In the history of United States, many structural failures happened due to geotechnical conditions and unfortunately those failures caused billions of dollars and more importantly lives of the people. The failure in our case was a bridge, Schoharie Creek Bridge that was located in New York collapsed due to bridge scour at foundation level after a heavy flood. The bridge was built in October 1955, its 155 meter span that was fitted onto 5 piers. The bridge was also supported with pier frames along with abutments at each end. The bridge was collapsed due to bridge scour that caused the pier 3 to crumble that tragic failure took 10 lives of people. Snowmelt combined with spring rainfall performed a 50 year record flood. That flood was the last straw of the foundation failure of the bridge. Bridge scour is move of rock and soil particles that covers the foundations and piers of the bridge due to high velocity water flow under the bridge. When designing a bridge foundation, it’s crucial to estimate scour depth to not expose abutments. One of the main causes of bridge failure is bridge scour. According to Bridge Scour Data Management, %60 of all bridge failures caused by scour and other hydraulic related issues[1]. So, it’s important to design a bridge that is located in a place like New York that experiences heavy floods. The main object of us engineers is considering the extreme situations while designing to minimize the catastrophic failures like the collapse of Schoharie Creek Bridge that took 10 lives and caused financial damage to the community.
Introduction
Bridge failure is a common catastrophic disaster in the history and they result in loss of human life, expensive repair costs and disturbance in daily life. Schoharie Bridge failure was one of those tragic accidents. Schoharie Creek Bridge was composed of steel girders and a system of floor beams and four concrete piers and two abutments. Piers were made of columns connected to a main beam and bottom of the columns were framed into a concrete shallow footing. On the day of April 5, 1987, one of the piers collapsed due to scour mechanism, the soil around the footing carried away due to stream erosion. Due to these unfortunate natural events plinth and the footing collapsed due to fracture failure. Since it was a fractural failure span 3 and span 4 collapsed immediately and span 2 collapsed within hours. The results were catastrophic, 10 people died and financial cost was millions of dollars. The catastrophic flood was later determined that it was a 50-year event with the maximum flow rate of 1800m3/s. [3] It collapsed approximately 24m to the overflowing creek, at the time of the incident 1 car and 1 trailer was on the bridge and before the authorities blocked the road, 3 cars flew to the river. 10 people died in the result of the failure. It’s said that riprap stones were not enough to resist the flood. Also, Schoharie River was known for ice jams and freezing conditions over the winter. Other studies and laboratory work stated that ice impacted and damaged the riprap layer over the years that contributed to the scour failure. Data from 1956 to 1986 states that over those years Schoharie River usually covered with ice with the average thickness is 9.56 inches. [4] So, we can say that even though the main cause of the failure was extensive scour, ice effects played a crucial role as a contributing factor.
Background
The Schoharie Bridge was built for the Thomas E. Dewey state thruway project for NY. The main purpose of the bridge was to cross the thruway traffic on the Schoharie Creek. It was just one of the bridges constructed by the NYSTA for I-90 which connects Buffalo to New York. The bridge didn’t have any problems until the disaster struck on April 5, 1987. NYSDOT contracted out Madigan-Hyland Consulting Engineers for the design job. The design firm used the 1949 edition of the AASHTO “Standard Specifications for Highway Bridges” for the preliminary design of Schoharie Creek Bridge. The preliminary designs were nearly the same except for the span length, one of them was 600 feet and the other one was 540 feet. Shallow footings on the piers were used in both plans to support the structure. NYSDOT approved the plan for 540 feet bridge span. It was submitted in January 1952 and it had five simply supported spans with the length of 100, 110, 120, 110 and 100 feet. The bridge spans along with abutments at each end that were supported by concrete pier frames. Also, the columns had a lightly reinforced plinth, which was located on a shallow reinforced spread footing. The spread footing had a protective layer of dry riprap. B Perini and Sons, Inc. started construction on February 11, 1953. The design firm performed construction inspection for the bridge. About a year after the bridge was opened to partial traffic. In 1955, the bridge got through a 100-year flood, but the damage of that flood may have had an effect on the collapse in 1988.
Events Leading to the Failure
The largest flood that the bridge was experienced was in 1955 and the second largest flood was in 1987 which was the year the bridge was collapsed. The velocities of those floods were 1758 m3/s and 2084 m3/s respectively according to National Transportation Safety Board [1]. The conducted investigations after the failure showed that a riprap was placed in 1955 to protect the footing from the scouring. After the flood in 1955, bridge pier plinths started to form vertical cracks. The width of those cracks were ⅛ to 3/16 inches, and the locations of the cracks were different from pier to pier. Cracks have formed because of the high tensile stresses in the concrete plinth. The plinth simply didn’t have enough resistance for the bending stresses between two columns. There were different problems that happened shortly after the completion of the construction. Inspection teams saw that the expansion bearings were out-of-plumb, roadway approach slabs had settled, drainage from the roadway was pretty poor, and the material that supports the west embankment dry stone pavement was deficient. But these small issues were fixed by fall of 1957. The photographs of the bridge in the 1970s showed that there was a movement at the rocks of the riprap possibly due to heavy flood in 1955. It is believed that the heavy flood in 1987 moved the remaining riprap away because if that was not the case the bridge would collapse before the events of 1987. When the footing was exposed due to the flood, it collapsed rapidly and caused a structural failure.
The Collapse
In 32-year lifespan, the bridge was in service, normally it was designed for the possible events including floods. Investigations show that the failure has been roughly caused by scouring, which affected the stability of the bridge piers, on the other hand, the event raises questions about inspection practices which have been done. It was assumed that it was constructed to specifications and also assumed that there were no major repairs and maintenance. In today’s world, bridges get inspected often and thoroughly but at that time inspection standards were absent.
Collapse is diagnosed as a system failure. The failure of the span 3 was a subsystem consisted concrete floor, main girders, and floor beams. The fall of the abutment and pier of the span 3 caused the failure of the pier 2 and span 2. One car and one-tractor-semitrailer were on the bridge when it collapsed [3]. But these cars weren’t the only ones. Before they can close the road, three more cars fell into the gap. Authorities found nine dead bodies. Unfortunately one body was never found.
Span 3 started collapsing first. In ninety minutes pier 2 and span 2 fell. After two hours pier 2 collapsed, pier one and span 1 shifted. Pier 2 collapsed because the wreckage of pier 3 and the two spans partially blocked the river, redirecting the water to pier 2
and increasing the stream v
elocity.[5]
Reasons for Failure
The main cause of the failure is flood. The second cause of the failure is that trucks not abiding truckload limitations. Also, some failure is due to the fatigue of the structural members of the bridge. Scour has always been the number one threat to the bridges on rivers. It is the number one cause of the bridge failures. Scouring is displacement of river bed material brought with the increasing flow rate and velocity of the flood. When in extreme cases those high velocity and flow rate could even displace large rocks with an immense drag force. In the case of Schoharie Bridge, the flow was directed downwards by the bridge piers and created a jet effect which removed more foundation material.
The reason that the problems of the bridge were not detected in the inspection because at the time of the inspection deep scours at bridges occur during flooding and by the time when floods subsided scour holes gets filled either partly or completely and this makes it undetectable by the inspectors. With today’s technologies radars and sonar equipment are used to measure scour holes but at that time those technologies were still at experimental.
Who was to Blame?
There was a court about the failure of the Schoharie Creek Bridge. In the court, it is said that underwater activities surrounding the bridge was involved in the collapse and scouring phenomenon was unique to the bridge and it was really hard to detect with an ordinary inspection. Even though AASHTO had a bridge inspection program at every 5-years but the importance of the underwater inspections was given in 1986 which was one year before the collapse. Normally they should be underwater inspections by sending a scuba diver to check the underwater bridge parts or by doing sound readings. It is said that of the 575,000 bridges in all of the US, 130,000 of them has some kind of defect. [2] Because of the flood, it is hard to detect the scouring of the footing due to high water levels.With the 2 days of heavy rain and the flood Schoharie Bridge collapsed and one truck and four cars carried the 10 victims to their deaths. There is a number of factors that could be blamed. It could be the construction of the bridge, high flow rate of the creek, the soil and rock type on the river bed, topography of the location, banks of the creek and another number of things. In New York State a 1 to 7 scale was used to group the bridges. 7 was a new bridge and 0-3 were hazardous bridges. Schoharie Creek Bridge was rated 5. The bridge was built in February 1953 and opened to traffic in 1954. There was a political pressure to finish the project on time and there was a repair contract after the 2 years when the bridge was opened to traffic. But, since the bridge was survived by a hurricane that caused a severe flood that is much wilder than 1987, it can be concluded that there was no fatal weakness during the construction process.
Investigation and Aftermath
NYSTA contracted two teams to investigate the Schoharie Creek Bridge failure WJE Associates, Inc. with Mueser Rutledge Consulting Engineers also New York State Disaster Preparedness Commission contracted Thornton-Tomasetti, P.C. These teams worked together to find out what was wrong with the Schoharie Creek Bridge. All the teams conducted a report that was concluding the reason of the failure was scour under pier three. This scouring was affected by four important factors according to the Thornton-Tomasetti, P.C. Since the footings were shallow they claimed that it wasn’t enough to have bearing on soil. Also the foundation of pier 3 was laying on the erodible soil which means high velocity floodwaters can penetrate the bearing easily. The as-built footing excavations and backfill could not resist scour. The area left around the footing due to excavation was backfilled with erodible soil and topped off with dry riprap, in turn rather than being backfilled with riprap stone to the entire depth of the excavation as design plans specified [3]. In addition to these they claimed the riprap protection wasn’t adequate.
Lessons Learned
Underwater bridge inspections are very important for the safety of the public because of the reasons:
– Providing guarantee that the bridge is safe to utilize.
– Identify and take precaution to the potential and existing problems of the bridge early.
– Transcribing the current state of the bridge in periodic timeframes.
– Providing valuable information to designers, contractors, and government about the problems that could occur and likely maintenance problems.
– Checking the effects of the allowable loads on the bridge.
References
[1] International Conference on Scour and Erosion 2010 : San Francisco, C., Burns, S. E., American Society of Civil Engineers. Geo-Institute., & Environmental and Water Resources Institute (U.S.). (2011). Scour and erosion: Proceedings of the fifth International Conference on Scour and Erosion, ICSE-5, November 7-10, 2010, San Francisco, California. Reston, Va.: American Society of Civil Engineers.
[2] United States. Congress. Senate. Committee on Environment and Public Works. Subcommittee on Water Resources, T. (1987). Collapse of the New York State Thruway bridge over the Schoharie Creek: Hearing before the Subcommittee on Water Resources, Transportation, and Infrastructure of the Committee on Environment and Public Works, United States Senate, One hundredth Congress, first session, May 4, 1987. Washington: U.S. G.P.O..
[3] Storey, C., & Delatte, N. (2003). Lessons from the Collapse of the Schoharie Creek Bridge. Forensic Engineering (2003). doi:10.1061/40692(241)18
[4] Haines, D. B., & Zabilansky, L. J. (2006). Scour under Ice: Potential Contributing Factor in the Schoharie Creek Bridge Collapse. Current Practices in Cold Regions Engineering. doi:10.1061/40836(210)17
[5] Lebeau, K., & Wadia-Fascetti, S. (2007). Comparative Probabilistic Initial Bridge Load Rating Model. Journal of Bridge Engineering, 12(6), 785-793. doi:10.1061/(asce)1084-0702(2007)12:6(785)