Inside the high-dollar race to sell natural gas as low-carbon

This is part one of an in-depth reporting project. Also read part two.

By night, the industrial operation sprawling across the marshland of Louisiana’s southwest tip seems otherworldly. The 1.6-square-mile complex of tubes, tanks and machinery emits a dull hiss, and its thousands of yellowish lights and three flame-topped flare towers cast an eerie glow. By day, when the facility is easier to discern, so is its earthly purpose. Natural gas piped from fields across the United States whooshes into the plant, where six 1,200-foot-long lines of engines, fans and compressors cool the gas into “liquefied natural gas,” or LNG. The frigid fossil fuel then shoots out of the complex and into gargantuan oceangoing tankers, each roughly as long as three football fields, waiting at the facility’s dock on a channel that feeds the Gulf of Mexico. Once loaded, the ships lumber into the Gulf and set a course based on the location of their cargo’s buyers. Some turn east and cross the Atlantic to Europe; others head south, then sail west through the Panama Canal and across the Pacific to Asia. When they reach their destinations — maybe France, or Spain, or China, or South Korea — their LNG is warmed back into a gas. Then it’s burned, producing everything from plastics to power.

Dominating a coastal flank of Cameron Parish, Louisiana, and squatting along Sabine Pass, a waterway that defines the Texas state line, this is the biggest LNG-export terminal in the U.S., the nation that this year is likely to become the world’s largest LNG exporter. In 2021, this complex chilled about 35% of all the LNG the U.S. exported, and about 3% of all the natural gas the U.S. drilled up. In the process, it sent skyward the equivalent of 5.6 million metric tons of carbon dioxide, roughly the annual output of Vermont. Yet even that, according to various estimates, likely represented less than one-tenth of the climate impact of the LNG this facility sold. The gas liquefied here is burned the world over, notably in fast-growing Asian economies and in a Europe newly thirsty for U.S. LNG to replace gas it previously had piped in from Vladimir Putin’s Russia but has been unable to get in the wake of that country’s invasion of Ukraine. Overseas combustion is where natural gas produced and liquefied in the U.S. does its greatest damage to the climate. The Sabine Pass LNG colossus is both a pivotal point in a wildly gyrating international energy business and a hot spot in the fight over how to fix the damage LNG is wreaking on a warming world.

Natural gas is humanity’s looming energy and climate dilemma. Coal, the dirtiest fossil fuel, provides a declining share of total energy. The share of solar, wind and other renewable energy sources is soaring, but from a comparatively small global base. In between coal and renewables on the spectrum of carbon intensity sits natural gas, a fossil fuel that, studies project, will be burned in massive quantities for decades to come.

Natural gas long has been marketed as a “bridge fuel” to a cleaner energy future — a dependable energy source that, when burned to generate electricity, produces about half as much carbon dioxide as does coal. But scientific studies over the past decade have underscored that neither natural gas in general nor LNG in particular is as clean as previously claimed. Each step in the production and distribution of natural gas leaks methane, a hydrocarbon that is, pound for pound, roughly 80 times as damaging to the climate as carbon dioxide for the first 20 years after it’s emitted, making methane a crucial target for near-term climate action. The International Energy Agency warned in 2021 that methane emissions are “the second-largest cause of global warming,” behind carbon-dioxide emissions. In addition, the extra steps needed to trade a load of natural gas as LNG — liquefying the gas, shipping it abroad and then warming it back into a gas once it’s delivered to its international destination — increase the gas’s total carbon footprint.

If today’s government policies and market trends endure, natural gas will account for 24% of global carbon-dioxide emissions from combustion in 2050, up from 21% in 2021, the International Energy Agency projects. That scenario envisions global carbon emissions falling 13% by midcentury — nowhere near the drop scientists say is necessary to avert climate disaster, but more aggressive than today’s reality, in which emissions are still slightly rising. Barring the rosiest predictions for renewable energy’s domination — predictions that aren’t out of the question but that are nowhere near certain enough to bet the planet on — humanity will stanch climate change only if it can ax emissions of natural gas. The gas industry, under mounting pressure from investors and regulators, is scrambling to shrink the carbon footprint of natural-gas production. But, to make natural gas climate-neutral, it will have to do far more.

Pressure is intensifying to curb greenhouse-gas emissions from all links in the LNG business chain: from the wells that produce natural gas, to the pipelines that transport it, to the plants that liquefy it for export, to the power plants that burn it. Methane emissions are sparking a particular backlash, because they are such an egregious and, at least in theory, fixable affront. Governments and corporations around the globe have pledged to cut their carbon footprints to “net zero” by midcentury, the target that science dictates is crucial for the climate, and they are increasingly aware that LNG, whose production leaks lots of methane, may render their promises little more than hot air. Over the past two years, some 130 countries have signed an international pledge to reduce their methane emissions by 30% by 2030. According to the United Nations Environment Programme, global methane emissions from human activity must roughly halve from their 2020 level by 2050 for the world to align with a net-zero path — and about one-quarter of such emissions come from the oil and gas industry.

All of this explains why regulators and some of the biggest foreign buyers of U.S. LNG are pushing the U.S. LNG industry to plug its methane leaks. The Biden administration, as part of the Inflation Reduction Act, the mammoth spending package passed last year, is moving to slap a fee on companies whose gas the government deems to have been produced with too high a methane-leakage rate. LNG buyers, meanwhile, are pressuring U.S. producers to curb their methane problem — and to substantiate that they’ve done it in a way that holds up with investors. To be sure, for the past year, with global supply of natural gas squeezed by Russia’s war in Ukraine, the U.S. industry has had ready buyers for basically all the LNG it can produce, carbon content be damned. But the industry sees market concern about LNG’s climate impacts as an inevitable long-term structural shift — and thus as an existential threat. In response, it is peddling a new commodity it hopes will register as credible with the market but won’t crimp its profits: ostensibly green gas.

Seemingly weekly, a major U.S. industry player — a gas producer, a gas-pipeline operator, an LNG exporter — announces a claim that some of the gas from which it makes money has been produced in a way that’s extra-climate-conscious. At least eight of the 10 biggest producers of U.S. natural gas — among them ExxonMobil, BP, Chesapeake Energy and Southwestern Energy — have made green-gas claims. S&P estimates that about one-quarter of all the natural gas produced in the U.S. is now being sold with some sort of certification of its carbon content, up from essentially zero a few years ago.

Few LNG companies are gunning harder to justify their green-gas claims than Houston-based Cheniere Energy, which owns and operates the Sabine Pass LNG liquefaction plant. Seven years ago, it became the first U.S. company to export LNG; since then, it has ridden the LNG-export business to riches. Today it operates two LNG plants — the one at Sabine Pass and another down the Gulf coast in Corpus Christi, Texas — which together accounted for 55% of all U.S. LNG exports in 2021. Cheniere’s shares, which go by the bumptious ticker symbol LNG, today are worth six times what they were when the company began exporting LNG in early 2016. “Future So Bright, Have to Wear Shades,” gushed the title of one JPMorgan report to investors about Cheniere in March 2022, the month after Russia’s invasion of Ukraine, which has sent LNG prices soaring — and, with them, consumer energy costs and the value of Cheniere shares.

Last year, bending to market pressure to fix methane leaks across the vast U.S. natural-gas-production network that feeds its liquefaction operation, Cheniere began providing the majority of its LNG customers what it calls a “cargo-emissions tag” — a chit the company says quantifies the carbon footprint of each load’s production and shipment to its foreign destination, calculated according to a methodology whose development Cheniere has funded and coordinated. Cheniere says that footprint is markedly smaller than prior research has found for U.S. LNG as a whole. But its methodology has sparked scientific debate, and Cheniere-funded research has noted “limitations” of the data on which its calculations are based.

Cheniere’s chits are part of a burgeoning green-gas market that is of questionable environmental legitimacy. The federal government, in preparation for its planned fee on gas whose production leaks large concentrations of methane, is just launching a process to set rules for how to measure and calculate those methane-leak rates. In the meantime, the new green-gas market rests on a crazy quilt of competing claims that involve differing methodologies, all of which are evolving. Some embody assertions that the producer or buyer of the LNG has bought “offsets” in a project elsewhere that is said to cancel out the greenhouse gas emitted by the production of the LNG. But offsets are themselves controversial, largely because of questions about the environmental credibility of the standards on which some are based. Other green-gas labels assert that the commodity has been produced in a way that has kept methane leaks or carbon-dioxide emissions to a supposedly verifiably low number. But the companies that sell such gas, as well as a raft of other businesses springing up to sell the gas industry their green-gas-certification services, disagree among themselves about their methods to measure methane leaks, about their standards to judge whether leaks are plugged, and about their calculations to justify their green-gas labels.

All of this, so far, is voluntary. Despite efforts by some industry and outside groups to cobble together uniform practices, companies can — and do — pick and choose among different green-gas-certification regimes. The result is that U.S. gas is now sold in many shades of “green.” Different companies are selling purportedly green gas with differing carbon footprints. Some companies, such as Chesapeake, are certifying different batches of their own gas with differing green labels. As the corporate green-gas claims multiply, the confusion mounts. Absent regulation, judging the legitimacy of individual companies’ green-gas claims, let alone the environmental efficacy of the whole green-gas market, is all but impossible. The green-gas market at this point is a Wild West frontier filled with mudslinging and smoke and mirrors — and the gunslingers are deciding the rules as they go along. At stake in the fracas is the climate.

For years, the natural-gas industry has marketed its juice as fossil-fuel lite: reliable, like coal, but cleaner. Over the past decade, a number of scientific studies on methane emissions have cast serious doubt on these claims. Computer modeling and, more recently, measurements from an array of high-tech devices, including towers, drones, airplanes and satellites, have revealed that methane is leaking from basically every crevice of the gas system. The research indicates the leaks are worse in certain parts of the U.S., places rich not just in gas but also in oil — notably the Permian Basin, the region in Texas and New Mexico that is the country’s biggest producer of oil and its second-biggest producer of natural gas. There and elsewhere, a revolution over the past decade in fracking and in other techniques to produce fossil fuel from previously tough-to-puncture rocks known as shale has spurred pedal-to-the-metal production techniques that have strained infrastructure, including storage tanks, processing plants and pipelines. That production frenzy also has worsened methane leaks. A number of studies suggest the leaks are so massive that some U.S. gas may not, after all, be much climate-friendlier than coal — and that it certainly isn’t climate-friendly enough to lessen the need for solar, wind and other renewable energy.

Cheniere was founded in 1996 and opened its Sabine Pass facility in 2008 to import LNG from other countries. Its plan was to unload LNG from ships; warm the chilled liquid back into a gas, a process known as “regasification”; and send the gas through pipelines to customers around the country. The company’s name is the Cajun-French term for a type of sandy ridge, sometimes also spelled as “chenier,” that’s common along the coast near Sabine Pass. The word comes from “chêne,” French for “oak,” a tree often found on those ridges.

Soon after the Sabine Pass regasification plant began operating, the U.S. fracking revolution swung into high gear, transforming the United States from a net importer of gas to a net exporter. In 2012, Cheniere became the first company to get federal approval to build a facility to liquefy U.S. gas and export it. In February 2016, Cheniere exported from Sabine Pass the first cargo of made-in-America LNG. Since then, as U.S. LNG production has soared, Cheniere has remained the undisputed liquefaction leader. It has added more lines of gas compressors at the Sabine Pass plant, which has gotten so big that, in 2021, the latest year for which statistics are available, it was the eighth-largest greenhouse-gas emitting facility in the country, excluding power plants, according to the U.S. Environmental Protection Agency. Cheniere’s LNG-export terminal in Corpus Christi accounted for 20% of all U.S. LNG exports in 2021.

Disagreements about U.S. LNG’s environmental impact have dogged the industry since its inception. In 2014, as part of a U.S. Department of Energy review of applications by Cheniere and then other companies to export U.S. LNG, the National Energy Technology Laboratory, a federal lab, published a “life-cycle analysis” of the carbon footprint of U.S. LNG — an accounting of its climate impact from the gas’s production at the well through its consumption abroad for electricity production. Over the next few years, as U.S. gas production ramped up, scientists began publishing studies that identified methane emissions as a major climate problem, and the federal government began moving to regulate methane emissions from oil-and-gas production. Cheniere’s LNG exports were surging, and the company recognized carbon concerns as an increasingly important determinant of the future demand for gas, so it launched an internal effort to assess its LNG’s carbon footprint. Fiji George, the gas-industry veteran whom Cheniere hired to lead the effort, told me the intent was to develop a method of calculating LNG’s carbon footprint that was “bespoke to Cheniere’s supply chain.”

Cheniere’s study still was underway when, in 2020, the stakes of methane leaks for the U.S. gas industry jumped. Though President Donald Trump’s EPA had rolled back government regulations of oil-and-gas methane emissions that August, Democrat Joe Biden made clear on the campaign trail his intent to restore and toughen those rules if elected. Beyond the political arena, meanwhile, the market was issuing a more immediate threat.

Engie, a major French energy company that both trades gas and buys it to fuel gas-fired power plants that Engie operates in Europe, had come under increasing pressure from European environmental activists not to buy U.S. shale gas, largely due to concerns about methane leaks, especially from the Permian Basin. In the summer of 2020, Engie’s board quietly put on ice an LNG-import deal that Engie had been negotiating with NextDecade, a Texas-based company building an LNG-export terminal in Brownsville, Texas. When, starting in late October, the media began reporting that Engie might back away from the NextDecade deal due to climate concerns, U.S. LNG companies did a double take. Ken Robinson, a veteran U.S. gas executive who had worked as chief corporate risk officer for Engie in Paris before moving back to Houston in March 2020 to head up the company’s U.S. energy-marketing business, told me the U.S. LNG industry saw Engie’s pullback as “writing on the wall.”

On Nov. 6, 2020, just days after Engie’s rejection of the NextDecade deal became public — and, as it happened, the day before the presidential election widely was called for Biden — Cheniere executives held a routine call with investment analysts to discuss the company’s latest quarterly earnings. On prior earnings calls, their comments about LNG’s environmental profile had been bullish. Anatol Feygin, Cheniere’s chief commercial officer, had asserted that a rising global focus on climate change, as evidenced by environmental policies in Europe and China, was good for Cheniere, because LNG was cleaner-burning than coal — and so, under carbon prices, was economically advantaged. This time, without naming the Engie deal, Jack Fusco, Cheniere’s chief executive officer, cited “an increasing number of questions from our shareholders” about LNG’s carbon footprint. Cheniere, he said, was “methodically reviewing our business to quantify our life-cycle emissions and to identify and analyze climate-related opportunities across our value chain,” moves that had “a potential to not only positively impact the environmental profile of Cheniere and our LNG, but also the industry at large.”

At Engie, Robinson and his U.S. team started getting calls from U.S. producers who had read in the press about Engie’s pullback from a NextDecade deal and wanted to know what they could do to ensure Engie would continue buying their gas. “‘What if I certify it to you as being low-methane, low-carbon, whatever?’” Robinson recalls one company asking him. Robinson’s response: “What’s ‘certify’? What does that mean?”

Robinson started investigating. His research soon led him to a number of new outfits, including startup Project Canary and not-for-profit MiQ. The two operations recently had sprung up to hire themselves out to companies across the natural-gas trade as certifiers of the environmental impact of specific quantities of gas — overseeing the measurement, quantification and verification, based on methodologies and technologies they were developing, of the methane emitted in the production of that gas. (Project Canary is, despite the name, unrelated to Canary Media. One of MiQ’s creators is RMI, a think tank that helped launch Canary Media and that continues to provide Canary Media funding and administrative support. Canary Media stipulates in its founding documents that it is editorially independent from RMI.)

The goal of both Project Canary and MiQ is to establish a market for gas certified as low-carbon — what many in the industry dub “responsibly sourced gas.” That gas, judged and labeled as to its carbon content, would cost somewhat more to produce. But, the theory went, it would command a commensurately higher price in an increasingly climate-focused market — raising the wholesale price of a thousand cubic feet of gas, now about $7, by perhaps a few cents. The question hanging over the prospective market was how much of a premium buyers would be willing to pay for greener gas — a question that, in turn, would depend on the extent to which sellers would be able to shift the cost to consumers. “Responsibly sourced” would be to gas what “organic” is to vegetables.

Engie had begun pursuing contracts for responsibly sourced gas when, in early 2021, it was negotiating a long-term contract to buy LNG from Cheniere. In those talks, said Robinson, the Engie executive, the French utility made clear that, to win the deal, Cheniere would have “to take steps to reduce their carbon footprint.” Cheniere’s response? “They committed to it,” Robinson said — but on Cheniere’s terms, based on Cheniere’s views about the best way to credibly count and validate methane emissions.

Cheniere, Robinson recalled, didn’t agree to subject the gas it would buy and liquefy for Engie to low-carbon certification by a third party — the sort of certifications Robinson was seeking in Engie’s other contracts for U.S. LNG. “They believe their approach is better than everybody else’s,” he explained of Cheniere’s methane-counting methodology. “I don’t necessarily agree with them,” he added, but Cheniere executives “at least are taking actions that a year ago, a year and a half ago, they weren’t as focused on.”

When I asked Cheniere about the suggestion that it had, in its contract talks with Engie, resisted promising to buy gas certified as low-carbon, Eben Burnham-Snyder, Cheniere’s vice president for public affairs, at first responded in an email that that version of events was “not factual,” without offering specifics. He also raised “potential conflicts of Canary Media,” citing Canary Media’s affiliation with RMI and RMI’s backing of MiQ. When I noted in reply that the description of Cheniere’s contract discussions with Engie had come from Engie, Burnham-Snyder emailed back: “I won’t have any further responses. We don’t comment on commercial negotiations.”

As the negotiations between Cheniere and Engie were underway in early 2021, the newly installed Biden administration was pressing forward with its effort to regulate methane emissions from the oil-and-gas patch. In June 2021, the same month Cheniere and Engie inked an LNG contract, President Biden, implementing one of his stated campaign priorities, signed into law a policy rescinding the Trump prohibition on methane limits, paving the way for a regulatory crackdown.

Engie’s Robinson speculated that the reason Cheniere chose to develop its own approach to measuring and quantifying methane emissions rather than accepting the approach of another entity, such as Project Canary, is that, as a dominant player in the U.S. gas industry, it figured it can “have a bigger say in how the market moves forward.” He added: “I think they want to genuinely build something that’s better than anybody else has.”

Robinson said he believes Cheniere is also concerned that its margins might suffer if the market shifted to natural gas whose carbon footprint had to be certified by an outside entity. Under the contracts that Cheniere has with the vast majority of its customers, the customer agrees, often over a period of 20 years, to pay Cheniere a price for LNG that’s indexed at a set percentage above a prevailing natural-gas-price marker — typically, the U.S. marker known as the “Henry hub” price — plus a set fee for Cheniere’s liquefaction services. If the LNG market shifted to LNG produced from gas certified by an outside party as low-carbon, he suggested, that might raise the price that Cheniere had to pay for the gas in a way that would make it difficult for Cheniere to pass that price increase to its customers. “Low-methane gas comes at a premium. They didn’t want to pay a premium,” Robinson said, “because they’ve already sold their LNG and they would be squeezed.”

Burnham-Snyder, the Cheniere spokesperson, rejected the suggestion that Cheniere’s business might suffer if the company had to pay more for gas externally certified as to its carbon content. He explained by email that, under Cheniere’s so-called “cost-plus” contract model, if a customer specifies that it wants gas that costs Cheniere more to buy — say, because that gas is certified as low-carbon — Cheniere raises, by a commensurate amount, the price it charges the customer for that LNG, “so we keep our basic margin.” Similarly, Burhnam-Snyder wrote, if broader market pressure or government rules for gas certified as low-carbon by an outside entity pushed up the price marker to which Cheniere’s already-signed contracts had been indexed, then the price Cheniere’s customers paid Cheniere for the LNG would rise along with the marker, keeping Cheniere’s margins intact. It would be “a pass-through cost to the customer as part of our contract structure,” he explained. He did not address what would happen in the event that such broad forces pushed Cheniere to buy gas externally certified for its carbon content at a price premium that was not reflected in the price marker to which the contracts had been indexed.

The combined pressure of regulators and a market ever more focused on runaway methane emissions has the gas industry roiling over the question of how to measure and quantify the pollution it generates. Traditionally, assessments of methane leakage in the U.S. oil-and-gas industry have been based on EPA estimates — the same ones used in the 2014 study that attempted to quantify the life-cycle emissions of U.S. LNG. Several studies since then have concluded that those estimates have underestimated emissions from natural-gas production in several important ways.

An influential scientific paper initially released in November 2020 calculated that actual methane leaks are as much as twice as high as the federal estimates indicate. Those federal estimates have a few foundational problems, said Arvind Ravikumar, a professor at the University of Texas at Austin who was one of the paper’s authors and who has spent years developing research expertise in methane-leak detection, working in the process with parties across the ideological gamut, from green groups such as the Environmental Defense Fund to companies in the gas industry including Cheniere.

The federal estimates are based on comparatively few, and dated, measurements — samplings taken from various oil-and-gas-field equipment in the 1990s and then extrapolated into national averages. That was before the shale boom dramatically increased the number of pieces of equipment in use and the number that were “super-emitters,” equipment whose emissions, either because of operator error or because of intense use, vastly exceeded the national averages.

As a result, the federal estimates appear particularly to be underestimating methane emissions from two parts of the gas-production process: storage tanks and “flaring,” the process of burning off excess natural gas in the field. Tank emissions were underestimated because the measurement technology available a quarter-century ago was inadequate and also because far fewer tanks were in use before the shale boom than are now. Flaring was underestimated, especially in basins such as the Permian, both because the shale boom increased its occurrence and because the prior data had failed to account for the number of flares whose flames had flickered out. Unlit, flares suddenly were becoming super-emitters; they were releasing not carbon dioxide from combustion but, in a far bigger hit to the climate, unburned methane.

Fusco, Cheniere’s CEO, has offered a different critique of the federal numbers: that they overestimate the methane problem. In a February 2021 earnings call, he told analysts that the 2014 study’s emissions estimates were “quite a bit higher than what we think our actual carbon emissions” at Cheniere are. He explained that in 2022 Cheniere planned to begin providing the cargo-emissions tags to buyers of its LNG. The tags “will be a carbon footprint of what the carbon — the CO2-equivalent offsets need to be for that carbon to make it carbon-neutral, OK?” Fusco said in the call, hours after Cheniere issued a press release announcing its plans to distribute them. “And we hope, over time, that we’re going to get better and better at these tags,” Fusco told the analysts. “And then our customers will be able to see more transparently how we’re calculating it and what’s all encompassed in it.”

Cheniere’s next quarterly earnings call was in early May 2021, by which point the Biden administration’s preparations to regulate oil-and-gas methane emissions were well underway. On the call, Fusco underscored how important it was to him that Cheniere unveil its long-running life-cycle analysis of its LNG. Cheniere’s weighing in was important to the company’s business, he said on the call, “because there’s a lot of disinformation out there” over how to compute LNG’s carbon footprint. “And there is not a very good process for defining terms that everyone agrees to.”

Two weeks later, in mid-May, the authors of the Cheniere-funded study submitted it to an American Chemical Society journal for peer review. On August 3, 2021, the journal published the paper. Two days after that, on Aug. 5, 2021, Cheniere issued a press release heralding the study as a “first-of-its-kind analysis” of greenhouse gas, or GHG, emissions from natural gas. Using “emissions data specific to Cheniere’s LNG supply chain,” it said, the study finds “a lower GHG intensity for Cheniere’s LNG” than prior studies had estimated for U.S. LNG as a whole.

Peer-reviewed journals typically require authors to note financial conflicts. “This work was funded by Cheniere Energy Inc.,” the paper disclosed. The paper’s nine authors were George, who directs Cheniere’s climate and sustainability efforts; employees of a consulting firm hired by Cheniere; and academics at Duke University, the University of Texas at Austin and Queen Mary University of London. The paper noted that seven of them — all but the two from Duke — had current or prior financial relationships with Cheniere. Some of those seven, the paper reported, also had current or prior financial relationships with other oil- and gas-industry companies or groups, or with environmental groups, or with both.

The paper’s headline conclusion was that the life-cycle emissions for LNG shipped from Sabine Pass in 2018, the plant and year it had studied, were about one-third less than the prior federal studies had calculated for all U.S. LNG. Underlying that conclusion, the paper explained, was the calculation that just 0.65% of the methane destined for Sabine Pass leaked during its journey up from the ground through the production well and all the way to its liquefaction at the Cheniere plant.

Multiple studies of U.S. LNG produced elsewhere have estimated the methane-leakage rate to be much higher than that. The 2020 study whose authors included Ravikumar found a methane-leakage rate of 1.3% for natural-gas production — slightly more than twice the rate the 2021 Cheniere-funded study calculated, given that the Cheniere-funded study, unlike the 2020 paper, included leaks not just from production but also from liquefaction. As it happened, the peer-reviewed version of the 2020 study was published, in Nature Communications, on Aug. 5, 2021, two days after the Cheniere-funded study was published.

The key to the conclusion by the Cheniere-funded paper that the company’s LNG had a lower carbon footprint was the methodology the authors had developed to calculate emissions. The paper relied on the same underlying government model used in both the 2014 federal study and a 2019 update of it. But the authors of the Cheniere-funded study made a number of “improvements” to the model and to the data underlying it — to “reflect the current state of the science” and “to better represent Cheniere’s SPL supply chain,” they explained in the paper, using an abbreviation for Sabine Pass Liquefaction, the big LNG plant. Their modifications sought to better account for methane leaks from storage tanks and from flaring during the gas’s production, for example, and to reflect the actual machinery at the Sabine Pass plant and the types of ships and the shipping routes on which the Sabine Pass LNG traveled abroad. “This work demonstrates that the use of current supplier-specific data is vital for accurately accounting for the variability” of natural-gas emissions, they wrote.

The Cheniere-funded paper also revealed several caveats about gaps inherent in the authors’ analysis. A 117-page methodological annex that accompanied the paper provided insight into them. For instance, Cheniere bought 42% of the gas it liquefied at Sabine Pass not from specific producers but from “traders who cannot reveal their exact sources,” clouding efforts to assess that gas’s carbon intensity. In addition, the paper noted “limitations” in the authors’ ability to compare the findings of prior LNG life-cycle-analysis studies, which assessed gas based on the basin from which it came, with their analysis, which examined gas based on the operators within the Sabine Pass facility’s LNG network. The paper explained tweaks the authors made to try to account for various shortcomings.

To assess how much their modeling assumptions affected their results, the authors had conducted a “sensitivity analysis.” The details of that analysis consumed eight pages of the paper’s explanatory annex, and today they stand as a tour of the methodological minefield that is the effort to calculate oil-and-gas methane leaks. The annex explored how the study’s calculated methane-leakage rate, and thus its estimates of the size of the carbon footprint of LNG from Cheniere’s Sabine Pass plant, might have been significantly higher had the authors made different choices about their model and their data. In one example, the annex explained, when the authors modeled the “uncertainty distribution” surrounding the estimates of leak rates from equipment — the rates that underlie the model — the exercise increased the study’s estimate of the carbon footprint of Sabine Pass LNG, from its production in the U.S. through its rewarming to a gas in the country that imports it, by as much as 37%.

Another example, the annex noted, concerned the provenance of the gas that’s liquefied at Cheniere’s Sabine Pass plant. On one hand, the authors reported that they had updated the federal data used in the 2014 study to reflect the higher methane-emission levels in the Permian Basin that studies since 2014 had found. On the other hand, those updates “did not have a significant impact” on the Cheniere-funded study’s determination of the carbon footprint of Sabine Pass LNG; as the main paper noted, the Sabine Pass facility “had relatively few Permian Basin suppliers” in 2018. But the paper’s annex noted an important caveat: “the magnitude of the impact would depend on the volumes of gas purchased from the Permian region and the amount of gas being flared by the specific suppliers.” As a result, Cheniere “will continue to investigate this emission source in the future.”

Three months after the paper was published, Fusco, the CEO, told analysts in a November 2021 quarterly earnings call that “about 100% of the gas at Corpus Christi right now is coming out of the Permian Basin.” In other words, in the wake of the Cheniere-funded study, which had stipulated that LNG plants using more Permian gas than the Sabine Pass facility was using might have higher carbon footprints than the study had posited, Fusco noted that Cheniere’s other LNG plant, in Corpus Christi, was processing primarily Permian gas.

In recent months, I had several conversations with David Lyon, who until earlier this month worked as a methane-leak expert at the Environmental Defense Fund. Lyon was another of the co-authors of the 2020 paper with Ravikumar that calculated a methane-leakage rate roughly twice the level the Cheniere-funded paper later found. In one of our discussions, Lyon told me that Cheniere’s effort is “on the forefront” of scientifically serious attempts to quantify and curb natural gas’s carbon footprint. He called Cheniere’s academic study “rigorous” and “really good.” But, said Lyon, the Cheniere-funded study’s use of a methane-leakage rate of only 0.65% represents a “major flaw.” As a result, “I think they’re drastically underestimating the upstream emissions,” he said of Cheniere. The solution: “They just need to replace that assumption with measurement-based estimates.”

Two other environmental groups, Greenpeace and Oil Change International, have criticized the Cheniere-funded study more deeply. In August, they published a report dismissing Cheniere’s cargo-emissions tags and the life-cycle-analysis methodology on which they are based as “greenwashing.” Among other criticisms, they argued that the Cheniere-funded study’s focus on the Sabine Pass plant elided the issue of Permian-heavy gas liquefied at Cheniere’s Corpus Christi facility. The carbon footprint of Cheniere’s Corpus Christi LNG, they contend, “could be very large.”

Last October, their criticisms sparked an epistolary duel in the pages of the American Chemical Society journal that had published the Cheniere-funded study the previous year. The journal published a formal critique written by two of the authors of the Greenpeace and Oil Change International report. Elaborating on the criticisms they had first lodged in their report, they noted that much of the gas liquefied in Sabine Pass came from marketers, meaning that “the true contribution of Permian Basin suppliers to the Sabine Pass is likely unknown.” And they criticized Cheniere for using the Sabine Pass study as the basis for carbon tags it’s issuing even for gas it liquefies at Corpus Christi, “where a large majority of gas is sourced from the Permian.” The critics argued that “such a methodology is not an appropriate baseline to begin certifying LNG shipments, much less combining those certifications with offsets to position LNG as ‘low-carbon’, as Cheniere executives have touted.”

The same day the journal published that critique, it also published a response by the authors of the original Cheniere-funded study. Those authors reiterated their paper’s assertion that the federal estimates, though they have “limitations,” are “currently the best available data source” for assessing emissions from the production of gas, and that the updated methodology the authors had designed to calculate the carbon footprint of LNG would become “more robust as emissions estimates improve and supplier-specific data are more available.” They also stressed that, for all the debate over how best to count methane emissions from gas production, that production constitutes only a small slice of gas’s full carbon footprint from the well where it’s produced to the power plant where it’s burned.

As for Fusco’s statement to analysts that the gas being liquefied at the Corpus Christi plant was essentially all from the Permian Basin, the authors of the Cheniere-funded paper declined to address it. They also declined to address the criticism that the Sabine Pass study was too flawed to justify Cheniere’s using it as the basis for cargo-emissions tags it’s issuing to its customers. The authors of the Cheniere-funded paper wrote that “statements made by executives at Cheniere Energy Inc.,” as well as the Greenpeace and Oil Change International authors’ “concern on use of the model for life-cycle emissions certification by Cheniere,” were “outside the scope of our paper.”

Over the past year, Cheniere has begun taking measurements to produce more-accurate data. It has gathered a team of consultants and academics to suss out and quantify actual methane leaks in the Cheniere system — at the fields of several U.S. producers from which it buys gas, in the pipelines of companies that transport the gas to its Sabine Pass liquefaction plant, and at that plant. I asked several of those companies to let me visit. All either rejected my requests or failed to reply to them. One company, Crestwood Equity Partners, which operates gas-processing facilities and gas pipelines, said no because, a company spokesperson told me in an email, Crestwood had signed a confidentiality agreement with Cheniere.

The team Cheniere is funding to conduct that monitoring is deploying sensors on towers installed at gas facilities, and on drones, planes and satellites that monitor such facilities from above. The equipment is operated by firms offering methane-measurement services — a business that has ballooned over the past decade, profiting from the push to hunt down plumes of the planet-heating gas. Ravikumar, the UT-Austin professor who specializes in methane measurement, is among the scientists Cheniere has funded to help with the program, though Ravikumar was not among the authors of the 2021 Cheniere-funded carbon-footprint study. What’s necessary, Ravikumar told me when we first talked several months ago, is “high-fidelity, continuous measurement” of methane — setting up a web of devices to measure emissions frequently. That’s important because leaks, including very large ones, can start and stop unpredictably. Yet even an array of whiz-bang equipment is only as good as the techniques used to deploy it and to analyze the data it collects. The field of methane measurement is young, with myriad players experimenting with myriad techniques. “It’s not yet clear who is going to be the most effective,” Ravikumar said. “It’s all trial and error and ‘Let’s see what works.’”

Continuous measurement of methane emissions using the latest gear is what Project Canary, one of the dominant new certifiers of the carbon footprint of gas, views as its competitive edge. That’s why Engie decided in 2021 to favor gas certified by Project Canary when purchasing U.S. LNG.

Denver-based Project Canary’s chief executive is Chris Romer, a voluble man who previously achieved financial success in corporate educational services and concluded a few years ago that green gas was going to be big. At the outset of our interview, over Zoom, he volunteered to me that he’s the son of Roy Romer, a former Colorado governor; the brother of Paul Romer, a former World Bank chief economist and a New York University professor who won a Nobel Prize in economics in 2018 for his work illuminating the role of government policy in fostering technological innovation; and a proud product of Colorado, “a blue state with a shit-ton of carbon next to homes.” The upshot of that background: “I’m a Democrat for more LNG.”

Romer said he talks often to Cheniere executives. Citing the growing list of deals by Engie and other buyers for LNG that Project Canary has certified, Romer asserted that his company presents a potential threat to Cheniere’s selling of gas whose carbon footprint it has estimated using Cheniere’s own methodology. “I think they’re afraid we could screw up their business model,” he said.

I asked Cheniere’s Burnham-Snyder how the company would compare the design and robustness of its methane-assessment method with those of Project Canary and MiQ. He emailed back that “we are not going to do a comparison.” Cheniere, he added, is “proud of the industry-leading, peer-reviewed work we have done through our life-cycle assessment” and emission-measurement campaign “to better measure and understand GHG emissions across the LNG supply chain, along with looking for opportunities for emissions reductions.”

Project Canary, too, has faced criticism over the environmental credibility of its methods. One concern is that Project Canary makes money by selling not just certifications but also the measurement equipment used to collect the data on which those certifications are based — a potential conflict of interest. “A lot of people have argued, ‘Aren’t you just trying to sell your hardware?’” Romer told me. In response to those concerns, he said, Project Canary recently decided to sell certification services even to customers that buy certification equipment from other vendors. Its core product, he said, is not the hardware but “the data.”

Another criticism of Project Canary is that its certification is easily gamed. The company lets firms in the gas industry seek its blessing for any subset of their gas they choose: one well out of 100, or three wells out of 1,000, or whatever. A company may hire Project Canary to certify a well that the company is pretty sure is an environmental top performer — and may avoid subjecting to similar scrutiny other wells that it’s pretty sure are not. “This is a legitimate criticism of Project Canary: Can companies just cherry-pick their best pads?” Romer told me. He contends that smart companies will “avoid the moral hazard of cherry-picking.” But, he added, “I don’t have the ability to mandate that people do 100 percent” of the gas in their system.

Engie over the past year has signed deals to buy gas from several U.S. firms, Robinson told me, after those firms agreed to subject themselves to Project Canary certifications. One is NextDecade, the Texas producer whose LNG Engie previously rejected as too dirty. Given the methane-based scarlet letter that burdens U.S. shale in the view of France and much of Europe, Robinson said, “I had to find a way to make it more palatable” to them. Project Canary certification has helped make Engie “comfortable” with the environmental profile of the U.S. gas it is buying to ship back home to burn.

MiQ, the other major green-gas arbiter, assesses its customers’ emissions across broad gas-operating areas rather than at individual facilities such as wells. MiQ created a methodology for quantifying methane intensity that it uses to slot gas from each area into one of six tiers: A, B, C, D, E and F. “But we’re not drawing a firm line” as to which tiers are green enough and which aren’t, said Georges Tijbosch, MiQ’s chief executive, who hopes the market will do that by paying more for the gas at the cleaner levels. MiQ has become the certifier of choice for a handful of big gas producers, including BP and ExxonMobil.

Tijbosch criticized rival Project Canary for its emphasis on continuous monitoring, which he suggested is a red herring because, in the field, technology “doesn’t always work right.” MiQ requires companies seeking its certification to undergo periodic, but in most cases not constant, emissions monitoring. It then uses a proprietary algorithm to categorize that gas into one of its six grades of methane intensity. The reasoning, Tijbosch said, is that MiQ’s goal is less to measure emissions with pinpoint accuracy than to push companies to cut those emissions. “We’ve got to create these competitive tensions in the markets,” he said, “so people start to abate.”

Project Canary’s Romer said MiQ would benefit from “a little more measurement.” He also told me, even as he took issue with aspects of Cheniere’s and MiQ’s methods, that he’s concerned the fighting among rivals threatens to delegitimize the fledgling green-gas sector. “Sometimes we’re jabbing MiQ. Sometimes they’re jabbing us. I’d like to stop that,” he said. The reason for all the jabbing? “It’s Moneyball,” he said, and then he likened the green-gas scrum to “a bunch of Salomon Brothers traders trying to stand up a market” — referencing two popular Michael Lewis books to make his point.

Some companies appear to be shopping the market, certifying some of their gas through Project Canary and some through MiQ. Chesapeake, among the biggest gas producers in the United States, chose MiQ to certify gas Chesapeake is producing between July 2022 and June 2023 in the Marcellus Shale, a geologic formation in Pennsylvania. But Chesapeake chose Project Canary in April 2021 to certify gas from certain Chesapeake wells in the Marcellus and in Texas’ Haynesville Shale. Tijbosch, MiQ’s chief executive, who’s a former energy trader at such companies as BP and Goldman Sachs, said MiQ required Chesapeake to exclude from MiQ certification the specific Marcellus gas that Chesapeake previously had certified with Project Canary, to avoid “the whole double-counting thing.” Chesapeake did not respond to requests for comment about why it certifies gas with different entities.

Cheniere, for its part, is determined to press ahead with its carbon tags, despite questions about the methodology underpinning them. Achieving “super, super-correct” emissions measurements is less important than understanding broad trends that can inform action, George, who directs Cheniere’s climate and sustainability efforts, told me in a Zoom interview from a nondescript gray conference room in the company’s downtown Houston headquarters. “We want to understand where the lowest-cost options are, and what can be done now and what can be done fastest.”

George is widely regarded as a national expert in the quickly evolving discipline of quantifying natural gas’s carbon footprint. He heads a subgroup examining such methodologies for a gas-decarbonization study that the National Petroleum Council, a body that advises the U.S. Department of Energy, is conducting at the request of Energy Secretary Jennifer Granholm. In the interview with me, George stressed what Cheniere’s effort isn’t doing as much as he discussed what it is doing, as if to minimize any threat that Cheniere’s carbon-counting effort might present to its suppliers or customers. Cheniere sources the massive amounts of gas it buys from around the country, and George made clear that the company doesn’t intend to wield its carbon calculations as a knife to cut off gas-purchase contracts from certain producers or from certain regions based on that gas’s carbon content. The carbon-tag effort presents “information for our customers to decide to use any way they want to,” he said. “It is not a certification program. We’re not giving a grade to anybody. We are not making purchase decisions based on the output. I need to emphasize that.”

Also present for the interview was Robert Fee, Cheniere’s chief of staff. Fee, who before coming to Cheniere worked for the U.S. Department of Energy’s fossil-fuel office, underscored to me that, of the approximately 30 million tons of annual U.S. LNG exports for which long-term contracts were signed in 2022, only about 1 million tons had climate-related terms attached to their agreements. To be sure, he said, “certain buyers do care,” and more “will care tomorrow,” which is why Cheniere’s leaders “want to make sure that we’re getting ahead.” But in the vast majority of LNG contracts, “climate is not a primary consideration, or a secondary consideration, for these buyers,” he said. Cheniere sells to 30 customers around the world; particularly with the global LNG market tight from the lack of Russian gas, Fee said, most of them have concerns they see as more immediate: “Price, availability of supply, commercial terms — these things are still paramount.”

In recent months, the fight over how to assess oil-and-gas methane emissions has intensified. Last August, Biden signed into law the Inflation Reduction Act, another turn of the regulatory screw against such pollution. That massive legislative package included a measure that lays the groundwork for the federal government to start slapping, in 2024, a fee on gas producers whose methane emissions are found to exceed a certain percentage of their total gas production. The law tasks the EPA with resolving between now and then a question crucial to the fortunes of Cheniere and other companies in the U.S. LNG business: how to measure emissions in a way that is accurate and environmentally credible, a determination that will shape the government’s ongoing decisions about whether a particular measurement methodology that a company proposes passes muster.

“That’s the challenge,” said Lyon, the former Environmental Defense Fund official. “How do you do a standard that says, ‘This method is good enough,’ and ‘This method is bullshit, and I’m not going to allow it?’” Lyon told me this in mid-January, about a week before his last day at the environmental group. As of Jan. 29, he works for the EPA — as an adviser on the federal effort to hash out which sorts of methane-measurement methodologies ought to be deemed acceptable.

As that approach is being developed, Cheniere continues to work to cement the influence of its methane-measuring methodology. In October, it announced it was joining a United Nations Environment Programme group dubbed the Oil and Gas Methane Partnership 2.0, which seeks to standardize practices on the reporting and reduction of methane emissions. On Jan. 10, the University of Texas at Austin announced a new $50 million research program to be housed at the university to improve ways to quantify the oil-and-gas industry’s emissions of methane and other greenhouse gases. The initial donors are Cheniere; EQT Corp., the nation’s biggest natural-gas producer; and the Williams Companies, a major U.S. gas-pipeline operator. One of the co-directors of the program, which also includes researchers from other universities, is Ravikumar, the UT-Austin professor.

Eight days after the announcement of the program’s creation, Cheniere submitted formal comments to the EPA weighing in on the approach to measuring and quantifying methane emissions that the agency is developing. In dry legalese, Cheniere’s letter nodded at the competition that has broken out among aspiring methane-measurement regimens. “Creating a like-to-like comparison between facilities or operators in an evolving environment of measurement technologies is challenging,” wrote Cheniere, which then, in the next paragraph, suggested a solution: data “compiled by credible third-party experts,” particularly the new UT program, which Cheniere’s letter singled out by name but did not mention is funded in part by Cheniere.

Looming over the industry’s bid to peddle greener gas is the question of whether it’s a diversion. Even if the industry’s green-gas claims proved credible as far as they went, they would have to go much further to spare humanity the full fury of climate change. The atmospheric imperative is to zero out LNG’s greenhouse-gas emissions. That encompasses not just methane released during LNG’s production in the U.S., but also the far-greater cumulative climate impact of the carbon dioxide sent into the air when the U.S. LNG is burned around the world.

To critics like Greenpeace and Oil Change International, natural gas, a hydrocarbon, is, as a fact of chemistry, incompatible with decarbonization. They contend the only responsible thing to do with it is to leave it in the ground. “While methane emissions reduction is fundamental to climate action,” their August report declares, “it must go hand-in-hand with winding down the fossil fuel industry, including the supply of gas and LNG.”

But the world isn’t buying the don’t-drill argument — at least not at a pace and scale sufficient to avert the most disastrous effects of global warming. Over the next quarter-century, although gas use is projected to fall in much of the industrialized world — places such as the United States, Europe and Japan — it is expected to soar in emerging markets and developing economies, notably across Africa, the Middle East and Southeast Asia, the regions where the climate battle will be won or lost. If current policies and market trends continue, the IEA projects, by 2050, even as renewables will have nearly tripled their share of global energy, to 29%, and as coal use will have nearly halved, to 15%, gas still will supply 20%, nearly the same proportion it provides today. Whether that gas can be decarbonized will to a large extent determine whether humanity stays within crucial climate boundaries or blows past them.

Yet LNG can’t be credibly called climate-compatible unless the burgeoning green-gas market is made harder to manipulate. That would require a robust system to quantify LNG’s carbon footprint, to measure and validate efforts to reduce it, and to police claims the industry makes to consumers about the supposed greenness of its gas. Such a system would need verification that is rigorous and independent. It would need to follow a methodology that is transparent and is consistent across the industry. And it would need to do all that with measurement technology and analytical techniques that are nascent, extraordinarily complex, and evolving faster than regulators can update rules. Otherwise, one company’s claims that its gas has a particular carbon footprint — or, indeed, that its gas is greener than another company’s — are at best opaque and at worst misleading to the market and dangerous for the climate. The EPA’s new methane rules may prove an important step in this direction. But the fight in Washington over the regimen’s fine print, which will determine its sweeping impact, has only just begun.

The stakes for the planet are clear from the juggernaut that is Sabine Pass. There, concerns about the legitimacy of the LNG industry’s climate claims appear to be imperiling sales not much at all. Cheniere declined my requests to visit the plant; though the company has organized tours for others, a spokesperson told me by email that Cheniere wasn’t going to “dedicate the resources” to show me around. Even from afar, however, it’s clear the place is going hard. Day and night, the facility’s compressors crank away. Out on the water, a steady stream of tankers ferries Cheniere LNG from the chillers of Sabine Pass to hot spots of fossil-fuel consumption around the globe. According to the latest federal statistics, the amount of greenhouse gas coughed out by the plant more than tripled between 2016, the year Cheniere began exporting LNG, and 2021. Fusco, the Cheniere CEO, reported on the company’s most recent quarterly earnings call, in November, that its LNG production had hit a new record high.

———

Read part two of this special reporting project: Can an economic giant clean up natural gas — and then swap in hydrogen?