The Quail Creek Dike failed because its lowest layer was not sealed off from water that seeped through the underlying rock and earth, a blue-ribbon task force has concluded.
Material used to build the dike was placed on the ground, including soil and rock, and "not protected from seepage erosion," said a letter to Gov. Norm Bangerter from the team of independent specialists appointed to examine the dike's failure.The team says the dam could be built again safely.
Asked by the Deseret News what he would have done differently in building the dam, taking advantage of hindsight, team leader Robert L. James said, "This contact (the point between the fill material and the ground) has to be protected."
The team's briefing document to Bangerter also said:
-Geologic conditions at the site were extremely challenging "and deserved special consideration in design."
-Three major sets of nearly vertical fractures were present in the foundation. They "permitted significant seepage flow" beneath the dike.
The report charged that "foundation exploration was not designed or complete enough to fully detect seepage problems associated with these joints."
-"The early assumption that there would be little or no seepage through the dike foundation below the shallow cutoff (a trench lined with clay, designed to prevent water from flowing below the dike) was not valid and had a profound effect on design of seepage erosion protection."
-Highly fractured, pervious rock and erodible overburden - that is, soil and rocks - were left in place below the dike on both sides of the cutoff trench. That permitted seepage along the contact line between the ground and the dike's fill material.
-The gypsum, a soluble salt, in the foundation wasn't a primary cause of failure. But when it dissolved as water percolated through the foundation, it hastened the erosion of the dike.
-Remedial grouting, the filling of hundreds of holes with concrete as was done when the dike leaked earlier, was not a long-term solution.
-"There is no indication that seepage through the dike embankment or the quality of its construction contributed to the failure."
In other words, the dike didn't fail because water worked its way through some permeable section of the earth-fill structure high above the reservoir's floor. It failed because water seeped through cracks under the dike and ate away at the dike itself.
In a press conference Tuesday afternoon in the offices of the Utah Department of Natural Resources, James said dam builders stripped off about a foot of soil, rock and other material. This was several feet thick in some places. He believes they may have excavated down to the rock in places, but not in other spots.
Also, he would have preferred that in addition to removing the material, called overburden, from the site where the dike would be built, that the top layers of eroded rock be taken out.
Then, if the planners had known about the vertical fractures, they could have put down a layer of cement, or at least gravel, as the base of the dike.
That material could have acted as a seal or a "filter" to prevent the eroding of the fill material when the seepage started. A filter layer could have allowed water to pass through, but not carry out the fill material, thus protecting the dike.
The clay liner was only at the bottom of the narrow cutoff trench under the dike, not under the entire dike.
James said the team prefers to call the dike a dam, as it was 75 feet high and "a significant structure." When it broke through Jan. 1, a breach 300 feet wide opened.
When the seepage became rapid on the day before the dike burst, it began "piping," he said. This involved the excavation of a cavity where the water seeped out of the toe of the dike.
The cavity then enlarged back into the dike as more and more fill material was carried out.
By 10:30 p.m. on Dec. 31, 1988, the pipe broke through to the reservoir. The flow increased to 70 cubic feet per second.
A 50-foot slab of the dike collapsed on the downstream face, temporarily blocking the cavity and nearly stopping the flow. "But unfortunately you never stop the seepage water on the downstream side," James said.
So "slice by slice," more chunks of the dike tumbled down until a 300-foot breech opened, shortly after midnight Jan. 1. Water roared out at 100,000 cubic feet per second, causing $12 million in damage.
Asked whether mistakes were made in the dike's design, James replied, "You may think I'm being evasive and perhaps I am." But, he added, "it would certainly have been nice to recognize seepage was going to go through this foundation."
Four experts addressed themselves to the dike's failure - James, a consultant from Lancaster, Texas; Richard B. Catanach, an engineer from Santa Fe, N.M.; J. Lawrence Von Thun, head of an analysis group for the U.S. Bureau of Reclamation in Denver; and Alan L. O'Neill, a consulting engineering geologist from San Francisco. Together they have 113 years of professional experience in the field.
The fifth reviewer, Bruce C. Barrett, was not involved in that phase because of a potential conflict of interest. He was once employed by the Provo firm Rollins, Brown and Gunnell Inc. Ralph Rollins, a principal in that firm, has been identified as the dike's designer.
However, Barrett participated in the study of the team's second major question, whether the dike could be rebuilt.
To that question, James responded, "The opinion of this panel is, yes, a safe dam can be built at Quail Creek Dike." The group advised against patching up the present dike, recommending an entirely new structure be built with more precautions.
"I can build a dam here, there or anywhere - you might not like the price tag," James said. He added that there would be seepage, because there almost always is with dams.
Dike study conclusions
- Geologic conditions at the site were challenging "and deserved special consideration in design."
-Three major sets of nearly vertical fractures were present in the foundation that permitted significant seepage beneath the dike.
- Foundation exploration was not designed or complete enough to fully detect seepage.
- The early assumption that there would be little or no seepage through the dike foundation below the shallow cutoff was not valid and had profound effect on design of seepage erosion protection.
- Gypsum, a soluble salt in the foundation, wasn't a primary cause of failure. But when it dissolved, it hastened the dike's erosion.
- Remedial grouting, the filling of hundreds of holes with concrete as was done when the dike leaked earlier, was not a long-term solution.
- No evidence was found that seepage through the dike embankment or the quality of its construction contributed to the failure.