London academic John Burland has spent the past 10 years on the toughest job in construction. Here’s how he stopped the Italian landmark collapsing without destroying that famous lean.

Ten years ago, John Burland was sitting in his office when the telephone rang. It was an old friend, Michele Jamiolkowski, president of the International Committee for the Safeguard and Stabilisation of the Leaning Tower of Pisa. “He’d phoned to tell me of the poisoned chalice he had just been handed,” says Burland. “He’d been told by the prime minister of Italy to take action to stop the Leaning Tower of Pisa leaning any further. I told him: ‘You have my sympathy, that’s an awful responsibility’.

“And he replied: ‘Keep your sympathy for yourself. Your name’s down to head the team that has to find the solution.’” With one phone call, Burland had been handed one of the toughest challenges in construction.

Burland is professor of soil mechanics at the department of civil and environmental engineering at Imperial College, London. Dressed casually, sitting behind his paper-strewn desk, he looks every bit the academic. “The problem with Pisa is that there are no precedents,” he explains in his quiet voice. “It is unique.”

After years of gradually leaning more and more, the tower was in imminent danger of toppling over. “The inclination had steadily increased to a point where it was just about to fall over,” explains Burland, leaning forward to emphasise his point. “Worse still, the stresses in the masonry walls caused by the lean were so intense that the marble was about to explode. To do nothing would have been a recipe for disaster.”

The tower was built as a campanile for the nearby cathedral. Construction started in 1173 but stopped abruptly twice before the tower was completed in 1370, nearly 200 years later.

The reasons for the stoppages in the tower’s construction are not clear, but without them, says Burland, the tower would have fallen over. The gaps gave the soil below the tower time to settle and compress, to better support the structure. “It’s a tower built on a bed of jelly,” he adds.

The 60 m high tower is actually built on highly compressible clay. So, no matter how carefully the structure is built, once it reaches a critical height, the smallest upset will make it lean dangerously. “It’s something small children building brick towers on a soft carpet will be familiar with,” says Burland. Over the years, the tower’s lean has gradually increased until the seventh cornice now overhangs the ground by a staggering 4.5 m.

The problem Burland faced was to correct the lean without causing the tower’s collapse. The tower was so delicately balanced that the team charged with saving it dared not touch the ground on the south side, the direction in which it leans. “It was not an underpinning job,” says Burland. “Any interference with the ground on the south would have brought it crashing down – and I’d have had to take up selling ice-creams in the piazza,” he jokes.

Although Burland had to decrease the stresses in the tower’s masonry and stabilise its foundations, he could not correct the lean too much. “There’s no way we could have had a not leaning tower,” he says, smiling. “Besides, it’s a historic structure of great significance – you can’t change its character. Nor could we pull it down and rebuild it.”

To give the team some breathing space, Burland introduced temporary structural measures. To reduce the likelihood of the marble failing, temporary steel tendons were wrapped around the tower’s first and second storeys. Then, to improve the stability of the foundations, a prestressed concrete collar was clamped around the base of the tower and 900 tonnes of lead were placed on its higher, north side.

With the tower now stable, for the short term at least, Burland started his search for a permanent cure. “It had to be an ‘ultra-soft’ solution, with no concentrated loads on the tower and its foundations,” he explains. Various options were assessed, including pumping ground water from one side of the tower and loading the ground around the tower to compress it.

Finally, though, the team opted for a technique called soil extraction, which works by creating lots of little tunnels to create a controlled form of subsidence. “The $50 000 dollar question was whether soil extraction would increase or decrease the tower’s lean,” says Burland.

Before the tunnelling rig was allowed anywhere near the tower, a series of tests had to be carried out. Computer and physical models of the tower were studied for clues. “Everything had to be pursued in enormous amounts of detail, and a proposed solution only evolved after years of intense argument.”

When there was nothing left to be tested in the laboratory, a trial was carried out on the piazza, close to the tower itself. A full-sized mock-up of the foundations was constructed by sinking a 7 m diameter shaft and filling it with a concrete imitation of the tower’s base.

This was then loaded with weights on one side to simulate the loads experienced by the tower’s foundations.

Next, a line of continuous-flight augers were positioned around the unloaded side of the mock-up and the soil was carefully extracted. The tests proved that the system worked and allowed the team to develop the drilling technology.

For Burland, however, the most important thing was that “a system of command was developed”. This involved the drilling team sending him two faxes every day, one in the morning and one in the evening, showing how the structure was performing. Burland would then fax back an instruction saying which augers should be used the next day to remove more earth.

Biting the bullet

With this test successfully completed, says Burland: “We had to bite the bullet and go for it.” Work started on the tower in February. This time, 41 augers were installed along its north side. Sensitive measuring equipment was fitted to monitor the tower’s response to the works. The burden was then on Burland to decide which auger to use to start the process of slowly and carefully removing soil from beneath the north side of the historic monument.

The augers were run slowly, one at a time. About 100 litres of earth a day were removed from beneath the tower’s foundations. Back in his office at Imperial College, Burland kept an anxious eye on the fax machine. As the results started to accumulate over the weeks, he could see the process was actually working. The tower started to right itself slowly.

But Burland could not relax completely. Every day, he had to look at the measurements from the previous day and decide which auger should be turned next to remove the soil. “It’s like riding a bicycle by fax,” he says, clearly relieved that things are going well. “I’m slowly steering the tower steadily due north.”

Now, six months into the project, Burland is still manning the controls. “So far, the tower has moved back about 190 mm at the seventh cornice,” he says, leaning back in his chair. The solution has worked so well that the team has even removed some of the lead ounterweights. “I’m not relaxing, though,” he adds. “You cannot afford to on a project as important as this one.”

The task of removing soil is expected to be completed by the end of February 2001, four months earlier than first planned. When the works are complete, the seventh cornice will have moved back about 390 mm, while the top of the tower itself will be 500 mm closer to the vertical. This is predicted to reduce the stress in the masonry by about 10% and will correct the lean by a similar amount – not enough to be noticed, claims Burland.

“That’s all there is to it,” he says. “Apart, that is, from the feeling of sheer terror and trepidation of being the man held responsible.”

Extracting the earth beneath the tower

At Pisa, 200 mm diameter drills are being used to slowly tease the soil from beneath the tower. These comprise two hollow sleeves, one inside the other, with a hollow-stem auger slotted into the centre core of the inner sleeve. To insert the drill into the ground, the inner sleeve extends, rotating as it moves forward. The auger moves forward with the sleeve, also rotating, but in the opposite direction. This ensures that surrounding soil is left undisturbed. As the sleeve and auger advance, a core of about 100 litres of soil is drawn into the sleeve. To remove the earth, the inner sleeve and auger are then carefully withdrawn, taking with them a plug of soil and leaving a 200 mm diameter tunnel in the ground. Under the weight of the tower, the remaining cavity gradually collapses. This process is carefully repeated, one drill at a time, until the subsidence is enough to right the tower.

How John Burland kept Big Ben standing

In Pisa, John Burland used a process of extracting soil from beneath the tower to correct its inclination. Closer to home, he is stopping Big Ben starting to lean by doing the exact opposite: filling the ground beneath the tower. The clock tower’s problems started with the excavations for the extension to London Underground’s Jubilee Line. As part of the extension, two new train tunnels were planned running one above the other next to the clock tower. Next to these, just 31 m from the base of the clock tower, the deepest open basement in London was to be sunk – the 40 m deep escalator box to connect the new platforms to Westminster Station above. “Whenever you dig a tunnel, it causes some subsidence,” says Burland. But he was unable to predict the combined effect of the separate excavations, so he decided to install a compensation grouting system. This involves injecting grout under huge pressure into the ground at critical areas to counter subsidence. To install the system, a large vertical shaft was sunk beside the clock tower. Then, a series of 125 mm diameter horizontal holes were drilled, fanning out towards the tower. A 73 mm diameter steel grouting tube was inserted into each hole. As work started on excavating the first tunnel, the tower started to move. Burland says: “It moved much more than we had predicted, but not enough to need the grouting.” However, following the tunnel drive, the movement continued to increase significantly so that the top of the tower, which had initially moved 4 mm, started to tilt at the rate of 1.1 mm a month. Burland hurriedly revised his calculations: “We realised the works could make the tower lean by up to 50% more than we had first expected.” With the first tunnel still to be widened and the escalator box yet to be sunk, the protective measures suddenly became essential. Before tunnelling could restart, the team had to test the grouting system to prove its effectiveness. A sophisticated tilt-measuring system was installed in the tower. As the grout was introduced, the tower responded and its tilt was reduced by 5 mm. As each subsequent tunnel was driven, the tower’s tilt was closely monitored. Every time the tower’s lean reached a danger level, grout was injected. The precise volume and location of each injection of grout was carefully managed to counteract the tower’s movement. In all, a total of 24 grouting episodes were carried out between January 1996 and September 1997, in which 122 m3 of grout was introduced. “This is estimated to have saved the tower tilting 100 mm,” says Burland.