What does it take to dig underground tunnels?

**Krishna** (0:00)
The Daily Brief Today, I'll talk about two interesting stories. In the first one, I'll talk about what does it really take to dig underground tunnels, and in the second one, I'll talk about the chemical sector's trade problem. Welcome back to The Daily Brief show by Zerodha, where we cut through the news to help you understand what's actually happening in the most important stories from business and markets. I'm your host Krishna, and today is Monday, 27th July.
This spring, workers at Gansoli in Navi Mumbai lowered the last parts of a massive machine into a shaft underground and assembled it there.
This machine is longer than a football pitch, about as tall as a four-storey building, and it weighs more than 3,000 tonnes.
On 18th July, the machine began its journey towards Vikroli, boring a 10-kilometre tunnel that will eventually pass under the Thane Creek.
It will become India's first undersea railway tunnel. Thirteen days earlier, a second machine had started digging from Vikroli towards Bandrakurla complex, passing beneath some of Mumbai's busiest neighbourhoods. These are the two tunnel-boring machines building the underground section of the Mumbai-Amdabad bullet train. Over the next few years, these two machines will dig through kilometers of earth that no one has ever seen before. But why build a railway underground at all? Wouldn't it be simpler to build it on the surface? Now it would, but sometimes the surface simply stops cooperating. Trains can't climb steep slopes or take sharp turns. They need gentle gradients and wide curves, and high-speed trains need them to be even gentler. Try fitting that through a city like Mumbai, and you quickly run out of space. The route runs into roads, buildings, utilities, water bodies, and everything else already sitting on the surface. Take the Mumbai-Amdabad bullet train for example. Between BKCA and Shilpata, the only practical route runs beneath Mumbai and the Thane Creek.
So, there isn't much choice but to dig.
Now, before the digging starts, engineers try to find out what's underground. They drill boreholes, run seismic surveys, and study the rock and soil. The International Tunneling Association is blunt about this and says that it isn't just paperwork. This stage decides the route, the construction method, and ultimately the cost. But there's a problem with this process. A tunnel is one continuous line through a huge volume of ground, while boreholes only tell you what's happening at the few places where you drilled. Now everything in between is an educated guess. And sometimes that guess can be wrong. A fault might lie between two boreholes, or a layer of water-bearing sand might appear where the model predicted clay. Now that's the reality of tunneling. No matter how much you study the ground beforehand, there will always be uncertainty. Now to do this inherently uncertain job, a contractor has to turn all that uncertainty into a single number. It has to estimate how fast the machine will move, how quickly its cutters will wear out, how much support the tunnel will need, and how many delays might crop up from problems nobody can predict yet. Then it has to put a price on all of that and promise a finish by a certain date. That's why the contract matters so much. It has to clearly define who bears the risk when the ground turns out differently than expected. Instead of leaving it open to endless disputes.
Modern underground contracts try to do exactly that. The FIDIC contract for tunneling is built around something called a geotechnical baseline report.
Think of it as an agreed description of the ground conditions that the contractor is expected to price for. If the actual ground broadly matches that baseline, the contractor bears the risk. But if conditions turn out to be materially worse, it allows the contractor to potentially claim extra time or money. It also helps the project owner. Without a shared baseline, cautious contractors would bid more than needed to cover for every possible surprise, while aggressive ones would bid low and fight over claims later. Now, the project owner doesn't want either. However, the bullet train project doesn't use the specialist ternling contract. The public tender refers to an older FIDIC contract where the employer provides their design. But the documents that explain how geological risk is actually divided between the two sides weren't made public. They were available only to bidders who paid a fee and signed a non-disclosure agreement. So, from the public record, it's impossible to know whether the project uses a formal geotechnical baseline or something similar. Now, let's say you have won the contract. You can still start digging. First, you have to build the machine that will do the digging. Before tunnelling begins, workers excavate large lawn shafts where the tunnel boring machine is assembled underground. Now, the machine itself doesn't arrive in one piece. Its parts are lowered into shafts one by one and bottled together below the surface. And despite the name, a tunnel boring machine isn't just a giant drill. The spinning cutter head at the front is only one part of a much larger system. Now, behind it are power systems, ventilation, equipment that installs the tunnel lining, and all the machinery needed to keep the operation running. Now, in other words, it's an underground factory. There are many ways to excavate a tunnel, depending on the ground and the project. Here, we are only looking at one of them, tunnel boring machines or TBMs. Once the machine is assembled, it settles into a cycle that it repeats thousands of times.