The decision to be a professional brings both benefits and dues, with one of the latter being at least occasional attendance at an annual awards dinner. Thus it was that a few years ago Gary Brierley, a consulting civil engineer, found himself listening to a speech by "Mr. Civil Engineer of 20xx." The happy winner stood up, thanked everyone, and unrolled the story of his project, which had to do with designing a high-security facility for the government. First the speaker listed the specs: The facility had to be able to withstand a direct impact from a big tornado, or to shrug off a crashing airplane. The usual. Then the engineer began describing his solutions: concrete walls four feet thick; layers of reinforcing steel and so on.
Brierley by his own account began to feel a touch of fatigue. Brierley is a specialist in underground construction, and he could see that the civil engineer was boasting about spending tens of millions of client dollars just to replicate on the surface everything that was already right there, literally under his feet.
This is an old and somewhat mystifying story for Brierley and other proponents of underground construction. In theory underground space hits almost every line on the CSO's wish list. It can be made nearly invisible from the surface, provides very tight access (and egress) control, and gives almost total insulation from surface turbulence, including fires, weather, riots and ordnance. Maintenance requirements are lower, which means fewer maintenance workers to pay. External support and insulation come free with the address, which means fewer worries about cracks and leaks. The ambient temperature is stable as a rock, pun intended.
Perhaps counterintuitively, underground space provides much better protection from earthquakes. Since a surface structure stands independently, with no external bracing, a wave passing through its foundation whips the structure back and forth, like yanking a rug from underfoot. And it gets worse: When earthquake waves hit the surface they interact with fields of objects that are complicated and heterogeneous, both geological and structural. These objects shatter the quake waves, bouncing them everywhere, twisting buildings back and forth in all directions. These are highly stressful experiences for any building, no matter how stoutly built. By contrast, underground structures live in a much quieter environment. A good analogy is the difference between experiencing a storm at sea under the water versus riding it out in a boat (on a cluttered surface). In 1999 professor Scott Kieffer of the Colorado School of Mines wrote a report comparing the damage wrought above and below ground by a major earthquake in Taiwan. There was no comparison. "The underground structures behaved so well," he says. "Worst case? Maybe it might leak a bit."
Security and Safety First: Benefits of Building Underground
Finally, underground siting responds directly to community concerns about security and safety: Nobody is going to be on your back about a facility that is 200 feet underground. There will be no fights about historically protected buildings or sun rights. Put a structure under the surface and the neighbors will forget about it, surely the happiest possible outcome from a security point of view. "If you have a security problem, you need a reason not to go underground," is how Brierley sees it.
The military has understood this for some time, as shown by the construction of the various underground cities intended to protect the entitled from nuclear war. (The first of these, Raven Rock, was built in the 1950s. Among other amenities it had cars, roads, designated smoking areas and a chapel.) Yet despite their example, the underground option is surprisingly absent from public discussion. How is it that the anxiety over the security and safety issues raised by nuclear power plants so seldom expresses itself as a demand that they be built underground? That the discussion over keeping high-speed trains safe from vandals and worse does not end with a consensus to put trains in tunnels (where they would not only be safe but could run a great deal faster)? Why do the authorities not insist that every phase of liquid natural gas management, right from the moment it leaves the tanker, take place under the earth's surface?
The reflexive answer would be cost, and while there is something to that, people are often willing to pay for public safety. Besides, the price per cubic foot of underground space is not all that far from the range of contemporary construction costs. In the mid-'90s an 11.5 kilometer road tunnel was built in Norway (between Flam and Gudvangen) for about $17 a real estate foot, or REF (a REF equals 10 cubic feet). That number is on the low side; it reflects the efficiency that can be built up by the same team doing the same thing under the same conditions for a long time. Not all underground projects have those advantages. More representatively, a metro extension dug in Madrid, Spain, during the late '90s came in at $150/REF (though note that figure is a total cost, including tracks, rolling stock and signaling). On the high end, metropolitan Boston just paid nearly $250/REF to push a water tunnel through exceptionally difficult geology. It's not instantly obvious that these prices should keep underground space out of the conversation, but so long as people think cost is a problem, it is.
Another problem with subterranean space is that the skills of making and marketing underground structures have little overlap with those required by surface buildings of the same volume. The structural analysis of a building that is surrounded and therefore supported by rock is totally different from that of a freestanding edifice. In any structure for which management of heat gradients is important (reactors, refineries), allowances will have to be made for the fact that rock is an insulator while air is not, and people will have to be shown why they should have confidence in those allowances. The same point can be made across a range of issues, from building integrity to fire prevention. All these problems have to be rethought and the pool of those competent to do so is small.
Both of these problems, high costs and small expertise pools, are affected by total market size; the larger the market, the more R&D capital and skilled labor flow into the sector. Fortunately, the list of applications for underground space has been expanding steadily during this century, beginning with utilities and transport to sidestepping the zero-sum games that surface structures must fight with the forces of preservation and conservation to city parking.
Tunnel Boring Machines
One consequence of this growth has been steady technical progress. The first of the two families of technologies employed in this work, drill and blast, was invented in the 14th century when Marco Polo brought black gunpowder back from China. The core procedure, which stayed basically unchanged for 600 years, was to drill holes in the working face, pack them full of explosives, run around the corner, touch off the charge, run back to the face (dealing with the fumes released by the explosion as best you could), try to get the roof supports shoved into place before the ceiling collapsed and then shovel out the fragments. It was a terrible job in every respect, not least because the system was impossible to scale. Only two or three people had room to work on a tunnel face at any one time, and conditions in chambers or caverns were not much better. Production was measured in a few linear feet a day.
In the 1950s an engineer named James Robbins hit on the idea of building a machine that would push cutter wheels or discs into the rock face and then rotate them in circles, like an immense glass cutter. Done with enough force, each of these rotations would shatter an inch or so of rock off the face. Improvements came slowly, but by the '90s tunnel boring machines, or TBMs, were handling the three key phases of the cycle: breaking the rock, picking up the pieces and dropping them on a disposal conveyor, and fabricating and installing a lining. A contemporary machine can excavate and line a tube with a diameter in the dozens of feet at an average rate of a hundred feet a day. (Actual rates vary widely.) The very largest TBMs have diameters approaching 50 feet, which is about what you need for a stack of two four-lane highways, one going in each direction.
TBMs will probably not get much larger, but they will certainly get faster. Rock sensingusing acoustics to give TBMs "headlights" so they can see and prepare for changes in ground conditionis becoming standard, as is remote control and unattended operation. Recent discoveries in materials science will allow much tougher cutter wheels, which means faster speeds through harder rock. New industry IT standards will allow machines to consult design and seismic databases on the fly. Under way are experiments such as equipping TBMs with lasers, since warmer rock shatters more efficiently. Today about 150 TBMs are working around the world at any one time, and that number may double over the next 10 years.
Meantime, drill and blast, now used mostly to hollow out caverns as opposed to tunneling, has picked up the pace impressively. Contemporary drills come stuffed with sensors that monitor their own vibrations and thermal behaviors. Since different kinds of rock make drills behave in different ways, these sensors allow the drill to tune itself to the material at hand, including controlling the explosives mix. This intelligence means the rigs can drive faster without damaging the ambient rock, historically the big downside of drill and blast. A modern drilling rig can create space almost as fast as a tunnel boring machine, with the advantage of requiring a lot less capital up front. The two technologies are competing to see which can lower costs faster.
Eventually the cultural barriers that so annoy Brierley will dissolve and underground space will pick up another driver: security. That day might arrive soon. An increasing number of disaster recovery services are marketing underground storage of both documents and backup servers as a feature. According to Mike Smith, proprietor of industry portal TunnelBuilder.com, weather security is rising in importance all around the world as a justification for underground highways.
For decades Boston was content to drink from a surface aqueduct (the Hultman Aqueduct). Anyone with the dollars to rent a backhoe for a few hours could have cut the water supply of the entire city, perhaps for several days (puncturing the system would probably have caused lots of ancillary damage). Recently the national culture changed, and this vulnerability became intolerable. As a result, the money was found to drive a tunnel through granite and quartz ground, which are about as unfavorable as ground gets. Smith thinks that eventually security will join rising real estate prices as one of the basic motives for going underground. At the very least this new application will draw in still more investment, driving costs down still further. In decades to come this trend might look less like a convenience and more like a migration.