Much of the UK’s North Sea natural gas comes ashore at the Easington terminal near the north bank of the Humber estuary. From there it flows through a pipeline that takes it south under the estuary of the River Humber to join the national grid of gas pipelines which carry it into the heart of England. The Humber pipeline carries up to 25% of the gas that is used in the UK; it is clearly a key part of the national infrastructure.
“This is a very important pipeline, of real strategic importance,” says Steve Ellison, senior project manager for National Grid. The current pipeline is of steel, concrete-coated, and was laid in 1984. It is now experiencing cover erosion due to the tidal nature of the Humber estuary. It was installed by cut-and-cover on the river-bed. As expected, currents and tides have played an active role over the years, with the ever-changing channels and flow patterns of the River Humber seeming to follow a 100-year cycle of activity. Over time, these currents have worked to scour different pathways; the result is that the river bed has eroded in places, leaving parts of the pipe uncovered and exposed. That of course presents a risk.
“The problem became apparent in the early 2000s,” says Ellison. Regular surveys of its condition were undertaken, and work to try to protect it began in 2009. Frond mattresses were laid over the exposed portions of the pipe: these are concrete rectangles with polythene or plastic ‘fronds’ on their upper sides that resemble and mimic the action of seaweed; the idea is that they trap tide-born sand and silt and so build up a protective layer. Installing them is a challenge in its own right, given the low visibility of the estuary and the need for it to be done at slack tide. A long-term solution was needed.
“The pipeline is called Feeder 9” says Ellison. “Another feeder, called Feeder 1, also crosses the Humber estuary close by. The same factors affected it, but to a greater extent – so much so that it had to be shut down. After it was decommissioned it was filled with nitrogen – and it failed. That was a warning that we might well have a real issue with Feeder 9 that needed to be resolved by a long-term solution.”
Several options were considered. Simply installing a new pipeline using cut-and-cover was ruled out, by legislation if by nothing else. The Humber Estuary, with its intertidal mudflats, sands and coastal lagoons, is hugely important for biodiversity making it a bird overwintering site of international significance. It is a designated Ramsar site (the UNESCO convention on wetlands); a Site of Special Scientific Interest (SSSI – a special area of conservation for flora and fauna) and a special protection area for biodiversity. Part of the northern bank is an RSPB nature reserve. “It is one of the most designated areas in the country” says Ellison. “Protections don’t get much higher.”
Given that level of protected status, repeating the open-cut technique of the 1980s would not today be allowed. In any case, such an installation might well suffer the same fate as its predecessor, of erosion and consequent exposure.
“We considered alternative routes” says Ellison. The estuary cuts deep inland. An offshore route of 130km, with 70km more onshore, would need added compressors to maintain the flow; a 190km inland route would bypass the estuary but the disruption and destruction would be huge. Other combinations were examined; but this was 5km of pipe that needed replacing, so building 100km-plus diversions was not sensible. All such options were overkill: prohibitively disruptive, expensive, and labour-intensive, and far bigger-scale works than called for. Clearly the river crossing was the straightforward solution. But if an open-cut on the river bed was no longer possible, how could the crossing be achieved?
“Horizontal directional drilling [HDD] was considered” says Ellison. “The pipeline needed to be around a metre in diameter. HDD has been done at that size but not over that distance. The record so far is 3km, and we had 5km of estuary to cross; so the technology for that solution is not yet with us.” Direct pipe-pushing was looked at, but again the distance and diameter ruled it out.
“Yet a third method would have been by immersed tunnel, which is sometimes used for railways and roads. But it is very, very expensive, and the needs case economy did not support it; and it too would have run foul of the Ramsar protective legislation.”
“So we were led to a different solution: a tunnel under the bed of the estuary, to be bored by TBM and lined with concrete segments with, once it was completed, the steel pipeline pushed through it from one end until it emerged at the other.” The £150m (US$200m) project would be a world record for this method of construction.
“The TBM tunnel was to be 3.65m inside diameter. It is in effect a very large piece of temporary works: it’s a hole for us to push our pipeline through.”
On any other project, a TBM boring under an estuary would be an achievement in itself; here, the 4.38m-diameter Herrenknecht mixshield TBM was just the preliminary enabler for the still-more-challenging pipe push involving two 500t-thrust machines pushing pipe at a rate of around 1m/min.
That TBM enabling tunnel was a challenge in its own right. “The TBM we used was a Herrenknecht slurry EPB machine” explained Ellison. “The alignment was 12m under the river bed and around 30m below ordnance datum, which is not a lot. On one occasion we had to get to the face of the machine to replace the cutters. It was done hyperbarically, under compression, and at low tide, when there was less pressure of water. The tidal range there is about 8m between low and high tides.
“The TBM cut at 4.4m diameter. The segments had a 4.2m outside diameter. We were keen to monitor the grouts on the annular fill; we inspected daily to make sure no material, sand or silt, was coming in, which would have meant ground loss around the tunnel and a possible threat to it.”
The geology was chalk and fractured chalk, with overlying glacial deposits and alluvial silts. “On the northern side, the alignment passes under an RSPB bird reserve, where we could not drill exploratory boreholes. So we did not have absolute knowledge of the ground conditions there. For safety’s sake, therefore, we stayed deeper in that particular section, in the known chalk, then came up at a steeper gradient than on the other side.”
That steeper gradient dictated the use of MSVs rather than rail for segment delivery. Bryony Brown was project engineer on the site. “We used filter presses to dry out the slurry and used it to restore a quarry on the Paull side” she says. “160,000 tonnes of TBM material needed to be taken down rural roads and used to help restore a former quarry; the whole project was in a very rural area, which added to the logistical challenges. The front section of the TBM alone weighed 225t, and that had to be delivered down the same kind of roads.”
The alignment is a straight line in plan, with gradients at each end. The crossing point itself was fixed, determined by existing gas installations at Goxhill on the southern side of the estuary and the village of Paull on the northern side. “We had to tie in closely with those, so that served to fairly narrowly define the route corridor.”
“The Humber hasn’t been tunnelled before” says Brown. “A tributary was tunnelled under in 1999 for a sewer; the geology is similar there, but they had a failure on it, so that was in the forefront of our thinking.
“There have been similar river crossings, across the River Ems in Germany and the Netherlands, which was 4km. In Queensland, Australia, Santos ING tunnelled 4.2km under an estuary and saltmarsh. Our project was for 5km. So this project was at the forefront of engineering.”
As well as delivering the project, the National Grid team had to design it. Furthermore, the second stage of design, for the pipe push-through, was not easy.
“For the project’s Development Consent Order, we needed to give quite detailed information” says Ellison. “We needed to deliver a 120-year lifespan; but the design life of the steel pipeline is 40 years. So we used a dual-layer fusion-bonded epoxy coating and a concrete coating layer to protect it during installation. The pipeline is also protected by a designed cathodic protection system for its operational life.
“We needed to design all this, but there were feedback loops. If we damaged the epoxy coating while we were pushing the pipe into place, then the cathodic protection (CP) system would have to work harder; which would mean a greater current [would have to flow] to protect the pipe. But a greater CP current has a negative effect on the concrete lining of the TBM tunnel and degrades it. So, damaging the pipe coating as we put it in place would have serious consequences for the lining of the tunnel. That was a very big concern for us.
“We had planned to put the pipe in place with the coating exposed but the risk of damaging it was too great: we needed to keep the CP current at a low value for the life of the pipeline. So, we gave each section of pipe a coating of concrete. That of course increased its weight and increased the power we would need to push it – and we would be pushing eventually 5km of it; so that was a big change and a big challenge.”
Delivery
Delivering the project was, among other considerations, a major task in logistics, explains Brown.
“The pipe itself is heavy-wall steel. It was made in Germany, and transported in 11m lengths to Edinburgh, where the coating and the concrete were applied.
Then they were transferred one by one to the pipe-laying yard at Goxhill on the southern bank of the estuary. We worked in strings of 620m, and the pipe was now weighing 836 tonnes per string. It comes out at 6,600 tonnes in total weight that we would be pushing by the time we had 5km of pipe in the tunnel. So the next problem was how to get it pushed.
“We had two thrusting machines” she says, “one to push and one to hold the pipe in place during the return stroke. The contractors at Edinburgh had weighed each pipe individually. There was some variation between them of course; and we arranged to put the lighter pipes in first, so that we would be pushing the lightest weights the furthest. As the string being pushed down the tunnel became increasingly longer, the heaviest pipes would go in last and only have to be pushed a small distance.
“We constructed a system of bespoke bogeys to move each string, all 620m of it, from the laying-out yard to the pipe-run entrance. That was complex engineering in itself, with pulley systems, and various motors. The distance they had to be moved was approximately 80m.
“Once the push had begun, we were conscious that we needed to know the parameters. Engineers from de la Motte, which had done the River Ems crossing this way, gave us advice and performed calculations for us of what was possible and what was not. That gave us the data to monitor or to expand our pushing forces.”
Friction, and the need to minimise its effects, was of course paramount. “We had polyurethane collars around the pipe, at about 6m spacing, to keep the pipe off the tunnel floor and take the frictional forces, but we took two special measures as well” says Brown. “During the TBM tunnelling stage, when the concrete segments were being installed, we took care to make the joins as smooth as possible. We inspected every one and we knew where the steepest bumps were, which told us where the danger of hang-ups would be greatest. We did not want a situation where two collars were hung up at the same time. To prevent that, we staggered the collars: we offset them by small, calculated distances so that they were unequally spaced. That meant that if one got caught up at a segment join, the others would be in middle of segments rather than at joins, so that only one would be catching at a time.
“Our second measure to minimise the effect of friction was to fill the TBM tunnel with water. The water acted as a lubricant and it also, through buoyancy, lowered the weight of pipe on the collars. The pipe was never quite floating, but its ground-weight was reduced. This water remained after installation, acting as an annular fill between pipe and tunnel, but that in turn required that other measures had to be taken: it had to be water of very specific purity.”
Several factors were involved in this. “The cathodic protection current imposed conductivity requirements on the water. In addition, microbiological activity can eat away at the pipe seals. Given that we were looking at a lifespan of 120 years, we needed to eliminate this. It was essential that the water contained no nutrients that micro-organisms could survive on,” said Brown. This requirement imposed on them one of the most difficult parts of the project. “A water-processing plant was set up at the Goxhill entrance. It took a long time and many adjustments to get the right standard of water from it.”
A nose-cone with a towing-hook was welded to the front pipe in case a pull-through from the northern end of the tunnel became necessary. “Happily, it was not needed,” says Ellison. “The pipe was pushed successfully, all 4,963.7m of it, which at the end qualified us for a Guinness Book of Records. They gave us a certificate to that effect, although because of the pandemic, the ceremony had to be a virtual one.”
“Logistically it was a monumental task, and we took good care to engage with the local community” he says. “They were of course affected by all the disruption; and we had open days, science days at local schools and colleges, all kinds of things like that. It felt really good to be with an engineering project that could enthuse these youngsters. It felt very positive; and some of them will become the engineers of the future.
“The country could one day abandon natural gas for hydrogen as a much more climate-friendly fuel” says Ellison. “If it does, the pipeline will be part of it. It could carry hydrogen instead of natural gas, so another satisfaction is that the project is future-proof.”
By December 2020, the project had been completed and the pipeline was delivering gas to the UK network.
