain losses. A 2% improvement in reported net emissions can translate to hundreds of thousands of dollars in additional annual credit revenue. And because no third party is required to verify actual field emissions with direct measurement, the honor system governs billions of dollars in environmental credit transactions.

EPA's proposed revisions to the Greenhouse Gas Reporting Program, published in late 2024, would require direct measurement at certain RNG processing stages. The RNG industry's trade association submitted comments opposing the requirement, arguing that it would impose "disproportionate costs" on small producers. The comment did not address the disproportionate climate impact of unreported emissions.

What Honest Accounting Would Change

If environmental credit programs required verified, end-to-end methane accounting from digester to pipeline, three things would change immediately.

First, the carbon intensity scores assigned to dairy RNG projects would increase by 15% to 40%, reducing their credit value and narrowing the economic advantage over conventional natural gas. Projects at the margin of viability, particularly those serving smaller herds, would no longer pencil out.

Second, the comparison between RNG and on-site methane destruction would shift decisively. Enclosed flares already win the cost-per-ton comparison by a factor of six to eight. Honest leakage accounting would widen the destruction-efficiency gap as well, making the case for RNG as a climate solution even harder to sustain at sub-scale sites.

Third, project developers would face real economic incentives to reduce supply-chain losses, which would actually improve the climate performance of the RNG projects that do make economic sense at scale. Better monitoring, newer upgrading technology, and shorter collection runs would become competitive advantages rather than optional expenses.

The methane leaving dairy lagoons is a real climate problem that demands real solutions. But the value of any solution has to be measured by how much methane it actually keeps out of the atmosphere, not by how much it captures on paper before the losses begin.