Marking 20 years
of bold journalism,
reader supported.
News
BC Politics
Environment

How Safe Is LNG? Not as Safe as the BC Government Has Claimed

New studies and past incidents raise questions about location standards and safety.

Andrew Nikiforuk 28 Apr 2017TheTyee.ca

Andrew Nikiforuk is an award-winning journalist who has been writing about the energy industry for two decades and is a contributing editor to The Tyee. His award-winning book, Slick Water, documents the Jessica Ernst case and the history of fracking. Find his previous stories here.

To support its struggling advocacy for as many as 20 proposed liquefied natural gas export terminals in British Columbia’s coastal communities, the provincial government routinely boasts that the industry has an enviable record of public safety.

“LNG is safe, on land and in transit as a result of detailed industry standards, strict regulations, and a commitment to risk management,” reads one government website.

But that’s not really the complete or truthful story.

As the U.S. Congressional Research Service tidily noted in 2004, there have always been public concerns about the siting of LNG terminals because “LNG is a hazardous liquid transported and stored in enormous quantities, often near populated areas.”

In fact, a recent and nasty explosion at an LNG facility in Washington State, combined with alarming British research on what scientists call “vapour cloud explosions” or the release of flammable gases into the atmosphere, has raised new doubts about the adequacy of safety regulations for siting LNG export terminals in North America.

Jerry Havens, a long-time expert on the catastrophic releases of hazardous materials and a chemical engineer at the University of Arkansas, doesn’t mince words about the problem.

He believes that “the present methodology for regulating LNG terminal import and export hazards to the public are overdue for careful review and assessment” in the U.S., and current standards “provide inadequate fire and explosion exclusion zones to protect the public.”

The U.S. agency in charge of LNG on-land terminal regulations, the Pipeline and Hazardous Materials Safety Administration, has begun a comprehensive review of its rules for citing LNG terminals.

The process of making LNG is complex, involving a highly flammable gas that is chilled and condensed. Once the gas has been cooled to temperatures below minus 160 degrees Celsius it forms into a liquid. Liquefied methane occupies one six-hundredth the volume of its gaseous state, and can be stored and transported in great volumes.

One LNG tanker, for example, can carry enough LNG to heat 10 million homes for a day. That’s a lot of energy tucked into a confined space.

Experts agree the chief hazard is a liquid spill, which would set off a series of events usually ending with fire or an explosion. At first the cold liquid, heavier than air, would form a pool that would evaporate into a cloud. Wind or a structural barrier such as building could direct that cloud to an ignition source. And that’s when a big fire or explosion could happen.

In 2008, Havens made this candid observation on LNG shipping safety while testifying in Ireland: “If an LNG carrier were to be attacked in the proximity of the shoreline, either while docked at the terminal or in passage in or out of the estuary, and cascading failures of the ship’s containments were to occur, it could result in a pool fire on water with magnitude beyond anything that has been experienced to my knowledge, and in my opinion could have the potential to put people in harm’s way to a distance of approximately three miles from the ship.”

No disaster like this has happened yet. Since international LNG shipping began in 1959, tankers have carried tens thousands of cargoes without a serious accident at sea or port. Tankers have grounded and even collided, though none has yet spilled its volatile cargo.

But experts have grown increasingly concerned about a phenomenon known as rollover, or the rapid release of LNG vapour that can occur as a result of the spontaneous mixing of layers of different densities of LNG in a storage or cargo tank.

Tankers so far safe, terminals a different story

Unlike tankers, LNG terminals have a mixed safety record.

In 1944, an explosion in Cleveland incinerated 128 citizens in a fiery methane cloud. Since then, industry likes to boast that all subsequent incidents have been minor.

But that’s not the full story either.

Since the Cleveland fireball, LNG terminals have experienced more than a dozen serious accidents, mishaps and explosions. One of the most notable was an explosion at an LNG terminal in Skikda, Algeria.

There, a large volume of liquid gas escaped from a pipe and formed a cloud of highly flammable and explosive vapour that hovered over the facility.

The blast broke windows more than mile away. The conflagration killed 27 workers and injured another 80. Half of the facility was destroyed causing nearly $1 billion worth of damage.

Skikda explains why the likelihood and severity of catastrophic LNG events has been the subject of controversy wherever the industry has wanted to locate a terminal in North America.

U.S. researchers and risk analysts started to rethink LNG safety’s record again after a catastrophic explosion that injured five workers and shut down the town of Plymouth, Wash., in 2014.

The site, operated by Oklahoma-based Williams Companies, didn’t export LNG, but converted methane from the Northwest Pipeline into a liquid for storage during summer months. It then re-vapourized the gas during peak demand in winter.

In March 2014, an operator error at the facility resulted in what investigators later called “a catastrophic failure and a resulting explosion on the portion of the facility’s LNG-1 purification and regeneration system.”

According to state and federal investigators, the company failed to purge a mixture of methane and oxygen from the facility, which resulted in a rolling detonation, fire and explosion. Blast debris as heavy as 250 pounds damaged the facility, punctured the outer shell of an LNG tank and destroyed nearby railway tracks.

All residents living within a two-mile radius of the plant were evacuated. For a day or two, Plymouth became a ghost town.

Although the accident released a dangerous LNG vapour cloud into residential areas, the event didn’t meet the definition of “a threat to public safety” and federal rules did not classify it as an LNG spill because the explosion didn’t kill anyone.

During the incident the plant lost about 168,000 cubic feet of gas. To shut off leaks caused by the explosion, employees donned cryogenic suits to close a valve between the damaged pipe and an LNG tank.

The incident took authorities two years to analyze, and the company $70 million to repair. It also cast another dark shadow over the industry’s terminal safety record.

At the time, Stephen Maloney, a senior risk consultant at Moody’s Analytics with a background in LNG risk, described the event to Reuters as “notable” but not particularly severe.

“But, when you are dealing with very low frequency events, even for an event of limited severity, one data point has the potential to really change statistics,” added Maloney.

Computer models and real explosions

While experts studied the explosion, other researchers raised concerns about the computer models being used to determine the siting of LNG plants near people.

After the terrorist attacks of 9/11, experts recognized that LNG carriers and terminals had a terrible vulnerability. They also admitted an attack on the industry could result in spills greater in volume than ever studied by LNG researchers in past experiments.

Because a major LNG tanker spill has never occurred, regulators and researchers must rely on computer models to evaluate hazards.

In 2015, Havens and James Venart, a mechanical engineer at the University of New Brunswick (now deceased), called those models inadequate and the siting policy for LNG terminals “faulty.”

In a paper submitted to the federal review of Oregon’s Jordan Cove Export Terminal, which the government later rejected, the experts noted that LNG ships and terminals in the last six decades had increased in size 10 times from its origins, while “regulatory guidelines had not been continually reviewed and updated.”

It is true that methane poses fewer hazards than heavier gases such as propane or butane, they wrote, but methane still ignites easily and burned hotly and those attributes “entail hazards that multiply with the amounts of fuel involved.”

Moreover, LNG export terminals didn’t just store methane. They also had to remove heavier gases and impurities from methane streams before cooling and liquefying the gas. That meant export terminals stored and handled a variety of explosive gases from butane to condensates — responsible for the some of the world’s worst industrial explosions. Those risks weren’t being properly considered, argued Havens and Venart.

The researchers raised some other issues. One concerned the mathematical models used to create safe separation zones from LNG terminals. They hadn’t been subjected to rigorous validation, said the experts.

The industry’s tendency to use proprietary models that can’t be verified by independent evaluation “prevents a basic public right-to-know,” they said.

The industry also underestimated the potential for leaks, they said. In order to maintain the illusion that the consequences of accidents will not extend offsite and harm the public, the industry and regulators had developed “design spills” involving leaks from small pipelines. “It is not good science or good engineering,” said Havens.

The 2004 review of LNG safety hazards by the Congressional Research Service highlighted the same dubious practice: “Critics of these LNG safety provisions argue that the thermal and vapour exclusion zones they specify may be too small, in part because the ‘design spills’ on which they are based are too small. They argue that catastrophes such as terrorist attacks on storage tanks could release far more LNG far more quickly than assumed in siting plans — resulting in larger, hotter pool fires or larger vapor clouds closer to nearby populations.”

Finally, the industry ignored a troubling global trend at facilities that routinely handle materials heavier than methane, such as gasoline. Since 2004, major “unconfined vapour cloud explosions at least four of large facilities resulted in heavy civilian and worker casualties and massive damage as a result of cascading events.”

Havens didn’t think LNG export terminals, which also handle a variety of explosive gases heavier than methane, could be immune from such catastrophic events.

New research confirms risks

Last year, a new study by the U.K. Health and Safety Executive, an agency devoted to reducing work-related injuries, supported Haven’s concerns about the transport and storage of energy fuels.

At the request of U.S. regulators, British safety expert Graham Atkinson looked at a cluster of vapour cloud explosions caused by leaks at facilities storing or transporting large amounts of fuel, creating exceptionally high energy densities that resulted in catastrophic explosions.

The severity of the explosions baffled safely experts because the incidents had not been identified previously in a major hazard assessment of these kinds of facilities, including LNG. All were catastrophic.

One took place at a British oil storage depot. The overfilling of a gasoline storage tank in Buncefield, England, in 2004 resulted in massive conflagration that injured 43 people, forced the evacuation of 2,000 and measured 2.2 on the Richter scale.

Another incident occurred at an Amuay oil refinery in Venezuela in 2012 when a pump failed and a cloud of propane rolled across the facility until it found an ignition source.

The explosion incinerated several residential areas, a military post, a hydrogen plant and killed 47 people. Damage exceeded $1 billion.

Atkinson’s review drew some disquieting conclusions. For starters, the worst incidents occurred in nil or very low wind conditions. Small but sustained leaks built large vapour clouds in a matter of minutes that spread through the facilities. In many cases, the flammable zone reached more than 500 metres from the source of ignition.

In other words, a small leak can result in a big bang given the right conditions.

A spill of LNG at minus 165 degrees Celsius can form a vapour cloud. Initially heavier than air because of its low temperature, the cloud, writes Havens, “will spread laterally and remain close to the ground, prolonging both in distance and time the potential hazard to the facility and to the public.”

In a presentation last year to U.S. LNG regulators Atkinson noted that his findings about the threats of vapour cloud explosions applied to LNG export terminals because these facilities all handled refrigerants such as propane and condensates which can form vapour clouds.

Last year, Havens sent a letter to the U.S. Pipeline and Hazardous Materials Safety Administration, which is now reviewing siting regulations.

It ended with one question: “Are the mathematical models which are being used as a basis for approving construction of LNG terminals, with the present focus on export rather than import, being subjected to the necessary scientific scrutiny to ensure that the hazards involved are being correctly identified?”

He hasn’t got an answer yet.  [Tyee]

Read more: BC Politics, Environment

  • Share:

Facts matter. Get The Tyee's in-depth journalism delivered to your inbox for free

Tyee Commenting Guidelines

Comments that violate guidelines risk being deleted, and violations may result in a temporary or permanent user ban. Maintain the spirit of good conversation to stay in the discussion.
*Please note The Tyee is not a forum for spreading misinformation about COVID-19, denying its existence or minimizing its risk to public health.

Do:

  • Be thoughtful about how your words may affect the communities you are addressing. Language matters
  • Challenge arguments, not commenters
  • Flag trolls and guideline violations
  • Treat all with respect and curiosity, learn from differences of opinion
  • Verify facts, debunk rumours, point out logical fallacies
  • Add context and background
  • Note typos and reporting blind spots
  • Stay on topic

Do not:

  • Use sexist, classist, racist, homophobic or transphobic language
  • Ridicule, misgender, bully, threaten, name call, troll or wish harm on others
  • Personally attack authors or contributors
  • Spread misinformation or perpetuate conspiracies
  • Libel, defame or publish falsehoods
  • Attempt to guess other commenters’ real-life identities
  • Post links without providing context

LATEST STORIES

The Barometer

Do You Think Naheed Nenshi Will Win the Alberta NDP Leadership Race?

Take this week's poll