Surveyors implement a 24-hour safety net after the Oklahoma I-40 bridge disaster.

ODOT engineers and surveyors stood on the bridge deck searching for signs of fatique that might signal futher collapse of the bridge.
At 7:30 a.m. on May 26, 2002, one of two barges traveling north on the Arkansas River slammed into a support of the I-40 bridge near Webber’s Falls, Okla., disrupting the quiet Sunday morning. The resulting impact caused one end of a nearly 600-foot section of the bridge to drop to the river below, taking unsuspecting cars, trucks and trailers with it. When the dust settled, 14 people had died and a large portion of the collapsed bridge section rested on the deck of one of the barges.

Emergency teams and nearby good Samaritans mobilized immediately to help pull survivors from the water. Engineers and surveyors from the Oklahoma Department of Transportation (ODOT) moved onsite to assess the integrity of the remaining bridge sections and the potential for more damage, and to ensure the safety of rescue teams. Their timely efforts and quick thinking provided a much-needed safety net for rescue and recovery teams, and set the stage for speedy reconstruction.

As a team of divers battled the murky waters searching for survivors, ODOT engineers and surveyors stood high above on the bridge deck searching for signs of fatigue that might signal further collapse of the I-40 bridge. It was feared that the slightest shift—even a fraction of an inch—in the barges, the buckled section of bridge or the still standing bridge sections would cause the deck to collapse onto emergency teams and survivors below.

While engineers carefully studied and analyzed the bridge for structural integrity, a team of ODOT surveyors had mobilized onsite to offer support. Per instructions from the bridge engineer, they were asked to set up a continuously operating control network to monitor the movement of the bridge for the entire rescue, recovery and demolition efforts.

Gordon Johnson, PE, PLS, chief of surveys for ODOT, recalls, “We didn’t have time to set up a control network and run static adjustments. These guys needed immediate information in order to give divers and other safety teams time to evacuate if necessary.”

Fortunately, ODOT had recently completed several projects in which comparisons were made between their Static GPS Network Adjustments and the positions obtained from the NGS OPUS (Online Positioning User Service) program. The two values differed by approximately two-hundredths of a foot on each of several projects, indicating that OPUS accuracy would be suitable for controlling projects of limited length.

They also knew they could rely on some in-place surveying equipment. The I-40 bridge at Webber’s Falls is routinely monitored for shift and is equipped with a series of reflectors along the entire span. On this first day of recovery and reconstruction, the surveyors recovered control monuments and tied the reflectors on the bridge piers.

Meanwhile, Larry Reser, PLS, assistant chief of surveys, and Johnson continued to work from the ODOT office in Oklahoma City to gather information for the National Transportation Safety Board and work with bridge engineers via E-mail and telephone to develop a comprehensive support plan. The plan included the need to develop profiles, match bridge abutments and monitor the barges for movement in three dimensions—horizontally and, more importantly, vertically.

Darin Stratton, PLS, takes observations of targets positioned on the barges and the bridge structure.

Continuous Operation In Minutes

On Monday, Reser and Johnson joined the other surveyors on the bridge to put in place a continuously operating positioning network. The overall network consisted of a GPS base station from Leica Geosystems (Norcross, Ga.) located about a quarter of a mile away on the east bridge abutment that surveyors had established on Sunday. Using points on the bridge deck that the bridge engineer had previously selected, the surveying team used autonomous positions to monitor selected points on the part of the bridge that was still standing.

Although the “true positions” of the points were unknown, relative to the NAD83 datum, their positions relative to the base could be determined very accurately—within two-hundredths of a foot. “We set 30 reflectors for monitoring the movements at selected points near the end of the cantilevered bridge section and at the tenthpoints (one-tenth of the span),” Reser says. “The bridge engineer also requested continuous monitoring of the still standing piers near the collapsed portion.”

Within one hour of arrival, surveyors were able to provide bridge engineers with position data.

Knowing the need for hourly information would extend through the night and the following days, the onsite surveyors split into teams of two for 12-hour shifts until other surveyors could arrive from around the state to help. “Between readings, these surveyors rested in their cars on the bridge deck less than 100 feet from where the bridge buckled,” says William (Leroy) Tackett, PLS, transportation survey manager and one of the first onsite. They brought coolers filled with food and drink so they didn’t have to leave the site between the one-hour readings.” Again, they continued to provide data to the bridge engineers on an hourly basis, day and night.

This information was particularly critical as the emergency teams and engineers set about trying to secure the barges and the collapsed portion of the bridge. “It seemed too simple,” Tackett says. “In a matter of minutes, we set up a base and set the GPS unit to record the reflector positions. The repeatability was impressive. We usually don’t shoot over and over with RTK—but with this framework, we were able to shoot those eight sets of points, and they never varied more than three- or four-hundredths of a foot.”

The monitoring continued on an hourly basis 24 hours a day for the following four days.

Kyle King, PLS, records observations in a field book.

Demolition Data and OPUS

On the night of the fourth day, the demolition teams were ready to begin. The survey team moved from the bridge deck to the west abutment and re-established survey control of the bridge and barges with previously placed reflectors. Using RTK GPS, they continued monitoring the bridge movement every hour, and during the most critical times, every 15 minutes using a Leica TC1000 total station from dusk to dark.

When it became apparent that the survey data would be critical for obtaining temporary easements for the salvage and reconstruction effort, the surveyors looked for a quick conversion to State Plane Coordinates. OPUS offered an easy, accurate and quick solution.

As the survey crew collected the position data, they downloaded the RINEX file to a PCMCIA card, and then sent the file to the NGS web page where it was processed automatically with respect to three National CORS stations. “The OPUS position for the base is E-mailed back within a few minutes, allowing us to rotate our autonomous positions to fit the OPUS-derived position. Since the reconstruction would have to match the existing bridge, we used the west bridge abutment as an elevation source,” Reser says.

Immediately following rescue and recovery, ODOT put a special team together to set up detours and begin planning for reconstruction. A special pre-bid meeting was called at the bridge site, bids were opened and the contract was awarded to Gilbert Central of Fort Worth, Texas. ODOT surveyors continued to offer support in the way of engineering and boundary surveys throughout the rebuilding effort.

When the demolition stage began, the surveying team epoxied four reflectors to the bridge deck, one on each side above the last standing pier and one on each side near the midpoint of the cantilevered span, located at the extreme west end of the portion of the bridge still standing. Other reflectors were epoxied to the four corners of the barge that had the bridge section resting on it. The surveying team followed up by setting control points on the west side of the bridge near the abutment using RTK GPS. From this point forward, the bridge was monitored using the total station, measuring to the reflectors from these control points.

After much responsive and accurate work, the bridge reopened on Monday, July 29, 2002, just two months following the barge accident.

Using previously established points, the surveying team used autonomous positions to monitor selected points on the part of the bridge that was still standing.
For the ODOT survey team the lessons learned were many. “We had a total of 17 surveyors rotating around-the-clock efforts for approximately two weeks following the bridge collapse,” Johnson says. “Most of us had never been involved in something like this that required such quick mobilization. We’ve learned a lot about the power of our technology and our role on the support team.”

As a follow up, ODOT has proposed the installation of a continuously monitoring system for all bridges on the Oklahoma navigation system that are currently being monitored for movement or settlement. This system will be monitored from a PC located in the Bridge Division offices and is a real-time digital messaging system that reflects any movement of those structures.

In addition, the survey division is currently looking into a cell phone/GPS link, which would greatly reduce the cost of GPS equipment purchases and hopefully increase productivity, particularly during emergencies. ODOT is also working cooperatively with NGS and the ODOT field divisions to establish CORS stations at all of the ODOT field offices.