Plans to reinforce the bridge were well underway when the project came to a screeching halt in January amid concerns about safety and cost.
The men and women whose job was to ensure the safety of Bridge 9340 were meeting once again. Just after noon on Dec. 6, they filed into a conference room in Roseville to divvy up the final prep work for a dangerous steel reinforcement project high above the Mississippi River.
A senior engineer was going to pull property records in order to contact landowners beneath the bridge. Detours were coming for West River Road. The Coast Guard was about to get heaps of paperwork on what tasks would be done from the river channel. Truck drivers would soon learn of pending weight restrictions.
It appeared that the most studied bridge in Minnesota, the focus of worrisome inspection reports for a decade, was finally going to have its most glaring weaknesses fixed.
But five weeks later, all those preparations stopped. In a single conference call on Jan. 17, the same consultants who said reinforcement plates were needed to strengthen the bridge cautioned MnDOT that drilling for the retrofit could weaken it.
"That was the turning point. That's where we turned the ship 180 degrees," said state bridge engineer Dan Dorgan.
Internal MnDOT documents reviewed by the Star Tribune reveal that last year bridge officials talked openly about the possibility of the bridge collapsing -- and worried that it might have to be condemned.
The documents provide the first look inside MnDOT's decision-making process as engineers weighed benefits and risks, wrestling with options to prevent what they believed was a remote but real possibility of the eight-lane freeway bridge failing.
Their concerns were not generalized, documents show. The San Francisco-based consultant, URS Inc., identified 52 crucial steel box beams deemed most susceptible to cracking. URS also had a specific recommendation that 24 of the 52 members be reinforced while the remainder would be kept on a special watch. Video of the Aug. 1 collapse being examined by the National Transportation Safety Board shows the bridge first falling on the south end over its shoreline pier -- a section of the superstructure where eight suspect beams were specifically tagged for reinforcing.
Dorgan and senior engineer Gary Peterson denied in interviews that money was a factor in deciding what to do with the Interstate 35W bridge, which was not due for replacement until 2022. They provided a written timeline showing that MnDOT supervisors on Nov. 1, 2006, funded the reinforcing project for $1.5 million, with work to begin in January 2008.
But at least three internal documents suggest that money was a consideration.
On Jan. 18, one day after MnDOT's Bridge Office opted to inspect rather than reinforce the bridge, Peterson apologized to an engineer in the department's Metro Design section that work put into the reinforcing project was for naught.
"We regret the additional work this has caused you and others in the district," Peterson wrote in an e-mail, "but I'm sure you agree that based on this new information it [is] appropriate that we postpone the project until we can determine if another option may [be] as safe and a more cost effective approach."
Earlier, when MnDOT and its consultants were zeroing in on reinforcing the bridge, an internal MnDOT "investment strategy" meeting was held on July 24, 2006, in which officials debated various approaches.
According to meeting minutes, officials said that immediately installing steel reinforcements would greatly reduce the risk of a crack forming "between now and 2022." That way, the agency could pick the ideal time and circumstances for carrying out the work.
The "risk" of that approach was described this way: "Must pay approximately 2 million dollars to get the job done."
A logistics and financial issue also was discussed. MnDOT officials said that if the bridge was simply inspected, the benefit would be: "Don't have to pay for steel, stockpile steel, or install steel."
The downside of such a decision, MnDOT officials acknowledged in the meeting, was that "If a crack is found it will take 4 months to order steel and reinforce the bridge, and the bridge will be closed to traffic for this duration. But there is a further risk that the damage is beyond fixing, and the bridge will have to be condemned. This means 35W will be closed for a minimum 5 years until the new bridge is finished."
The harshness of Minnesota winters also came into play. Minutes show that MnDOT officials saw a benefit to immediately reinforcing the bridge because it could choose the exact time, conditions and method of work.
"If we wait until an inspection finds a crack before we reinforce the steel, then random chance and weather will dictate ... This will negatively impact cost, quality and safety," according to minutes of the July 2006 meeting.
MnDOT officials didn't dispute the accuracy of the minutes. But they pointed out that the risk-benefit analysis was undertaken before URS made assurances in January 2007 that cracks could be found and neutralized before they reached a dangerous length. URS has declined to answer Star Tribune questions since issuing a brief statement in the days after the bridge collapse.
And in interviews, both Dorgan and Peterson scoffed at the notion that a $1.5 million expense to fix one of the state's busiest bridges was an issue in a department that annually spends hundreds of millions. Overall, MnDOT spends about $2 billion a year.
A month after the "investment strategy" meeting, Dorgan wrote a memo in response to the first draft of URS' written crack study of the I-35W bridge. Dorgan didn't want the consultant to mince words when it came to explaining what would happen if a fracture-critical beam in the superstructure failed.
"If the conclusion is the instability would likely lead to the collapse of the bridge, that should be stated clearly," Dorgan wrote.
It was part of the extensive give-and-take that went on between the consultant and MnDOT ever since URS received the contract in July 2004 to identify the structural members of the main spans of the bridge that were most susceptible to cracking.
For instance, in September 2006, three months after URS issued the first draft of its final report, MnDOT asked the consultant to verify the toughness of the steel used to build the bridge. The issue of toughness was raised because it affects how quickly cracks could spread and the time inspectors would have to find and react to them, according to documents.
But Dorgan said MnDOT never got an answer before making its decision to proceed with the inspection option.
According to records and MnDOT's timeline, here's how the agency turned, as Dorgan said, "180 degrees" to embrace inspections over reinforcements:
June 2006 URS recommends reinforcing 52 truss members most susceptible to cracking.
Mid-October 2006 The Bridge Office requests money for reinforcing and the project is funded with $1.5 million.
Late November 2006 URS is told to make plans for a contract letting in October 2007.
Dec. 19, 2006 URS informs MnDOT that ultrasonic inspections are an alternative to installing reinforcing plates. URS sends a draft of revised recommendations to MnDOT. The three options are to reinforce 52 beams, inspect the beams with ultrasonic equipment or a combination of reinforcing 24 beams and inspecting the other 28.
Jan. 17, 2007 The turning point occurs during a conference call. Dorgan and staff opt for inspection only. He says the decision is based on URS assurances that inspectors can detect and isolate cracks before they reach a dangerous length.
Jan. 18, 2007 Gary Peterson tells MnDOT's Metro Division that the plate-installation project will be delayed until at least fiscal year 2008-2009.
May 2007 Inspections of critical beams begin on half of the designated beams. Inspections were halted when concrete repairs began and were to resume in the fall.
According to the MnDOT timeline, engineers were scheduled to meet Aug. 20 to determine if inspections should continue or if they should go back to the plating concept. The bridge collapsed Aug. 1, leaving 11 people dead and two missing. The National Transportation Safety Board is investigating the accident and has made no preliminary finding on what may have caused the collapse.
Dorgan reflected on MnDOT's change of course when it came to steel reinforcements for the bridge.
"You can't help but ask yourself ... what should have been done differently," he said. "As an engineer you can't be at peace until the cause is found. And even then I have doubts that will bring peace."