Gorgon carbon capture and storage project: The failure of the world’s largest CCS facility
After decades of abject failure, these days even the most enthusiastic proponent of carbon capture and storage can at best mount a muted endorsement of the technology. At least among those willing to be bound by facts.
The standard lines today have shifted from where they were in the late 2000s, where seemingly plausible — though never actually plausible — visions of the future saw every coal-fired power station in the world being retrofitted with carbon capture to deliver guilt-free fossil power.
Today’s CCS advocates have retreated a long way. Nowadays even the most enthusiastic arguments that they can run amount to little more than: “Sure. It has failed so far. But CCS will help us to reduce emissions at a scale and on timelines consistent with the climate crisis for one simple reason: it must.”
What is almost never asked in response — and certainly never answered — is what is going to happen when the technology fails to provide us with the deus ex machina so many have been wishing for. None have wished harder than the fossil fuel industry hoping against hope that it can continue business as usual as if the climate crisis didn’t exist.
As Ketan Joshi has repeatedly. pointed. out. — including in a new post he released while I was drafting this very piece — the rhetorical use of carbon capture and storage is at least as interesting as its technical use. CCS is always and forever a technology that the fossil fuel industry will deliver to us next year, in two years, in five years, in ten, but it is never one that will be delivered today. Over the past several decades, nearly every single carbon capture project that has ever been planned has never arrived.
This is what makes the few that have arrived fascinating to me. Each CCS project that exists has been birthed into the world against the odds. Each one a success of-a-kind, if only you can ignore the all-consuming failure.
Most of the largest “success” stories are in enhanced oil recovery, a technique that uses injected carbon dioxide to extract more oil and gas — lifting emissions overall. Ultimately, EOR is a solution to improving the economics oil and gas exploitation, and not climate change.
Many far-smaller success stories exist as well. Invariably these are covered with breathless headlines and little context. One article from the other week described a new CCS test facility as ‘A Secret Weapon Against Carbon’. Are secret weapons usually effective? The article goes to great lengths to make that one test facility sound like the cure to the climate crisis. However, it captured an amount of carbon dioxide equivalent to 3% of one Australian’s emissions for one year.
Even those few projects of high repute are increasingly being found to be far less impressive than has been made out. Equinor’s Sleipner and Snøhvit are widely cited as ‘the good ones’ of CCS. Indeed, Australia’s Resources Minister Madeleine King spruiked these Norwegian projects in a speech earlier this year before she discussed Australia’s major project.
How are they going? Well, the Sleipner operation had an equipment malfunction that meant the Equinor was massively overestimating their capture rate and it took three. years. for anyone to notice. These projects are currently running at very far below their capacity. There are reasons for this, but as this post will show, there are always hyperlocal reasons that CCS underperforms. At a certain point, we need permission to start generalising from underperformance at every single site to what that says about the use of the technology by the oil and gas industry overall.
Chevron’s Gorgon carbon capture and storage project on Barrow Island — a 22 km long island of the coast of Western Australia couldn’t possibly be doing worse than that, could it?
What does carbon capture and storage do in the petroleum industry?
When gas and oil naturally form underground, it is natural for carbon dioxide to be held in the same reservoirs along with it. After the fossil fuels have been extracted, an essential early step in processing them is to strip carbon dioxide and other potentially acidic gases out of the mix to limit the harm to the infrastructure. This occurs at the creatively named ‘acid gas removal unit’.
Some fields have much higher or lower amounts of carbon dioxide in them. Santos’s Barossa gas field, located about 300 km NNW of Darwin, has a carbon dioxide content of 18-20%. If you extract 10 litres of gas from that field up to two litres will be carbon dioxide. Meanwhile, Woodside’s Scarborough project, just under 400km off the Pilbara in northwestern Western Australia, is claimed to have a carbon dioxide content of just 0.1%. If Woodside claims are correct, then extracting the same 10 litres of gas from Scarborough will result in less than half a cup of CO₂.
It is important to clarify something here.
As you no doubt know, when coal, oil and gas are burned, the fossil fuels are converted into carbon dioxide and other pollutants. These combustion emissions are always by far the largest share of the emissions attributable to a fossil fuel extraction project. However, the carbon dioxide and other gases that are released from an oil or gas field when these fuels are extracted is an additional burden on the climate: a so-called fugitive emission.
Traditionally after the carbon dioxide is stripped from the gas, it is released directly into the atmosphere with no obligation on the oil and gas field operator to either capture it or use it. The operators of both the Scarborough and Barossa gas fields mentioned above intend to release their reservoir carbon dioxide directly to the atmosphere, at least in the early stages. Santos has made some vague overtures that it might one day, maybe, possibly, with the right amount of somebody-else-is-paying-for-it, capture the carbon dioxide from Barossa. This will involve Santos transferring the CO₂ 500 km by pipe to a depleted gas field in Timorese waters. Santos are quite clear that while they might do this one day, when pressed for an answer about when they will start, they are quite clear that they sure as shit aren’t planning to do it any time this decade.
Enter the Gorgon
Chevron’s Gorgon liquefied gas terminal sits on Barrow Island, itself located off the northwest coast of Western Australia 145 km west of Karratha. Barrow is a flat, spinifex-covered place with a desert climate. It has played several roles since colonisation. This includes being a hub for the trade in Australian Aboriginal slaves in the middle of the 19th Century. In the past it has also played a home to the whaling and guano mining industries.
Since the 1960s, Barrow Island has hosted the oil industry. At times Barrow Island has been Australia largest active oil field. A little context is important here. Despite being the world’s third largest fossil fuel exporter, Australia doesn’t produce very much oil. Indeed, the country imports most of the oil it uses. As such, while a lot of oil has been produced from Barrow Island, the oil reserve beneath its rocks isn’t at all large by world standards.
There is no permanent human population on Barrow Island, and its isolation makes it a unique refuge for species that are threatened or have gone extinct on the mainland. It has a number of endemic species that are found nowhere else.
In the Walkley-nominated new book Slick: Australia’s Toxic Relationship with Big Oil, author Royce Kurmelovs discusses the manipulation of public opinion that was required to convince the Australians to allow oil exploitation in a place of such high conservation value. The petroleum industry’s efforts largely revolved around the recruitment of celebrity naturalist Harry Butler. Employed as a consultant to the growing industry, Butler unsurprisingly found that the arrival of the petroleum industry on Barrow would leave ‘no lasting or serious damage’ on the island.
With these words being widely published across the country to allay the fears of the public — as Kurmelovs says — this was a fine example of the co-option of environmentalists by the petroleum industry to buy social licence. Elsewhere it has been said that the Gorgon gas terminal ‘wouldn’t exist today if it wasn’t for the efforts of Harry turning an A-class nature reserve into something that could be compatible with the oil and gas industry.’
What has been left by the onshore oil industry on Barrow Island is a scarred landscape, pock-marked across its length and breadth by the petroleum industry, particularly in the south and east.
And of course, if we can do that to an area of such high conservation value area, why not add an even larger industrial facility there?
In 2009, Chevron Australia received approval from the Federal government’s offshore petroleum regulator NOPSEMA to develop the Gorgon gas field, around 75 km off the coast of Barrow Island. In 2010, a second approval came through to develop the Jansz-Io field which is further afield at about 130 km away from Barrow. Chevron’s plans to exploit these fields required the construction of liquefied gas (aka LNG) terminal on the eastern side of the island. Approval for this development was granted by the Western Australian government in 2009. This facility would process the gas from these new offshore fields and compress it for export via ship.
On the scale of low to high carbon dioxide content, the gas from the Gorgon field is at the Barossa end of the scale. While it isn’t quite as high as the Santos field’s 18-20%, the Gorgon field is nonetheless stunningly high at around 14%. That means that if you remove 10 litres of gases out of the underground reservoir, about one seventh — five and a half cups full— will be carbon dioxide.
With Gorgon and Jansz-Io fields operating at full capacity, each year over 5 million tonnes of carbon dioxide are produced the from the fields. Even before considering that the carbon dioxide isn’t the only major source of emissions from the operation of a liquefied gas terminal, this is enough carbon dioxide to place Gorgon comfortably inside the top five highest emitting industrial facilities in the country. Just the reservoir carbon dioxide from this one terminal is more than the entire Qantas fleet of aircraft emits from its Australian operations in most years.
As mentioned above, before the gas can be compressed and sold, the carbon dioxide must be stripped out. This is both to purify the product and because high CO₂ levels will lead to the infrastructure slowly being eaten away, especially if the CO₂ gets the opportunity to mix with water and form carbonic acid. If you’ve ever heard people discussing the issue of ocean acidification occurring with climate change, it is increased carbonic acid in the ocean that people are raising concerns about.
Because of the field’s extraordinarily high carbon dioxide content, approval to build the Gorgon liquefied gas export terminal was only given to Chevron by the Western Australian government on clear conditions. On the terms of its original approval, Chevron would need to (a) build a CCS facility that is capable of capturing and safely storing all the reservoir carbon dioxide removed by the acid gas removal units, and then (b) to operate that facility in a way that ensures no less than 80% of the carbon dioxide is captured.
With the goal of capturing at least 4 million tonnes of carbon dioxide per year, this would make Gorgon CCS the largest ‘pure’ CCS facility in the world, at least in terms of capacity. While there are at least two larger CCS operations — ExxonMobil’s Shute Creek in Wyoming and Petrobras’s Santos Basin Pre-salt oil field CCUS — both of these are enhanced oil recovery operations, where — as discussed above — the captured carbon dioxide is used to extract more oil and gas, leading to higher overall emissions rather than lower.
Like many Arctic Indigenous communities, the Sami Peoples have dozens of words relating to snow and ice, simply because their proximity and familiarity over innumerable generations to the cold leads to linguistic diversification. When you are surrounded by ice and snow, are living within it, and have developed a deeply intimate knowledge of it, it is a given that you would develop a rich and enduring language to describe it.
We need a rich language based on proximity to abject incompetence to describe the contours of Chevron’s failure to meet the targets that were set for Gorgon CCS.
Chevron’s basic plan was as follows.
Raw gas from the Gorgon and Jansz-Io fields would be sent via pipelines to the liquefied gas terminal on Barrow Island. There, it would be fed through the onshore facility’s acid gas removal units to strip out the carbon dioxide and other impurities.
Instead of venting the carbon dioxide to the atmosphere, the gas should be sent to one of three injection sites between 1 km and 6 km north of the terminal. There the CO₂ should be pumped into the Dupuy Formation, a saltwater aquifer that sits about 2.5 km below the surface of the island. Across the three injection sites, nine wells were drilled down to the aquifer for carbon dioxide injection.
Adding carbon dioxide at a rate of several million tonnes per year to an aquifer — which by definition is full of water — will quickly begin to increase the pressure to unmanageable levels. Unchecked, this risks doing irreparable harm to the aquifer and the carbon capture infrastructure itself. A 2010 presentation from Scott Ryan, Senior Reservoir Engineer for the CCS project, concluded with a simple takeaway: “Pressure management wells are fundamental to the development plan”. Chevron obviously wouldn’t want to run their facility without some form of pressure management in full working order. Foreshadowing is a literary devi — Wait. I already made that joke.
It is worth noting that this part of the plan was high-uncertainty experiment with high stakes. At the time that Chevron was creating its plans for Gorgon, relatively little was known about how saline aquifers would work as a long-term storage medium for carbon dioxide. While they had been identified as a potentially promising alternative to injecting the CO₂ into depleted oil fields, there were few hints about what unknown complexities would arise. There are important questions to be asked about why a project of this scale was approved with mitigation conditions that had such a high chance of failure.
To manage the pressure that would inevitably be created in the aquifer by the injection of millions of tonnes of carbon dioxide every year, Chevron planned a further six wells at two sites further to the west of the carbon dioxide sites. Four of these wells would pull the salty water out of the Dupuy Formation to relieve the pressure on the aquifer. Carbon dioxide in. Water out.
But now Chevron is stuck with millions of litres of salty, probably contaminated, water on its hands. To get rid of that they planned to use the other two wells to insert the water that has been removed from the Dupuy Formation into higher layers in the rock: the Barrow Group which is 1.5 km down. The Barrow Group is where a lot of the oil that has been extracted from the island had been held, and so it represents a low-pressure area that can hold the water that has been produced far from where it can do harm.
It all sounds very complicated, right? Here’s a simplified diagram of the CCS operation.
At this stage it becomes important to acknowledge the work of Peter Milne. Pete’s ongoing reporting on the Gorgon project — through his role at the West Australian, then through his personal website Boiling Cold, and more recently through the formerfax outlet WA Today — is the reason that we know most of what we know about Gorgon’s missteps by dutifully chasing documents through FOI. His work has also served as a valuable archive to have at the foundation of this extended history. A huge amount of the below is based on his public archive of documents related to Gorgon that you can find here.
I’m really glad that Pete is re-launching Boiling Cold in the next few days. It’s definitely something to add to your regular reads list: https://www.boilingcold.com.au. You can also follow him on Bluesky: @petemilne.bsky.social
Reality bites
As happens with many of these projects, Gorgon CCS’s arrival was being lauded long before the project was anywhere near completion.
At the time of Tony Abbott’s federal election win in 2013 — based as heavily as it was on his opposition to a carbon price — Fairfax papers ran an article with following headline:
It was a bit much. Even according to the plans Chevron made when Gorgon received its first federal and state approvals, the first gas shipment wasn’t due to be made for another year.
In reality, for reasons that go beyond the scope of this post — but that have been very well covered by Pete Milne himself — the first shipment was more than 18 months late, with train 1 firing up on March 7 2016, two and a half years after Fairfax’s premature celebration. A year later, all three LNG trains were ably (though not safely) placing super-cooled liquefied gas onto ships to send overseas.
As noted previously, there were two main prongs to the approval conditions on Gorgon related to carbon capture. In full, the first two conditions related to CCS read as follows:
26 Reservoir Carbon Dioxide Injection System
26.1 The Proponent shall design and construct Carbon Dioxide Injection System infrastructure in conjunction with the Gas Treatment Plant on Barrow Island that is capable of disposing by underground injection, 100% of the volume of reservoir carbon dioxide to be removed during routine gas processing operations on Barrow Island and that would be otherwise vented to the atmosphere.
26.2 The Proponent shall implement all practicable means to inject underground all reservoir carbon dioxide removed during gas processing operations on Barrow Island and ensure that calculated on a 5 year rolling average, at least 80 percent of reservoir carbon dioxide removed during gas processing operations on Barrow Island and that would be otherwise vented to the atmosphere is injected.
For those who don’t speak fluent bureaucrat:
- Build a CCS facility that can capture all the carbon dioxide that comes out of the reservoir, and
- Operate it to aim for all the reservoir carbon dioxide to be captured, but make sure it hits at least 80%.
At the time gas production from the terminal reached full capacity, Chevron was still constructing its carbon dioxide injection and pressure stabilisation wells, and many of the well-completions — basically, all the bits and bobs that convert a hole in the ground into a working well — hadn’t been… well… completed.
By not having the carbon dioxide injection system ready at the time LNG production began, the company had breached the conditions on its environmental approval. By not having the injection system ready at the time gas shipments began it had failed to install a carbon dioxide injection system capable of capturing 100% of reservoir carbon dioxide.
Notwithstanding the problems that would follow for the CCS facility — some of which were legitimately unforeseeable — not having the wells drilled and completions done for the carbon capture system doesn’t appear to have been anything more than an issue of resource allocation at Chevron. Liquefied gas production started 18 months later than originally planned, and full production occurred the best part of a decade after the first approvals were received. This is more than enough time to drill 15 wells. It’s not like NIMBYs were an issue. Out on Barrow, the most vocal community opposition is a judgemental wallaby that is as tall as a desk ruler.
Chevron was not nearly so lackadaisical when it came to sharpening their pen to claim taxpayer dollars though. The company was a proud recipient of $60m from the federal government under the Low Emissions Technology Development Fund. This bespoke scheme only ever funded this one project and provided with Chevron with three payments of $20m for the development of the carbon dioxide injection system. While most of these payments were made before completion, the threshold for the final payment was “Ready for start-up of the first carbon dioxide compressor”.
By the time of the first gas shipments, Chevron had already noticed issues with the project design. A working assumption through the design phase was that water condensation was not going to be a significant issue at the injection wells. Pre-startup tests had shown that this assumption held true if the injection wells were operating at a steady state. However if the injection operations were forced to scale down or shut-down temporarily, condensation — and so rust — became a real issue for the plant. Fixing this issue — along with certain others identified at this early stage — would require engineering modifications to the compressors. These modifications would introduce a delay to the start-up of the carbon dioxide injection of at least one year. CCS would have to wait until after July 2018.
As of July 2017, some wells had not been drilled, many completions hadn’t been installed and — crucially — according to Chevron’s reports to the department overseeing the Low Emissions Technology Development Fund, the compressors weren’t going to ready to start-up for another year. Even though Chevron absolutely had not met the final hurdle of their grant, Chevron submitted its final request for payment on 14 April 2017. The company received $20m for a job well done. The compressors wouldn’t actually be ready to start for another two years.
While Chevron initially claimed that modifying their three compressor trains — each with two compressors — would introduce a one-year delay beyond their original CCS start-up date of 30 September 2017 and be implemented by 30 September 2018. One year later they revised this date to 31 March 2019. A year later still, they revised it again to 30 September 2019. In the event, the first carbon dioxide injection occurred on 6 August 2019. In another six months, all three acid gas removal units were linked up to the carbon dioxide injection system.
Between the first liquefied gas shipment and the time when the first molecule of carbon dioxide hit the Dupuy Formation, 9 million tonnes of the gas — equivalent to the emissions of about a million cars over the same period — were vented directly to the atmosphere from just one industrial facility on an island most people have never even heard of.
Under pressure
Before the compressors could be fixed, a bigger and more stubborn problem than rust had begun to raise its head.
Drilling Centre D is located about 8 km northeast of the gas terminal and is one of the two sites responsible for pressure management in the Dupuy Formation. There are three wells here. D-WP1 and D-WP2 are water production wells that pump salt water up from the saline reservoir far below. D-WI1 sits alongside them. It receives the water from D-WP1 and D-WP2 and inserts it into the depleted Barrow Group.
Here’s that diagram from before again.
Those identification codes for the wells are a lot less complicated than they first appear. The “D” in “D-WP2” identifies that this well is at Drilling Centre D. There are five drilling centres associated with the carbon capture operation on Barrow, running anti-clockwise from the terminal from “A” to “E”. The “WP” stands for “water production”. The other codes are “I” for the carbon dioxide injection wells, and “WI” for the two water injection wells. Finally, there is a number at the end: 1 through 9 for each of the nine carbon dioxide injections spread across Drilling Centres A, B and C, 1 to 4 for the four water production wells at Drilling Centre D and E, and 1 and 2 for the water injection wells sitting alongside them.
Sometime before the first compressor turned on, and while the wells at Drilling Centre E were still being completed, Chevron found that D‑WI1 had lower than expected injectivity. This is a jargonised way to say that the well couldn’t handle the amount of water that they were trying to shove down it. That shouldn’t have happened. Head scratching ensued. It’s not mentioned in the reports, but it seems likely that at least one engineer gave wellhead an experimental kick.
Over the next year, a working theory began to form that the well had been clogged with grease and debris deep underground from construction of the water production wells. During the testing phase — the theory went — the crap that had accumulated along the length of the two 2.5 km water production wells had been pulled up and stuffed into the injection well blocking it at the end and preventing water flow.
Remediation works were planned, with wells at the two sites responsible for managing the pressure in the Dupuy Formation expected to be completed — late — in the first half of 2020.
It was at around this time that the first compressor started up and carbon dioxide from the Gorgon field progressively began to be inserted underground via the nine carbon dioxide injection wells. Each of these wells was capable of injecting between 30 and 40 million cubic feet of carbon dioxide per day. To bring a little bit of perspective — and metric! — to number that big: across the nine wells an amount of carbon dioxide gas capable of filling a cube 200 metres by 200 metres by 200 metres was being inserted underground every day.
Soon, the same injectivity issues that had affected D-WI1 began to affect E-WI2, the water injection well at Drilling Centre E.
Despite having identified as early as 2010 that running Gorgon CCS without working pressure management was a very bad idea, Chevron decided to take a wait-and-see approach to pressure management once operations had started. Through the 2020 calendar year, the company ran the carbon dioxide injection wells at near maximum capacity.
Over time the consensus shifted on why the water injection wells weren’t working: it wasn’t grease and debris from construction, by sand sucked out of the Dupuy Formation. In a report dated 30 September 2020, almost a year and a half after the injectivity issue had been identified and with no clear remediation plan in place, Chevron stated simply:
It is yet to be confirmed whether sand production is likely to be a long-term issue. If sand production is found to be persistent issue [sic] it is possible changes to the surface facilities may be required.
This had obvious consequences. Obvious — that is — to everyone except Chevron.
At the end of 2020, pressure in the Dupuy Formation began to build to unacceptable levels and the Western Australian government was forced to intervene to protect the integrity of Chevron’s own infrastructure. The government placed a hard limit on the CCS facility until the pressure management system was operational. This restriction limited carbon dioxide injection to 70 million cubic feet per day, a little less than one-third of the facility’s nominal maximum capacity.
Temporary sand management equipment was installed at the injection sites in the first half of 2021, and Chevron expected to return to full capacity later that year. It did not. Injectivity problems continued in 2021 despite efforts to re-perforate the injection wells to find a way past the blockage. In late 2021, despite the WA government doubling the hard cap on the injection rate, carbon dioxide injection was only able to operate at 84 million cubic feet per year.
The sand problem continues to this day, and while some long-term solutions have been proposed in more recent work plans reported to the Western Australian government, these are not attached to concrete work plans.
Like oil and water
Sand isn’t even the only problem with the pressure management system.
During testing, one of the four water production wells, E-WP4 (that is, the second water production well at Drilling Centre E), was found to be bringing a nasty surprise up with its briny payload. Instead of straight water, the well was also producing a meaningful quantity of hydrocarbons — that is gas and crude oil — along with the water.
Oil popping up where you were planning to get water feels like it should be a happy accident for a petroleum company. However, the presence of the gases interferes with both water production and injection in the pressure management system. This further limited the ability of the already strained-slash-borderline-nonexistent pressure management system to do its job.
Across various public documents, and several others obtained under Freedom of Information laws by Pete Milne, I have never seen an attempt by Chevron to account for the scale or impact of this issue. I have also never seen details of any measures put in place to resolve the issue. It seems that, to some degree, the scale of the issue ebbs and wanes, and we know that it continued until at least the 2022 calendar year.
Given Chevron’s enormous contribution to the climate crisis both through its direct emissions and through the eventual combustion of the fossil fuels that it produces, it almost feels petty to mention this next part. I will anyway because while the scale is certainly trivial in the context of Chevron’s overall emissions, by any other standard other than comparing these emissions to those of Australia’s largest industrial source of climate pollution they are meaningful.
Chevron never specifies what hydrocarbons that are being produced by the well, but it is almost certain that a significant proportion is methane. Methane is the simplest hydrocarbon, the most common gaseous hydrocarbon and the main component of fossil gas. As well as being a fossil fuel, it is also a greenhouse gas in its own right. It seems likely, given the state of our emissions reporting scheme that these methane emissions are not being accounted for under our existing emissions reporting scheme at all. This is important because unburned, methane packs a hell of a punch on the atmosphere.
The issue here is that the logical answer to the problem of gaseous hydrocarbons being produced in the water production wells is to vent these gases. And while the amount of methane being emitted very likely won’t be large compared to the facility’s total emissions, if they are being vented this would represent an unaccounted-for source of emissions at a facility that is already having a significant emissions problem.
A problem with the pressure management system
Are you tired of talking about the pressure management system yet? How do you feel about puppies? Here’s a puppy.
This puppy is concerned because he knows that I’m going to tell you about yet another problem with the pressure management system. This one is completely separate to the sand issue, and also separate to the accidental-oil-well-that-should-have-been-a-water-well issue. He loves you and wants to support you through this, but he knows you can do it.
So, I regret to tell you that Chevron has also realised after all this that even if they didn’t have the sand or accidental oil well issue, that they just flat out didn’t drill enough water producing wells.
Wait.
I misspoke.
Chevron “just flat out didn’t drill enough water producing wells”, said Matt Hatfield, Managing Director of Chevron Australia at the World Energy Cities Conference in November 2023.
In the 2022 calendar year, the “fully operational” water production wells Chevron installed on Barrow Island as part of the pressure management system at Barrow Island produced 1.34 gigalitres of water, and while we don’t have concrete figures, a similar amount was anticipated for 2023, and most likely 2024.
Chevron has now decided that maintaining the pressure in the Dupuy Formation is going to require the company more than tripling that capacity, taking it from 1.34 gigalitres per year to 4.65 gigalitres per year. The solution seems simple. If you need three times more water, triple the production at you wells and/or dig more wells.
But tripling the water production and injection rates triples the possibility of further complications exactly like those that have occurred so far. The sand issue at D-WI1 and E-WI2 hasn’t been resolved, and there is nothing on the horizon — at least nothing publicly announced — that is galloping down the hill as a likely candidate to save it. There is no guarantee that increasing production will fix the problems at all.
It certainly won’t do so in the near term. Commentary from the company has indicated that it is certainly not in a rush to resolve the issue. The most we have heard is that they are “working toward the placement of long lead equipment orders” and that even in the best case, the solutions have been determined to be something that will take “years, not months”.
Meanwhile, millions of tonnes of greenhouse gas that Chevron promised to capture — indeed, that they were legally required to capture — are making the way to the atmosphere every year and pushing the global climate off its axis.
And while Chevron has been “punished” in a manner of speaking for its failure so far, that punishment almost certainly costs Chevron less than it would cost to fix the pressure management system.
Gorgon CCS by the numbers
Between the start of liquefied gas production in 2017 and the end of the 2023–24 financial year, the acid gas removal units on Barrow Island have stripped 30,760,271 tonnes of carbon dioxide from the raw gas produced in the Gorgon and Jansz-Io fields.
At least 9,000,000 tonnes of this was stripped out before the carbon capture system was ready. Chevron was just late, and because of this the global atmosphere became the company’s dumping ground. There was no effective regulator to hold it to unequivocal conditions in its environmental approval.
Then Chevron fired up the CCS and for one year (2020) it operated at full tilt without a working pressure management system. The pressure management system is something that Chevron’s own engineers had said was fundamental to the plan ten years earlier, but really? Who needs it? Without a doubt, the CCS project had one very good year where it injected a lot of CO₂. But in acknowledging that one good year, you also must acknowledge that at the end the year the Western Australian government — which is not known for over-regulating the gas industry — was forced to intervene to protect Chevron from itself.
Since then, the amount of carbon capture at Gorgon has seen a slow decline.
While more carbon dioxide was captured in 2022–23 than in 2021–22, 2023–24 set a record for the lowest amount of carbon dioxide captured since the first molecule was inserted into the Dupuy Formation. Almost as much reservoir carbon dioxide was dumped into the atmosphere in the past year (3,723,328 t) as was dumped into it in the year before the CCS system first fired up (3,741,777 t).
Over the past three years, less than one-third of all reservoir carbon dioxide has been captured. As a reminder, Chevron was required under its approvals to build a facility capable of capturing all the CO₂ removed from the field, and also required to operate that facility in a way that ensures it captures at least 80% of that CO₂. Between the commencement of carbon capture operations on Barrow, Chevron has captured a total of 9,836,017 t CO₂. That sounds impressive until you realise that—in addition to the roughly 9 million tonnes of vented CO₂ that occurred because the carbon capture operation was late — Chevron has also fallen 8,113,888 t CO₂ short of reaching 80% target in the years where CCS was operating. Its failure to build a CCS facility capable of capturing all the reservoir CO₂ has led to another 4,487,476 t.
Putting Gorgon’s failures into context
As long as this story is, and as breathtaking as Chevron’s repeated failures have been, this is also story about a niche technology that could never have made a meaningful contribution to an enormous climate burden that began under the waters off Barrow Island and will affect our lands, skies and waters for generations to come.
Even if Chevron was capable of meeting its legal obligations in a way that it has shown itself to be unable or unwilling to commit to do, Gorgon CCS could never have made a dent in the climate crisis.
To begin with, even though the reservoir carbon dioxide burden at Gorgon is enormous and is many years is the largest source of emissions from the project, it is only narrowly so. In 2023–24, 5,317668 t CO₂ was stripped from the acid gas removal units on Barrow, of which a paltry 1,594,340 t CO₂ was successfully injected. Meanwhile, Chevron’s other sources of emissions on the island — like Chevron burning through its own product to run the liquefied gas terminal — produced approximately 4,600,000 t CO₂e of climate pollution. This additional 4.6 million tonnes of climate pollution is something that Gorgon CCS could never mitigate because the capacity and systems are not in place. Honestly, it very likely could not ever be put in place.
This chart below puts Gorgon CCS into context with emissions from the facility as a whole. In 2023–24, Gorgon was the highest emitting industrial facility in the country. A fully operational CCS facility would have dropped Gorgon from first to fourth, and still higher than the annual emissions of Qantas’s entire Australian operations at about 5.5 million tonnes per year.
In the chart below, if the CCS facility was working perfectly — capturing every single tonne of reservoir CO₂ — that would mean none of the emissions shown in red would make it to the atmosphere. However, that would still leave all of the other sources — shown in blue — unaddressed.
But even this paints a manifestly insufficient picture. The climate burden created by Chevron’s operations in and around Barrow Island do not end after a ship laden with liquefied gas from Gorgon leaves Australia’s waters. Gas is a fossil fuel, and it will be burned. The economic reality is that wherever it is burned in the world, it will never be hooked up to any kind of meaningful mitigation tech, whether CCS or something else. (For what it’s worth, those who try to convince you otherwise — that CCS at end-use is feasible in anything other than a few edge cases — are simply trying to buy your permission to ignore their climate destruction.)
While Gorgon primarily produces liquefied gas, it also produces piped gas and condensate, and I don’t have precise figures for the quantity of the other fuels that are produced in most years. That said, here is an indicative assessment of the total climate burden from Gorgon CCS.
Whether 80% of the reservoir carbon dioxide is captured or all of it is — heck! — even if all the emissions from the operations on and offshore at Barrow Island are captured — it fundamentally doesn’t make a real difference. Ending the age of climate destruction requires an end to fossil fuel projects. How bad it gets depends entirely on how fast we can bring fossil fuel production down to zero.
Gorgon’s failures are hyperlocal in many ways. Each problem that has been faced on the site is so specific to its location that it is unlikely to repeat itself elsewhere and there is very little that can be learned from this project to apply to the next attempt. But this is itself a generalisable lesson. Carbon capture and storage will always face these hyperlocal problems. This is especially so in an era when operators aren’t required to meet their commitments.
Alongside the fact that many carbon capture projects are simply narrative devices — with just enough work put in to make it look like there may not need to be a future beyond coal, oil and gas — these issues are why projects consistently fail and why the cost of CCS is not meaningfully decreasing over time.
Ultimately, I hope Chevron do get their arses into gear and get Gorgon’s CCS working, not least of which from a stubborn sense of legalism. They are required to do this and went ahead with the project on that understanding. It is only right that they should be held to that commitment. But whether carbon capture and storage works at a site level or it doesn’t is in many ways simply irrelevant. The fact that it has been a catastrophic failure is just a shit cherry on top of a cake made of shit.
One day soon, I hope that we can move away from arguing that carbon capture will work simply because it must, and start the hard work on what we will do when — not if, but when — it doesn’t.