March 2017, Vol. 244, No. 3

Features

Richard Kuprewicz: The Man Who Knew Too Much About Pipeline Safety 

As the pipeline industry was being thrust into the final days of America’s political drama last fall, environmental activists at Earthjustice reached out to a pipeline safety icon known for his expertise and independence, Pacific Northwest-based Richard Kuprewicz.

The environmental group’s support for the Standing Rock Sioux Tribe’s opposition to the $3.8 billion, nearly 1,200-mile Dakota Access oil pipeline project prompted the call to Kuprewicz, who chose to be a paid third-party source on the safety review of the project’s underwater pipeline crossing.

Kuprewicz, president of Redmond, WA-based Accufacts Inc., is a chemical engineer with an MBA and a deep resume dating back to the early 1970s with some of the ARCO companies. He has earned the respect of both industry and government clients for his judgment and expertise in the world of pipeline integrity management programs. He has been part of investigations of recent pipeline ruptures and participated in advising the federal government’s Pipeline and Hazardous Materials Safety Administration’s (PHMSA) latest attempts to craft updated rules for gas transmission and gathering pipelines.

Kuprewicz ended up writing a 10-page technical review that was critical of an environmental assessment (EA) that the U.S. Army Corps of Engineers (USACE) did initially in reviewing and granting an easement for the water crossing by a unit of Texas-based Energy Transfer Partners (ETP) that built the Dakota Access pipeline.

“I have concluded that the EA is seriously deficient and cannot support the finding of no significant impact, even with the proposed mitigations,” Kuprewicz wrote in his summary. “Important details are missing.”

He submitted his assessment to an Earthjustice official in Seattle, but the Sioux tribal leaders ultimately told the government that they had commissioned the third-party report, which didn’t sit well with supporters of the oil pipeline.

“I had no interaction with the Sioux on this matter; I was paid to do the analysis, and no party gets to change my independent technical findings and observations in my reports,” said Kuprewicz, offering a glimpse into his ever-challenging world of pipeline safety analysis and interpretation.

A complete and prudent risk analysis can’t cut corners; it has specific minimum types of information and analysis, Kuprewicz said, whether referring to Dakota Access pipeline or generically to hundreds of other pipeline projects. Any analysis should include route/design/operation/maintenance activities to make the risk-analysis credible and the oil-spill response plan “likely to be effective if ever needed,” he said.

Kuprewicz emphasizes the need for independent verifications of equipment placement/types, operational procedures, and finally integrity management applications and effectiveness through application of a seven-part checklist for any particular pipeline:

• Pipeline elevation profiles (approximate elevation vs. milepost for the pipeline segments between the nearest upstream and downstream pump stations).

• A line indicating the maximum operating pressure (MOP) on the elevation profile.

• A hydraulic profile at the design rate case on the elevation profile.

• Location of mainline valves and their type of operation (manual, remote, automatic)

• The general locations/types of critical leak-detection monitoring devices by milepost.

• Identification by milepost range of high consequence areas (HCAs).

• Given the many pipeline failures following inline inspections (ILI), further requirements are warranted on the type of ILI tool to be run, its frequency, and tool limitations for the segments that could threaten or affect the surrounding resources/environment.

“Knowing the possible threats is very important and is the foundation of integrity-management regulation,” he said. “More importantly, since MOP is usually set, you mainly want to avoid an anomaly that can grow until the pressure, even if it is reduced, causes a rupture.”

He said the industry carries its own myths that can prove harmful. An example is that certain anomalies are stable, which Kuprewicz considers “very dangerous,” depending on the type of threat and the material being moved within a given pipeline.

In today’s world of social media and around-the-clock news cycles, the responses of energy companies to basic integrity-management rules created a decade earlier varies greatly. Regulators have vacillated between prescribing detailed safety steps and requiring companies to upgrade operating performance to meet higher standards. The difference is mandated actions vs. performance-based rules.

“It can vary among companies and within different parts inside a single company,” said Kuprewicz, noting the relatively long time frame for PHMSA to develop a notice of proposed rulemaking (NPRM) for transmission/gathering pipelines was attributable to the cognitive dissonance still lingering in the industry. ”There are many people who get it and stay ahead of the curve, and unfortunately there are way too many others, based on the many public tragedies we’ve all seen, who don’t.”

The advent of increased merger-and-acquisition activity in the face of the two-year low commodity price environment has added to the problem in Kuprewicz’s opinion, resulting in some companies “losing their focus and series of internal checks and balances” needed to stay well ahead of pipeline failures. As a result, he believes PHMSA has had to come up with expanded integrity-management requirements and more prescriptive approaches to pipeline safety.

As a result of the spate of failures in recent years, PHMSA has sponsored several studies that have shined a useful light on the core bases for some of the problems, and the federal agency is incorporating those insights in the NPRM that Kuprewicz has helped influence through his analysis and review. He is hoping “good, effective regulatory guidance” evolves from the process that culminated in 2016, noting that gathering pipelines – long left out of the regulatory mix – are being included in the new rules.

“Not all gathering pipelines take on the level of transmission pipelines, but many of the gathering lines now are sized and operating at pressures akin to a transmission line would,” Kuprewicz said. “No matter what they are called, they will fail based on their threat characteristics, just like a gas transmission pipeline.”

With the advantage of being in close touch with the incidences and players involved from time to time, safety engineering experts like Kuprewicz can position themselves to gain insider information on what is being done right or wrong when it comes to safety.

“I always tell people to never underestimate the ability of a group of very smart people to make the stupidest decision, if they have the wrong motivations,” he said. “There are many companies that do it [integrity management] with wisdom, and then there are ones so focused on cost reduction they raise the risk of a disaster.”

A common trap is to start believing internal “myths” that crop up in various organizational cultures, according to Kuprewicz, who encourages clients to question their own mindsets. Operators, he thinks, need to be careful not to overstate their capabilities within their own teams and to the general public. Stay hungry and questioning seems to be his mantra. Resist complacency.

As an example, even though Kuprewicz is a long-time supporter of ILI, he urges operators not to overstate the capabilities of this tool within their own teams and to the general public.

“Inline inspection has made tremendous advances, but I have seen in too many of the rupture investigations that an ILI tool had been run, and the pipe still failed by rupture shortly after the test,” he said. “I’m not talking about 10 years after, but in really short periods of time.”

Kuprewicz has found a tendency to “oversell” ILI capabilities. “We want to foster the advances, but need to ensure operators are running the right ILI for the right threat,” he said, encouraging more questioning of whether the selected tool and threat have been correctly identified and properly evaluated by the engineering assessment. In Canada, there are expectations being made that ILI tools can detect all of these possible causes of ruptures, he said.

“We want inline inspection to work; it works really well depending on certain types of threats, but fundamentally, industry needs to acknowledge in making decisions that ILI isn’t ‘free.’ Cheaper than other options, but there are places where inline inspection just can’t do the job,” he said. “The rash of pipeline ruptures after ILI – Marshall, Mayflower, Refugio, PA, etc. – had a combination of poor ILI detection and very bad so-called ‘conservative’ engineering assumptions that were anything but conservative.”

Part of the ILI effectiveness verification is tied to efforts to advance certain types of crack detection. Inline inspection tool work needs to be made public regarding the industry standard (API 1163) calling for “unity plot” or “graphs” to double-check the ILI tool’s findings. This compares what the smart (ILI) pig uncovered with what was found externally in terms of corrosion, tracking and damage to pipe.

“Nobody is selling inline inspection tools claiming they don’t work, and if you have a good story somewhere along the line, you have to make certain information public,” Kuprewicz said. “So I see more of the unity plot/graph information being made public, and this could set the public’s mind at ease by demonstrating that the tools are actually working. It is when they don’t make them public that problems arise.”

“Unity plot/graph” is an API-recommended practice requiring specified numbers of field verification digs to verify the ILI’s identification of certain corrosion or other problems.

“I have seen some recent versions of various ILI runs where they have misused unity graphs comparing one ILI run against another, introducing built-in bias from previous poor ILI; it just isn’t that complicated,” Kuprewicz said. “Remember, after one gets the ILI runs right in the determination department, a pipeline company can still mess up time-to-failure estimates from the pig calls via very poor engineering critical assessments.”

As a young engineer, Kuprewicz encountered a seam failure that in retrospect could have wiped out a refinery that was part of his responsibility. He and his colleagues were able to shut it down before the failure did any serious damage. That lifelong lesson provided him with a greater understanding of cracks and failures involving electric resistance welding (ERW). Because there has been a long history of seam failure in the industry – both low- and high-frequency – most parts of the industry have a good grasp of the risk.

“It is the basket of cracking failures, of which ERW is just one, and others would be environmental ones, such as SCC [stress corrosion cracking], or transportation cracking. Depending on how you have done hydro tests, you could leave some large cracks that could grow and come back to haunt you,” he said.

Generally, seam failures are the most difficult to ascertain and evaluate with the latter being the most important aspect, Kuprewicz said. “There have been industry standards issued if you have a certain type crack or crack family, but in many cases the guidance is incomplete, and if the operations people are inexperienced, you could be running recommended practices and not coming to the best solution that protects the company.”

He labels cracking as the industry’s biggest challenge, noting that it comes in various forms. Ideally, he would like the industry to be more public about what is being discovered in terms of tools and capabilities, including advances in engineering assessments.

The base for all this is the science underpinning regulations and industry best practices, and while the U.S. industry has led many advancements in recent years, the tool kit is still incomplete. Inherently, what Kuprewicz calls the “linearity” of the pipeline industry compared to other hazardous cargo conveyances, such as rail and trucking, creates bigger integrity-management challenges. Still, today’s advances can allow operators with the right quality assurance and quality controls to virtually eliminate the risk of cracking in their new pipelines, he said. Most likely, cracking will remain an issue for older pipelines (5 to 10 years in age), and definitely for the vintage ones.

There are key parameters for engineering critical assessments (ECA), but these have to be understood and applied, according to Kuprewicz. PHMSA has incorporated them into the NPRM in areas such as vintage pipeline ERWs, he pointed out.

An ECA identifying corrosion above 80% wall loss would be in violation of the new PHMSA regulations, assuming they get published. Anything with at least 80% wall loss is assumed to have corrosion in the new regulations. API and the American Society of Mechanical Engineering (ASME) standards already make this clear, said Kuprewicz, noting “all bets are off if you’re over 80% on general corrosion calculations.” Nevertheless, he said some companies will try to justify situations in which a pipeline is well beyond 80% wall loss for general corrosion.

Environmentally based cracking or “stress corrosion cracks” (SCC) has a lot more science backing it, much of it driven by the types of pipeline coatings deployed, asserted Kuprewicz. “If a gas pipe has a basket of coatings close to a compressor station, it might be a candidate for SCC cracking.

“You can check for it, and the environmental conditions needed for SCC might not be present in your case. It is an issue that is not difficult to ascertain and defend in an integrity-management program. A number of companies, especially on new pipes, can design, install and transport their pipe in such a way as cracking and corrosion aren’t there,” Kuprewicz said.

Use of the more durable “fusion-bonded epoxy coatings” can essentially eliminate the threat of SCC with the new pipe. “Companies are doing that – not all of them – but many are,” he said.

The scenario today calls for companies to apply effective QA/QC programs on large-diameter D/t ratio pipe whether they move it by truck, rail or barge. This avoids placing extra stresses on the new pipe that can introduce cracks that can survive a 1.25 times-MOP hydro test. Similarly, operators during construction will take precautions in which pipe is not placed or damaged by rocks that can foster dents with stress risers (cracks, corrosion or gouges).

“It currently is an illusion that ILI can find such threats and that engineers can predict time-to-failure for such anomalies from ILI data,” Kuprewicz said.

Pressures and cracks muddy the operating waters, and he warns that pressure reduction is not a substitute for integrity-management solutions. He cites the recent example of Enbridge Energy Partners LP’s 30-inch oil pipeline rupture and spill into the Kalamazoo River in Marshall, MI as an incident involving two different types of pipeline cracks. It failed at very low pressures and even very low percentage specific minimum yield strength (SMYS) tests – what Kuprewicz calls “way below MOP.”

Operators often set MOP at 110% and operate at 100%, but recent pipeline failure investigations turn up ruptures that happened at 50% MOPs.

In Kuprewicz’s world, the pieces are all available to virtually eliminate energy pipeline incidents but putting them together correctly has proved difficult, given conflicting forces in play. Some of them may eventually be eliminated by the new Trump administration. Climate change initiatives, split regulatory duties and insufficient training are some of these forces.

Kuprewicz has seen the effects of the climate change push on pipeline safety from his past roles with clients on both sides of the issue. “It is not illegal to leak natural gas right now, although it is a greenhouse gas (GHG). We’re not trying to stop all leaks; we’re zeroing in on hazardous leaks. Safety vs. GHG has to find a balance. But as more information has come out, people are beginning to realize that this could hamper safety efforts. We have to be careful here.”

When there is a pipeline rupture, its release will be many times that of what comes out of a leak, said Kuprewicz. “A rupture becomes a ‘super-emitter’ for methane releases,” he said. “All parties should be aware we’re not going to stop all leaks, and everyone should go after the super-emitters, who are readily easy to identify – for the production, transmission and distribution sectors.”

Another “force” needing to be recognized is the regulatory split between FERC and PHMSA, a safety – not siting – agency, and yet some of Kuprewicz’s clients regularly mix up these roles. “I can understand the disconnect here; I’m comfortable with that,” he said. “Intervenors at FERC, the siting agency, don’t necessarily inherently understand the difference between the two.”

Integrity management is not a siting issue except for identifying high-consequence areas, he said. “The emphasis of integrity management is to get to those failures early, well before they go to rupture.”

In the NPRM there will be more “prescriptive guidance” that has been identified as needed, given some of the recent failure investigations, Kuprewicz said Is he optimistic? Only cautiously.

Richard Nemec is a Los Angeles-based correspondent for P&GJ. He can be reached at: rnemec@ca.rr.com.