Shifting Sands of Tunnel Techniques

The first technological venture of my life failed. I was 4 years old, my brother was 6; starting at sites about three feet apart on the beach, we tried to dig a tunnel. What happened was two tunnels. Our efforts paralleled and we never met.

The problems of tunneling are familiar to any beach engineer who has ever dug a tunnel in the sand. The area has to be surveyed to decide the best path, and the tunnel’s roof and sides need temporary support so they don’t collapse while you’re digging.

My brother and I had the right instincts. We used water to make a kind of mud for temporary support, and installed a “permanent” lining of driftwood and shells. We were right to begin digging from both ends too. We just didn’t know much about surveying.

From Temple to Spring

From the beginning of history, tunnelers have found it economical to start digging in two teams. Before modern surveying methods were developed, lots of people had trouble meeting in the middle. Trial and error ruled, with error often triumphant. In the 7th Century BC, for instance, King Hezekiah of Judea, anticipating a siege that did in fact happen, commissioned a tunnel connecting the Temple Mount inside the walls of Jerusalem with the Siloan spring outside, so that his people would have water. It worked, but it was a close call.

Archeologists excavating the site not long ago unearthed the biblical tunnel. It isn’t straight. They also found a message scratched in ancient Hebrew on the wall that explains why. It seems that the workers heard each other’s voices just in time, changed direction and so met in an “S” near the inscription.

Roman Engineer Saves the Day

More than 800 years later, a semi-retired Roman engineer, Monius Datus, arrived at a hilltop near Algiers to find a crew of workers literally weeping beside an excavation. They had been digging in a circular path and had dug half again as far as the distance through the hill. Each team had gone off line and they couldn’t seem to meet. Monius set them tunneling straight, saved the aqueduct and installed a stone marker in his own honor recounting the feat.

By the early 19th Century, surveying allowed tunnelers to dig straight, so builders were confident that if teams started digging from both sides of a river, they would eventually meet.

Today, as in the past, time is the most expensive part of tunneling. The slower a tunnel progresses, the more it costs to keep the equipment going and the longer it takes to begin earning a profit from the enterprise. When in April “groundhogs” went the last mile in the longest tunnel ever built--the 32-mile Seikon Tunnel connecting the Japanese islands of Honshu and Hokkaido--it was 10 years behind schedule and cost 10 times the estimated price. A technological feat, it is an economic nightmare for the government that commissioned it.

It may be more than coincidence that in April, when the English and French transport ministries invited competitive bids for a fixed link between France and England--a span only two kilometers shorter than the Seikon Tunnel--they insisted that it be privately financed. Investors without the resources of a national treasury can’t afford such miscalculations.

Submerged Caisson System

If two teams are twice as fast as one, why not, suggests the Mid Channel Access Corp., proceed even faster with three teams, or even four? Although MCAC is not connected with any of competing bidders, the corporation hopes that whichever group wins will use their newly patented submerged caisson system.

Caissons were developed in the 19th Century during construction of the tunnel connecting Jersey City with Manhattan under the Hudson River. They allow workers to adjust slowly to the change in air pressure, avoiding the bends. The new technique provides portable caissons that can bore through the chalk in the middle of the channel so that drillers can proceed laterally in both directions, at the same time as other teams are tunneling out from sites on shore. Without having to build an artificial island that would obstruct shipping, submerged caissons allow tunnelers to be lowered from a ship to a hatch on the bottom of the sea. If the builders use several caissons, a tunnel only two kilometers shorter than Seikon could be finished in less than five years.

Of course the range of error is theoretically enormous. If the instrumentation somehow goes awry, the English “chunnel” could end up with four parallel tunnels, each taking four times as long to complete.