Jackson Hole news | Jackson Hole feature | Jackson Hole environmental | Jackson Hole sports | Today's News

Science to settle dam debate
Safety of Jackson Lake Dam is being reassessed as new technology comes to fore.

By Angus M. Thuermer Jr.

New scientific methods that have been refined in the last decade are being used to recalculate the safety of Jackson Lake Dam, which was reconstructed in the 1980s.

After the dam was rebuilt at a cost of $78 million, a four-year project that ended in 1989, experts with the Bureau of Reclamation outlined their challenges and achievements in a memo. They listed forces that the Teton Fault, 7.5 miles to the west, might generate at the dam.

"Mean peak horizontal bedrock accelerations at the damsite were estimated to range from 0.07 to 0.58 g [gravity]," Bureau experts wrote, implying a manageable temblor. The forces are expressed as multiples of gravity, the same measurement fighter pilots use to quantify pressure experienced in high-speed turns.

In 2000, Bureau consultants URS Greiner Woodward-Clyde, a California company, determined that peak ground accelerations at the dam might reach 1.19 g, more than twice the original Bureau figure. Taking other factors into account, the consultants said forces impacting the dam could reach 1.42 g.

Scientists today have a clearer picture of how quake waves act, how soils like those in the valley respond to shaking, and how basins like Jackson Hole focus and amplify earthquake forces, Bureau experts and consultants are heading back to their computers to reexamine whether the dam is as safe as they once thought.

All this caught the attention of Wyoming Gov. Jim Geringer, who listed them as grounds for worry in a letter to Secretary of the Interior Gale Norton, who oversees the Bureau. "A considerable difference may exist between estimates of design ground accelerations used in the late 1980s and today's models," Geringer wrote in September 2001. Given those differences, plus "residual uncertainty in the engineering analysis," Geringer asked Norton to make the Bureau prove that the dam is safe.

Fourteen years after spending $78 million to reconstruct the mile-long dam, the Bureau of Reclamation is undertaking a reassessment of the hazards associated with the dam. Bureau officials say the study, to be released in March, will show the dam won't fail.

The dam rises 49 feet above the outlet to Jackson Lake and its rupture would cause the uncontrolled release of 300,000 cubic feet per second, many times larger than the Snake River in western Wyoming has seen in thousands of years. The effects would be devastating, ruining environmentally sensitive portions of Grand Teton National Park, flooding posh subdivisions and Wilson, disrupting a multi-million-dollar recreational fishing and scenic-float industry and crippling irrigation districts in Idaho.

Would the Teton Fault, the geologic structure that displaced bedrock more than 20,000 feet to create the 13,770-foot-high Grand Teton and its surround range, ever shake the groung violently enough to damage the dam? At the Bureau's Safety of Dams office in Denver, Chief Bruce Muller would not rule out damage, but said residents and visitors need not worry.

"I guess damage is relative," he said in a recent telephone interview. "We don't believe a large earthquake would do any detrimental damage. There would be no uncontrolled release of the reservoir."

Muller said the new studies show "there may be localized damage" on the structure that stretches across mud flats at the east end of Jackson Lake. Some repair might be needed, he said.

"The dam would be expected to respond favorably to these events," Muller said. "That's what the engineers have told me."

New science raises questions

What is this new science that has people worried and that the Bureau used when it launched a review? How was the dam rebuilt and does that reconstruction guarantee the safety of residents? And why, when new questions are being raised about dam safety, has the Bureau chosen to shut off a network of 20 seismic sensors it placed along the Teton Fault to expose the mysteries of that feature?

Even before it chose to rebuild the dam, the Bureau, in the early 1980s, conducted a seismotectonic study on the Teton Fault and the forces it was expected to produce. The authors made several recommendations, the first of which was to "install additional seismometers in the vicinity of Jackson Lake Dam to collect high-quality regional seismic data for the Jackson Hole area on a long-term basis."

In the fall of 2001, after 16 years of operation and recording little more than a single quake on the Teton Fault, the Bureau turned the network off. The agency deals with 100-year probabilities, faults that might move once every 1,300 years, mountain ranges that take millions of years to form. So its interpretation of what the study means by "long-term basis," might be surprising.

"The technical staff ... they proposed the long-term network being a five-year installation," Muller said. "We turned it off after we got 16 years of data. We felt we went well beyond the original intent of a long-term network."

Referring to the 1980s recommendation for research, Muller said. "I guess our interpretation is it's been completed."

Construction history

To understand more about the fault, the network and the dam, an account of the reconstruction project is helpful. When dam construction was analyzed by the Bureau of Reclamation, the agency estimated that the maximum credible earthquake on the Teton Fault would be 7.5 magnitude on the open-ended quake scale. Such shaking would cause unconsolidated soils underneath the dam to act like a liquid and collapse, causing the dam to settle and fail from the pressure of the water behind it. Under some portions of the dam, unconsolidated soils extend to a depth of more than 600 feet.

The agency calculated that the aging structure first built in the early 1900s, had a 40 percent chance of failure in 100 years ­ "two orders of magnitude greater than the failure rate for dams in general," the Bureau estimated.

So began the reconstruction on both major components of the dam ­ the concrete structure at the south end and its associated earth dike that extends 0.8 miles north. To buttress the concrete segment, where spillways are located, workers poured more than 1,000 cubic yards of concrete.

The earthen dike also needed more work. First, it had to be completely removed. Then workers began solidifying the foundation. In one section, they hoisted a 32-ton weight 100 feet in the air by crane and dropped it, driving water out of the ground and solidifying the soils. Despite bent cranes and other glitches, this "dynamic compaction" process was repeated more than 36,000 times as the mammoth weight beat holes nine-feet deep, 16-feet wide. They were filled with sand and gravel and pounded some more.

Elsewhere, workers used augers three feet in diameter to drill holes up to 110 feet into the mud. They pumped cement grout down the voids. On the upstream foundation, these columns were linked together to form an underground wall at least 24 inches thick. Under the eastern edge of the dam, the columns formed a honeycomb pattern. Engineers hoped to contain the soupy soil during an earthquake and prevent liquefaction ­ the process where the foundation might ooze away and cause the dam to collapse from the weight of the lake behind it.

Was the program effective? Some of the techniques ultimately employed had never been used in the U.S. The Bureau launched the project before it even knew what method contractors would choose for solidifying parts of the foundation. All the while, both money and time were key factors in Bureau decision-making.

According to its reports, the Bureau chose to undertake some of the foundation work according to a "fixed-price negotiated contract." The method was employed to solicit "innovative ideas" from contractors. Safety wasn't the only factor. The contractor and method for one phase of work were chosen to ensure the dam would be rebuilt "at a reasonable cost within a two-year period."

The Bureau stated that "Portions of the work were relatively complex or could have been constructed in a variety of different configurations." And it chose one contractor "on the basis of technical merit and cost." The project was finished $2 million under budget.

All of this was enough to worry Geringer, who said as much in his letter to Secretary Norton in 2001.

"Jackson Lake Dam was reconstructed using a stabilization technique called dynamic compaction," he wrote. "While this technique is often used for stabilizing roads and other structures, experience in the United States with dams that have been subjected to large earthquakes is limited."

Structures have frequencies

Today's earthquake and engineering science deals with more than just the forces a quake will produce on bedrock. State Geological Survey hazards geologist Jim Case said scientists have drawn a new series of quake maps that form the basis for design required by the International Building Code. Instead of simple bedrock motion, scientists today must account for soil types, the wave period of an earthquake shock, plus the height of structures, among other factors, he said in a published paper.

Those forces can build to levels much larger than those an earthquake exerts on the bedrock. Scientists have published "probabilistic spectral acceleration maps," that combine many of these variables to predict the probability of earthquake-generated g-forces at various sites.

"Spectral acceleration is the acceleration [or force] experienced by a building when it is subject to an earthquake," Case said in an interview. "Ground acceleration is what you feel on the ground. It is generally what rock is subjected to."

Today, engineers know that structures have what is called a natural frequency or period. Like a tuning fork, they will sway or shake at a certain frequency once they are pushed. "If you could tap a building, it would vibrate with a natural frequency similar to what a tuning fork does," Case said.

Tall tuning forks, and tall buildings, have "harmonic vibrations" that are slower than short ones. These periods may range from 0.2 seconds for a structure only a few stories high to a full second for buildings that are seven stories or so. The material and type of construction also can be factors.

What is dangerous is when the frequency of a quake is the same as the natural frequency of a structure. "If a building has a natural frequency that is similar to the frequency of the seismic waves, you can set up harmonic vibrations in the building that would be greater than the acceleration [or quake force] on the ground," Case said.

This became apparent in quake-ravaged cities where buildings of a certain height toppled, while those taller or shorter remained standing. Thus, scientists today must go beyond simple predictions of maximum credible earthquakes or even ground acceleration to make the best predictions of the forces that come to play.

Teton Fault ripe for action

In the URS Greiner Woodward-Clyde report of 2000, authors Ivan Wong and others list spectral acceleration at Jackson Lake Dam for a quake with a one-second cycle at between 0.91 and 1.42 times the force of gravity. Again, these figures exceed the Bureau's calculation of bedrock forces in its post-construction report, where it said peak ground acceleration would only be 0.58 g.

"Back in the '80s, I don't think there was much discussion about spectral acceleration," Case said. "It was mostly ground acceleration." State officials have wondered about the differences between Bureau figures and those available today.

"Those accelerations exceed the design accelerations that were used back in the '80s to reconstruct Jackson Lake Dam," Case said. "That was part of our concern.

"That doesn't mean that the dam is going to fail," Case said. "It does mean the accelerations are greater than expected."

Teton County deputy attorney Keith Gingery put the fears more bluntly in an exchange with the Bureau seeking information about dam safety. "The Jackson Lake Dam can collapse either by the actions of a large earthquake or by a tidal wave caused by the earthquake," he wrote on Dec. 24, 2002.

Safety calculations also include the element of time and the probability of a certain magnitude quake occurring. Case said the Teton Fault is ripe for action.

"There is some question when it was last activated," he said. "It may have been 7,500 years ago, may have been 3,600 years." In any case, in recent geologic times, the fault hasn't moved as much as it generally does. "We have to assume the Teton Fault is overdue for activation," Case said.

Muller said Wong's estimates of g forces at the dam are being analyzed by the Bureau. "The information has been generated to assess the issue [Wong] has raised," Muller said. "It is addressed in the analysis," which is due in March.

Wong said the Bureau is being careful. The question to be answered is: Given the source of the earthquake and some probability of that quake occurring, what would be the level of ground motion, he said. The Bureau's assumption of quake probability is "very conservative, more than any federal guideline," Wong said.

"There's no doubt the levels of ground-shaking in the Jackson area are some of the highest in the Intermountain West," Wong said. "Everybody knows that. The BOR knows that the USGS knows that.

"Obviously, the Bureau's concerned with the safety of all their dams, not just Jackson Lake. Throughout the western U.S. in recent years they've been doing studies. They're just starting to understand the Teton Fault. Their program is a good program."

Wong said the controversial seismic network is expensive to maintain. The Bureau has put the operating cost at an estimated $100,000 annually.

"They've been running it for several years," Wong said. "Given the expense, they think they have sufficient information. But I can also understand the state's position. "It always boils down to science versus economics."

Fault angle clue to stability

Case and other Bureau critics believe continued operation of the network could reveal a key personality of the Teton Fault ­ the angle at which the fault plane dives underground at the base of the range.

Does the angle make a difference? A big difference, some geologists say. If the fault tilted at a 45-degree slope to the east, It would be 7.5 miles deep under the dam. But if it was at an angle of 35 degrees, it would be much shallower and closer to the structure.

"If the fault dips at 35 degrees, peak ground acceleration [would be increased by] a factor of two stronger than if the fault is dipping more steeply," Case said. How steep it actually is, "we really don't know." But if the network were in place and there was some small activity on the fault, that angle might be computed.

"If we started to get some micro-seismic activity on the fault, we can delineate it," Case said. But only if the seismic network is operating.

Ralph Archuleta, a professor at the University of California Santa Barbara, recently co-authored a paper that calculates the effects of a quake on soils in Jackson Hole. It was scheduled to be delivered to a symposium last December until the Bureau blocked presentation, claiming the work had not undergone proper peer review.

In an interview, Archuleta said that most normal faults, like the Teton Fault, dip at steep angles on the order of 60 degrees. Also, shallow faults, those at 35 degrees, are more difficult to dislodge. He agreed, however, that the distance between structures and faults is important.

Muller said such factors are being considered. "The angle of the fault we fully have accounted for," he said. "We estimated the likelihood of different orientations and included that in the likelihood of certain types of ground motions at the site."

Valley may shake like Jell-O

The structure of Jackson Hole and the surrounding mountains give rise to even more factors affecting dam stability. The top 1,500 feet of the valley basin is filled with Quaternary debris - rubble-like sand, gravel rocks, silt, and so on, deposited by rivers and glaciers. The water table is two meters below the surface, making soils prime for liquefaction in the case of a quake.

Bureau scientists have recently reported that parts of Jackson Hole, including the area under Jackson Lake, have a tendency to "trap" seismic waves, resulting in extended shaking. According to the abstract of one paper scheduled to be delivered in December, but put on hold by the Bureau, quake waves would bounce off bedrock beneath the Tetons, extending the period of shaking in the middle of the valley by more than 10 seconds.

"If you were to hit Jackson Hole from the side, the bedrock would give you a little jolt," Case said. "The basin sediments ­ they're going to resonate, almost like a bowl of Jell-O. The bowl itself is just going to have a little jolt, but the Jello will keep on moving. The movement [of the Jell-O] is greater than the bowl itself."

Archuleta's paper, published only in abstract form so far, analyzes the liquefaction potential of soils in the valley. He said the work was novel.

"Nobody had ever done the calculation 'If you had a magnitude 7 on the Teton Fault, what is likely to be the ground-shaking at the surface of the dam?' That was the purpose of the abstract.

"There's a sand layer between 10 and 27 meters depth," Archuleta said. A quake would put a shear strain approaching 20 percent in the layer, his abstract said.

"Strains that large are almost impossible to imagine," Archuleta said. "What it indicates is nonlinear behavior, material changing its physical properties.

"This indicates there's a fundamental change in the sand layer, from a solid to a liquid, or acting in that way," he said. "This is what I think anybody would recognize ­ the onset of liquefaction. It [the sand layer] is behaving like a liquid. It's starting to flow. Under a large earthquake, what we expect to see is liquefaction."

Given uncertainties about the Teton Fault, Case and others in Wyoming believe the network should have kept operating. The state and the Bureau have traded criticism, the Bureau saying it offered others a chance to take over the system, the state saying it didn't have a proper opportunity.

"I disagree with the Bureau's decision to abandon the network and believe that a strong case can be made for federal operation of the Jackson Lake Dam Seismic Network," Geringer wrote to Norton. "We in Wyoming believe that the Bureau of Reclamation has a responsibility to maintain the network until the Teton Fault can be adequately characterized using data from events directly associated with the fault, or in close proximity to it. ... Rather than risk a problem, I much prefer that additional information be gathered from the seismic network, including the Teton Fault."

Muller said the network would not have helped map the angle of the fault as the state believes it would.

"I don't think it's the seismic network that determines that," Muller said. "You would have to have a number of ruptures of the Teton Fault to be able to determine source locations. The Teton Fault ruptures on an average of once every 1,300 years. So we don't anticipate that the data needed would be generated in any near-future monitoring."

Underground secrets

The network may reveal some underground secrets ­ basic earth science ­ the Bureau contends, but that is not necessary for the safe operation of the Jackson Lake Dam.

"A case can be made that basic earth science may be valuable," Muller said. "The disagreement is not so much about the value of the network. It is more a disagreement about Reclamation's role and responsibility."

The agency is charged to determine whether additional modifications are required. "We've collected enough data to make that determination," Muller said.

Last September, on the eve of the termination of the network, Dr. Bob Smith, professor of geophysics at the University Utah, a part-time valley resident and a supporter of the network, made an observation that he called a "strange reminder."

"It seems quite poetic that three days before the planned termination of the Bureau of Reclamation's Teton Seismic Network that we have a magnitude 4.7 earthquake in the Gros Ventres," he wrote. He was referring to the Gros Ventre Range that makes up the east edge of Jackson Hole opposite the Tetons.

"Without a seismic network in the Teton region, a key element of the 1300 km long Intermountain Seismic Belt containing one of the major faults of the western US and several Bur of Rec reservoirs ... will be without earthquake monitoring capabilities!" he wrote in an e-mail note. The decision, he said, was "A step back in the public safety efforts that we as a community have built up for so long."

Case said that in addition to helping define the fault, the network could have signaled new movement, could have warned residents that the long-dormant crack was again coming to life.

"If it's the peak of tourist season and there are small-scale earthquakes occurring on the Teton Fault, they may be nothing more than that," Case said. "They may be precursors to a larger event ­ they may be. If they are and we don't have that knowledge, we can't put up any red flags and that's criminal."

News reports questioned

Case said his characterization was a forceful one.

"That's pretty strong wording," he said. "But if there's something you can do and you don't do it, I don't know what other word to use."

Bureau Commissioner John Keys wrote a guest opinion for the News&Guide this week, stating that the Jackson Lake Dam is safe (see page 5A). Without naming specific stories or instances, he said some reports of dam problems are wrong.

"Regrettably, much of what has been published has been misleading and inaccurate," Keys wrote. "The microseismic network that is presently being debated in the local press has no role in the safe operation of Jackson Lake Dam. ... This equipment is by no means an early warning system as published reports imply.

"The safety of the residents of Jackson Hole is not put at risk by our decision to remove the microseismic network," Keys wrote. "To suggest otherwise is simply inaccurate and inflammatory."

"Jackson Lake Dam has the same high level of safety as the other 369 dams covered under the Bureau of Reclamation's Safety of Dams program," Keys wrote. "It is safe."

Respond to this article by e-mailing publisher@jhnewsandguide.com