February 22, 1994

THE TRUTH AT LAST: MANY CAUSES SANK THE L.V. MURROW BRIDGE

BY CLAIR ENLOW

Journal A/E editor

On Thanksgiving holiday weekend three years ago, Seattleite got up in the morning and looked out their lake side windows to find that the view had changed -- dramatically.

The Lacy V. Murrow Bridge no longer spanned Lake Washington. The only reminders of the 50-year-old landmark were the shore structures and a few remaining pontoons at each end. The rest was gone -- swallowed by Lake Washington.

History had sunk, too. The Lacy Murrow was the first major floating concrete bridge in the world when it was constructed in 1940.

The bridge had been undergoing a major renovation designed to bring it up to interstate standards as part of the $1.5 billion Washington State Department of Transportation (WSDOT) project to complete the 7-mile stretch of I-90 linking Bellevue and Seattle. The $30 million renovation of the Lacey Murrow Bridge involved converting the two-lane roadway to one-way traffic and structural alterations to accommodate a breakdown lane.

On the fateful weekend three years ago, Seattle was recovering from a severe windstorm that had whipped Lake Washington and the bridge two days before. And many of the construction workers were taking a brief break from hydrodemolition work on the bridge and the concurrent struggle to pump accumulating water out of the bridge pontoons, where it had to be stored while awaiting envrionmentally-required transport away from the lake.

That day, tugboats began a heroic struggle to rein in drifting breakage and prevent even more disastrous damage by free-floating pontoons.

As work on the renovation of the ``Old Bridge'' came to a halt, lawyers, forensic engineers and public officials set to work building cases for the relative liability of the parties involved: Washington State, as the owner of the bridge, and Traylor Brothers, Inc., contractor, of Evanston, Illinois. Both the contractor and the state attorney general's office commissioned investigations into the causes of the bridge failure.

Investigators immersed themselves in records, plans, notes, memories -- and, of course, the waters of Lake Washington, where the primary evidence lay.

In May of 1991, a blue-ribbon commission appointed by the governor found that both WSDOT and the contractor ``apparently underestimated the risk of the bridge sinking. That June, WSDOT filed a $69 million lawsuit against Traylor Brothers.

The dispute was resolved in August of last year, through a mediated settlement in which Traylor agreed to pay the state $20 million.

Civil Engineering magazine, a publication of the American Society of Civil Engineers (ASCE), published a summaries of the State's case and the contractors in this month's issue.

The following is a comparison of the state's case, as presented by Donald O. Dusenberry, and that of Traylor Brothers, described by Charles R. Firth.

Dusenberry is vice president and senior associate with Simpson Gumpertz & Heger Inc., Arlington, Massachusetts, which was retained by the state to investigate the sinking. Firth is vice president with Ben C. Gerwick Inc., San Francisco, the firm retained by Traylor Brothers to determine the cause of the failure.

The contractor sought to prove that the failure of the bridge was primarily the result of structural weaknesses that accrued over many years prior to the sinking and as a result of the storm immediately preceding.

The state's case is based on proof that the bridge was stressed beyond the breaking point by the particular methods of demolition and construction and not the design of the project, that these methods were mismanaged by Traylor, and that the cracks and structural weaknesses in the structure that accrued over time were not sufficient to sink or even to severely compromise the bridge.

Of particular interest to both investigations was the source of the water that was in or on the bridge at the failure point.

According to Dusenberry's summary of the state's case, ``The sources of the water were the hydrodemolition machines being used to remove concrete from the pontoons, rainwater that fell on the bridge, and, to a lesser extent, wave splash water.''

The state faulted the contractor for not responding adequately to the accumulation of water in the pontoons: Choosing to pump cells that were unevenly distributed, which added to the stress of the weight of the water; and simply not keeping up with the intake of water in the pontoons.

``Although a large volume of water was removed from the bridge, pumping was not systematic or thorough,'' wrote Dusenberry.

Although the state claims that wave splash water entering uncovered hatches the sides of the pontoons contributed to the critical amount of water in the bridge, the contractor, according to Firth, claims that eyewitness accounts of the freeboard on the pontoon A-5, (the first to sink) was not nearly sufficient after the storm to show the deflection corresponding with failure.

According to the contractor, the water sources listed by the state as contributing to the failure of the structure were not sufficient to sink it. Instead, Firth argues for the contractor, existing cracks in the bridge were themselves the source of the water that swamped it, and the cracks widened through cyclical, long and short term stresses on the structure -- not, primarily, through the stresses of construction.

Central to the case of the contractor is the concept of bond break down -- that is, the failure of the bond between the rebar in the concrete in the 1940 structure. According to Firth, the contractors investigating commission studied bending moments and stresses of wave action; signs of fatigue in the bridge and their correspondence with the first reported cracking (in 1960); models of bond-slip according to the nature of the re-bar construction in the Lacey Morrow Bridge and various observations and calculations of the rate of flow through cracks in the concrete.

``The probable cause of the failure was the progressive bond slip at lapped splices in the bottom of pontoon A5, due to fatigue in the bond,'' wrote Firth.

This lead to permanent -- and finally destructive -- widening of the transverse cracks on the bottom slab of the pontoon after the November storm of 1990, according to the contractor's case.

The state's investigation hinged on the distribution and measurements of the cracks in the pontoons that existed prior to the sinking.

It focusses on the relative width of the existing cracks in the bottom slabs of the pontoon, showing that those in the middle (where failure occurred) were not as wide as those at the sides, which did not figure in the sinking. Therefore the state concluded, according to Dusenberry, that existing cracks in the bridge were not the cause of the sinking.

``We concluded that prior cracking was not indicative of diminished strength or performance and that the causes of this failure were not strongly influenced by the mechanisms that formed the cracks in the bridge,'' wrote Dusenberry.

Although several of the causes for the bridge failure were acknowledged by both state and contractor's commissions, their cases are stated in terms of primary causes. For the state, the primary cause was recent.

According to Dusenberry, ``The primary cause of the sinking was water that had accumulated in the bridge pontoons during the (20 month) renovation.''