Several problems may have led to collapse last August of the I-35W bridge across the Mississippi River in downtown Minneapolis. However, based on public information appearing in a recent newspaper study and available at press time, it seems to me that overloading the bridge on the southbound (closed) lanes was the main cause of the bridge failure. Earlier accounts of the deadly disaster pointed an accusative finger primarily at the design and construction features of the bridge. Some accounts, not necessarily submitted by professional civil engineers, stated that the under design of gusset plates was the culprit in the failure. While it may still be proven that such under design contributed to the disaster, recent public information suggests to me that poor construction management procedures during a bridge rehabilitation project may have been the real cause of the tragic event.
According to a story in the New York Times, not only have investigators been "extremely critical of the Minnesota Department of Transportation," but "the 40-year-old span, one of the busiest in the state, had been deemed in poor condition during regular inspections over 17 consecutive years." Forty years is not very old by bridge standards.
This recent investigation is by a Minneapolis law firm retained by a joint committee of the legislature to study the collapse. The findings "were intended not to point to a cause of the disaster, but rather to review the state transportation department’s handling of the bridge." However, many facts about the tragedy seem to point to the construction methods and (possibly) politics as being the proximate cause of the collapse.
Perhaps one of the most telling acts discussed in the New York Times article is that loading on the bridge at the time of the disaster was not only unbalanced, but also may have exceeded reasonable safety limits. A table in the article indicates that loading on the lanes closed to traffic during the maintenance construction project carried a far greater portion of the total load than did that part of the bridge that was carrying moving traffic. According to the newspaper, the total weight superimposed on the open traffic lanes in both directions was only about 137.5 tons, and the total weight on the closed lanes (in both directions) was 408.5 tons.
From the data, it is obvious that the center span alone (a closed lane) carried 289.4 tons. This was apparently the weight of the construction material and equipment placed on the bridge during construction. Also, the southbound (closed) lane alone carried a high percentage of the total load of 578,735 pounds on the bridge. No other lanes, either closed or open to traffic, carried more than 112,200 pounds. It seems clear that the 578,735-pound load was largely due to construction equipment and materials, which perhaps were being stored on the closed lane, at least temporarily, during construction operations.
Of course, I am not privy to details of the loading procedures; that information will certainly be made public when more data is available. But I think further study will show that dead-loading problems in at least one of the closed lanes was the main cause of the failure. If that turns out to be true, then the original design characteristics of the bridge may not have been the primary cause of the failure.
Additionally, torsional considerations could have been a secondary cause of the collapse. For example, if large quantities of construction materials slated for uniform distribution on the structure were placed at one of the edges—the outermost lane or lanes of the bridge spans—the "twisting" tendency may have been the straw that broke the camel’s back.
Engineers, especially those in public works and bridge engineers, are looking forward to more studies and reports that will surely become available. Disasters such as the I-35W bridge collapse are preventable, so it is important that we learn from such events.
For example, in my own area of expertise—hydraulic engineering—we learn through studying catastrophic engineering failures such as the Johnstown flood in 1889. It was caused by the sudden collapse of an earth-embankment dam. Much of the blame for the failure of the South Fork Dam, which initiated the failure of the dam on the South Fork of the Conamaugh River upstream of Johnstown, Pa., has to be borne by poor construction management.
The original design and construction of the dam would, almost certainly, have survived the rainfall event that resulted in its failure if the structure(s) in place at the time the dam was constructed had remained pristine. Years after the dam was built, modifications were made to the outlet structures of the dam, and the dam crest, itself, was lowered. The combination of changes made after the dam was constructed is the proximate cause of its sudden collapse, taking more than 2,100 lives.
The parallel of what happened in Minneapolis, when compared with Johnstown, is striking. Only qualified persons (engineers) should be permitted to meddle with adequate designs that were performed by other engineers at an earlier time.
Did the unknowing hand of a construction contractor cause overload and failure of the Minneapolis bridge? That question will, I am sure, be answered in the future.
Alfred R. Pagan, P.E., is a consulting engineer in Hackensack, N.J. He can be reached at 201-441-9719; or e-mail him at email@example.com.