Engineer's report finds removal of supports caused Kingston causeway bridge to buckle
Analysis also concludes work plans 'omitted a necessary step'
Workers tasked with repairing the LaSalle Causeway bridge not only removed "critical" supports from the top of a member that helped bear its massive counterweight, they also pulled some off the bottom, reducing its strength to less than half and causing it to buckle, according to a report obtained by CBC.
The 20-page document, titled Failure Analysis of the Lasalle Causeway Bascule Bridge Diagonal 15-17N, provides an expert opinion on the likely cause of the March 30 failure, which ultimately led to the demolition of the century-old span in Kingston, Ont.
The report was commissioned by Public Services and Procurement Canada (PSPC), which owns the causeway, and completed by engineer Colin MacDougall for consulting firm Sigma Risk. While the analysis aimed to provide clarity about why the bridge failed, its conclusions are already being contested.
Representatives for Landform Civil Infrastructures (LCI), the contractor for the project, describe the report as "exceptionally one-sided" and "full of inaccuracies" that favour Parsons Corporation, the engineering firm.
LCI said it was simply following plans prepared by engineers and does not believe it bears any responsibility for what happened.
Parsons did not respond to questions, saying its contract requires it to direct all media inquiries to PSPC.
The report concluded deviation from the work procedure — not following directions in the prescribed order — contributed to the bridge failure.
However, it also found the work procedure "omitted a necessary step" and that drawings prepared by Parsons did not show extra bracing would be needed after the supports were removed.
The buckling ultimately led to the demolition of a bridge relied upon by thousands of people to cross the Cataraqui River daily. It's caused months of headaches for commuters, stranded tour boats and limited traffic to downtown businesses.
Only 41% strength
CBC obtained a copy of the bridge failure analysis through access to information laws.
It references a health and safety incident report prepared by LCI, following consultation with Amson Welding, a subcontractor that was carrying out work on the member when it buckled.
That document, which was previously reported on by CBC, pinpoints the moment on the evening of March 30, 2024, when the failure occurred, while workers were removing a lattice of supports from a member that formed part of a steel truss.
The incident report said the lattice (also referred to as lacing) was being removed from the top of the member, then, as the last piece was removed, it "buckled instantly."
While that report suggested only the top lacing was removed, MacDougall found evidence lacing was also removed from the bottom.
"Rivets have been removed and replaced with bolts," he wrote. "In one case, the ends of the lacing were unconnected and therefore would provide no structural integrity."
The report author described the lack of mention of lacing having been removed from the bottom as one of the "inconsistencies" between LCI's report and his observations, noting "photos taken immediately after the incident clearly indicate that 2.9 m of bottom lacing had indeed been removed."
MacDougall calculated removal of those supports would limit the member's strength to "only about 41% of the fully laced member."
He concluded the immediate cause of the bridge failure was the complete removal of all top lacing then, without installation of replacement reinforcement, "excessive removal" from the bottom.
LCI did not dispute that finding. Instead, vice president John Almeida said the failure happened while it was following plans it received from Parsons.
"There was no limit that said, 'One must remove the top. One must put the plates on top. Then, only then, remove the bottom,'" he said in an interview with CBC. "That is not there."
Despite the member that buckled playing a key role in holding up the bridge's heavy counterweight, Almeida said LCI did not have concerns about the level of support without the lattice.
The contractor submitted a Work Method Statement (WMS) to Parsons for review and "stood by the engineering," he said.
"As far as we're concerned, there was really no reason to doubt that anything beyond that was required."
LCI also shared several emails it sent to PSPC, where it states the engineering firm had "full time inspection/supervision" while work on the bridge was underway, but "never raised a concern" with removing lattice from both sides at the same time.
"I don't feel that LCI bears any responsibility," said Almeida. "There was a plan. There was an identified sequence. We followed the sequence."
Need for bracing not shown in plans
The failure analysis describes the lacing as "critical for ensuring the integrity and strength" of the member that buckled.
That means the steps followed to replace it were critical too, and would require planning to maintain its stability during repairs, the document continues.
The work underway when the bridge failed aimed to replace the lacing with steel plates that would serve a similar function and to install "diaphragms" – lateral bracing between the two pieces that make up a member.
A review of correspondence between LCI and Parsons showed they discussed the WMS for the member that buckled between Oct. 1 and Dec. 7, 2023. The discussions ended after staff at Parsons said they had no further comments.
In the copy of the analysis PSPC provided to CBC, the entire work plan is redacted. However, correspondence between LCI and the department that the contractor shared includes what appears to be an unredacted version of the plan.
MacDougall devotes a section of his analysis to the WMS and whether it was followed, noting differences between the plans and how work seems to have been conducted.
Step 5 reads: "Once all diaphragms are installed remove all existing lattice on bottom plates of the diagonal."
Despite that direction, MacDougall wrote, it appears workers "decided to just start with removal of the lattice on the top." The next three steps should then have consisted of replacing the lattice with new steel cover plates.
"This clearly did not occur," his analysis reads.
LCI alleges directions should have stopped at "remove all existing lattice" and any reference to only taking it off the bottom of the member wasn't supposed to be there.
"It was always intended to remove all lattice top and bottom after the installation of the temporary channel diaphragms … because that was Parsons design to provide stability of member 15-17," its email to PSPC reads in part.
MacDougall's analysis notes it would then be expected that the next step would describe how to take lacing off the opposite side of the member, but that's "missing from the procedure." Though, he wrote, the work hadn't reached that stage when the bridge failed.
"It would appear [redacted] believed that lacing could be removed from both sides of the diagonal without first installing one of the new steel cover plates," the document reads.
MacDougall said he spoke with engineers from Parsons, who emphasized a requirement in their plans that the maximum unsupported length for the member should remain under 1.45 metres at all times.
"The thinking of Parsons was that an external bracing system (or equivalent) needed to be in place to satisfy this requirement," MacDougall wrote.
While that would have prevented the member from buckling, regardless of how much lacing was removed, "this reviewer found that the fact that an external bracing system was needed (and not simply the internal diaphragms) was not shown on the Parsons drawings, nor in the work procedure," he concludes.
In its emails to the government, LCI describes the discussion about a need for external bracing as a "clear admission" by Parsons that they made a mistake.
"The report by Sigma Risk is an important document, PSPC are making decisions on payments and assignment of responsibility for the failure that have serious financial penalties which if inaccurate will lead to pointless litigation," its message continues.
"This report is full of inaccuracies and clearly written to favour or protect Parsons."
Almeida said senior managers with LCI also weren't consulted as part of the analysis.
A LCI worker did speak with Sigma Risk, but it was only a "two to three minute ad hoc interaction," he added, leaving the report without the contractor's perspective.
PSPC not seeking further analysis
A spokesperson for Sigma Risk said PSPC had directed it not to respond to media questions about the analysis or LCI's concerns and instead allow the department to do so.
It's not clear what steps, if any, the government is taking in response to the report.
PSPC declined a request for an interview.
In a brief emailed statement, the department said it "cannot comment on potential financial liabilities or legal matters or aspects relating to this incident or the report."
The message added PSPC does not anticipate seeking any further analysis into what caused the bridge failure.
Molly Gill, director of operations for Amson Welding, said the subcontractor completed its work on the bridge "in accordance with the approved specifications and procedures" provided by the contractor and engineering firm.
Why the bridge had to be 'killed'
Elena Dragomirescu, a civil engineering professor at the University of Ottawa, reviewed a copy of the analysis obtained by CBC.
"The plans should have been clear," she said.
"We're not talking about inexperienced people here. We're talking about companies who work on most of the bridges in Canada, both contractors and the designers."
Despite what happened, Dragomirescu said the bridge failure could have been worse.
"If they would have completed the removal of both lattices on top and bottom, that bridge would have had … victims," she said. "So in a way, it's lucky that it buckled when it buckled."
While removing lacing from the top or bottom first may seem like a small detail, each step of a work procedure would have been carefully pre-calculated, according to the professor.
"Even IKEA, you have to use the instructions in order, right?" she said. "Those kinds of physics we cannot trick."
Dragomirescu said she believes once the bridge buckled it could no longer be saved, adding that trying to salvage it could have had deadly consequences.
"This bridge had to be killed because otherwise it would have killed people," she said.
Temporary replacement to be installed
The causeway was closed for months before the bridge was demolished in June.
Meanwhile, local businesses suffered and drivers have been subjected to daily traffic jams — as costs continue to mount.
PSPC said it paid a demolition company $1.7 million to knock the bridge down and another $3.9 million to set up a temporary replacement. That price tag also covers moving the new span twice to allow boats through.
The replacement is scheduled to open on Oct. 3. By that time the crossing will have been closed for more than half a year.
Meanwhile the debate about who's to blame for the loss of the bridge is ongoing. LCI said it intends to pursue its legal options and has filed a complaint with the Ontario Dispute Adjudication for Construction Contracts.
ABOUT THE AUTHOR
