On Shaky Ground
Altitude is an astronomer's delight. At elevation there's less atmosphere to distort starlight and absorb infrared energy. Mountaintops tend to have unobstructed 360-degree views of the sky and aren't compromised by light pollution. Some of the best, in astronomical terms, are dry and barren with few cloudy nights.
Building at height seems like an obvious choice now, but before the Lick Observatory was constructed at 4,209 feet atop Mt. Hamilton in San Jose, California, in 1888, cities were the preferred placement for permanently occupied installations. (Lick, an eccentric businessman seeking immortality, wanted to build a telescope "larger and more powerful than any existing" in downtown San Francisco until George Davidson, president of the California Academy of Sciences and a trusted adviser, convinced him of the advantages of altitude.)
Today, the Chilean Andes and Mauna Kea, a dormant volcano in Hawaii, are two of the best sites in the world for ground-based astronomy. And there's a building boom at both: a new generation of extremely large, extremely complicated, and extremely expensive visible- and near-infrared-light observatories, uncreatively referred to as "extremely large telescopes," are in planning or under construction. The Extremely Large Telescope (ELT) and the Giant Magellan Telescope (GMT), both in Chile; and the Thirty Meter Telescope (TMT) on Mauna Kea are the three largest.
Excellent though these sites may be, Chile and Hawaii share a downside: earthquakes. The strongest quake in modern history, a 9.5 on the Richter Scale, hit southern Chile in 1960, killing more than 1,600 people and creating a tsunami that raced across the Pacific and swamped coastlines in New Zealand, Japan, and the Philippines. Hawaii averages ten earthquakes a year of magnitude four or greater due to volcanic activity, plate tectonics, or the flexing and bending of the earth's crust. There's no free lunch, astronomically speaking.
Protecting these new mega-observatories from the capricious sites they'll inhabit is a challenge made more complicated by the nature of the telescopes themselves. To mitigate the risk, designers are employing techniques that have been proven in civil engineering but are untested in astronomy.
Cross section of the GMT enclosure section, with seismic isolation system below the concrete wall. Credit: Eric A. Manuel/M3 Engineering & Technology Corp. SPIE Proceedings doi: 10.1117/12.2314654
Earthquakes have always been a consideration for observatories in seismically active areas. Observatories are, first and foremost, buildings that have to remain standing after an earthquake, and no astronomer wants a telescope down for repairs.
But "legacy" observatories—famous names such as Keck, Gemini, Subaru, the Very Large Telescope—are of another era, said Frank Kan, a seismic analyst and principal investigator with the engineering firm Simpson Gumpertz & Heger. "Being smaller meant they were lighter and cheaper," said Kan. "The design philosophy was to make them as strong and stiff as possible while incorporating safety measures, designing them with sturdy concrete piers anchored into solid rock."
Primarily, they adhered to building codes which typically allow a degree of inelasticity in design as a means of absorbing energy. "So a building may not return to its original shape, it may be highly deformed after an earthquake," said Kan.
Yes, a telescope is a building. But it's also much more. "This is a gray area," said Kan. "Is it a building or is it a machine?"
It's a distinction with a difference, apparent now that legacy telescopes have been through a few earthquakes. On 15 October 2006, a magnitude 6.7 quake and a 6.0 aftershock a few minutes later, struck Hawaii, knocking out power in Oahu and causing $120 million in infrastructure damage on the Big Island, where Mauna Kea is located. It was the largest earthquake since telescopes were built on the dormant volcano's summit.
With the exception of minor damage to office facilities—books knocked off shelves, fallen ceiling tiles, damaged desktop computers, etc.—most of the 13 observatories on the summit were unscathed and quickly back on science. Two of the larger telescopes weren't quite as lucky. Gemini North, which has a twin in Chile that rode out a large earthquake in 2015, was down for 26 days due to a failure in the secondary mirror system. And the W. M. Keck Observatory's two telescopes lost about five weeks of science due to damage of azimuth encoders, which helps determine their exact orientation, bent seismic restraints that kept the telescope from jumping off its track, and damage to other equipment.
Kan studied the damage to Keck and later recommended seismic retrofitting to prevent future earthquake damage. "We used site-specific seismic and applied it to our model and found out that if we don't do any upgrades the telescope is vulnerable to damage."
Diagram of the European Extremely Large Telescope. The vertical gold supports at the bottom are the seismic isolators. Credit: ESO
The new generation of extremely large telescope observatories differ from one another significantly in design, but share a common feature legacy telescopes lack: isolation between the telescope itself and the pier anchoring it to the earth. The size and weight of these instruments necessitates it.
"We believed we could resolve our seismic challenges without an extreme measure of seismic isolation," said GMT project designer Dave Ashby. "It's actually a pretty compact design. However, once we actually started to explore the risk exposure in a quantitative way, we rapidly came to the conclusion that that wasn't practical."
In this case compact doesn't mean small. The GMT, under construction at Las Campanas Observatory in Chile's Atacama Desert, is 48 meters from the base of the pier to the top of the secondary mirror support. Each of the seven 8.4-meter primary mirrors weigh 17 metric tons. The telescope itself is enclosed in a 22-story dome.
It's similar in design to the Large Binocular Telescope in southeastern Arizona, which Ashby also worked on. But the LBT isn't in a seismic hot spot, so it's not designed to withstand a big earthquake.
At a cost of more than $1 billion and 16 years to build, GMT designers have no such luxury. To better understand their location, they modeled earthquakes on a continuum, from events small enough to go unnoticed to a "survival-level earthquake" during which the goal is avoiding collapse. They wanted to understand the probability density of peak ground acceleration levels possible in a broad range of quakes to quantify how the telescope would respond. "All of this is a probabilistic treatment to understand what kind of risk exposure we actually have," said Ashby.
GMT designers settled on friction pendulum bearings, which consist of top and bottom plates separated by a slider, to isolate their telescope from the ground. Both the slider and the bottom plate it moves on are concave, inducing a pendulum-like movement when the ground moves. Common in civil engineering, friction pendulum bearings range from about 3-feet in diameter to the 13-foot-diameter bearings, the world's largest, that seismically isolate the Benicia-Martinez bridge near San Francisco.
"They're very simple devices that provide many different functions," said Ashby. "They provide the actual mechanical isolation like a bearing, they provide return force like a spring, and they also dissipate the energy of the earthquake. To provide those three functions in a simple device is, I think, a relatively impressive feat."
So equipped, the GMT should withstand a survival-level event, which on Las Campanas has a return time of 2,475 years, meaning a quake strong enough to put the telescope's structure to the ultimate test will happen at that time interval on average.
The TMT, proposed on Mauna Kea but currently on hold due to protests at the building site, incorporates 492 actively controlled mirror segments into a 30-meter, 121-metric-ton goliath.
To seismically isolate the TMT, engineers incorporated four viscous fluid dampers arrayed in a horizontal plane where the telescope's azimuth ring—the track that enables horizontal rotation—meets the pier. These are similar to shock absorbers on a car on a grand scale. The idea is fairly simple: Fluid, typically silicone oil, is held in a pressurized chamber divided internally by an orifice. One end of the damper is connected to a piston that forces the fluid through the orifice when movement occurs, converting dynamic energy into heat, which is dissipated by the damper itself.
Masao Saito, manager of the TMT's telescope structure team, said the system is designed to survive a 1,000-year-return level quake. "There's a very stringent requirement of the telescope structure about a seismic event," said Saito. "The lifetime of the TMT is 50 years, so in a 1,000-year return there is a two percent chance that we may be hit with such a gigantic earthquake."
And then there's the ELT, "the world's biggest eye on the sky," under construction on Cerro Armazones in the Atacama Desert. Its 39-meter mirror, close to half the length of a soccer field, will be made up of 798 smaller mirrors that can actively adapt to conditions 1,000 times per second. It will gather more light than all 8-10 meter legacy telescopes on the planet combined.
Like the other two mega-projects, the ELT incorporates seismic isolators into its foundation. These allow up to 30 centimeters of horizontal movement. Unlike the others, ELT designers are also planning to temper the vertical acceleration associated with earthquakes with 80 "shock absorbers" built into the hydrostatic pads of the telescope's azimuth ring.
Vertical damping is more difficult to achieve, and no off-the-shelf solution exists for a machine this size. But ELT engineers had a bit of relevant experience to pull from. The Atacama Large Millimeter/submillimeter Array, or ALMA, designed and built by the ELT's parent organization, the European Southern Observatory (ESO), is a series of mobile, 100-metric-ton antennas moved on 14-wheeled transporters. Doing so on unpaved desert roads subjects the delicate antennas to potentially damaging road shocks, so each wheel is vertically damped.
ESO engineer Max Kraus explained the system, and how it might be adaptable to the ELT, in a blog post on the observatory's website. "The goal of the system is to limit the forces transmitted from the road to the antenna by changing the oil volume in the hydraulic cylinders that connect the wheels and transporter frame—essentially, the system absorbs the shock before it reaches the antenna. If configured correctly, it doesn't need to be linked to an active sensing system. We realized that if we could also develop the ELT seismic damping technology without using an earthquake detection system located at a large distance from the ELT site, it would simplify the process and avoid expenses and risks."
The challenge of building behemoth observatories in astronomically exquisite places that could destroy them is fundamentally about mitigating risk. Telescopes are both massive and delicate, powerful and finely tuned. Each is thousands of tons of weight moving precisely in two planes, aligned exactly with the stars. To operate correctly, an observatory's foundation must be absolutely stiff, resisting wind loads and nuisance tremors, until the exact moment it needs to be flexible. Then when the shaking is over, all that slewing mass has to return, or be returnable, to its exact position in order to get back to doing science. Bent or broken machinery would be a setback. Damaged mirrors could take years to replace, and would be a catastrophe.
"There is no industry data to provide stiffness of these bearings in these conditions," said the GMT's Ashby. "We just actually completed tests to directly measure the stiffness of these bearings in these conditions. We selected a set of prototype bearings and we did a series of rigorous tasks that include nominal seismic isolation performance to explore how they would impact normal operation. We've convinced ourselves at this point that they pose no increased risk in that respect."
It's uncharted territory, but so is the Universe these telescopes will explore—as long as the engineers can protect the machine. Ashby summed up what's at stake: "It's not much of a telescope if you break your mirrors."
Bob Whitby is a science writer based in Fayetteville, Arkansas.
|Enjoy this article?
Get similar news in your inbox