Arctic pipelines and the future
The development of Arctic petroleum pipelines has always suffered from huge mood swings.
In North America, the earliest Arctic line was the 100-mm Canol pipeline from Norman Wells to a refinery at Whitehorse, completed in 1944 and required in order to support the war effort in Alaska. There was much development in the 1960s and 1970s, starting with the marine pipelines in the Cook Inlet of Alaska, not by any means a straightforward project and the subject of a remarkable film. The presence of oil close to the Canning River in northern Alaska had been known since the early 1900s, but already in 1925 geologists had recognised that it would be difficult to export oil by tanker.
In the late 1960s, it turned out that the Prudhoe Bay field close to the Arctic Ocean shore of Alaska was very large indeed. Tanker transport was a possibility, either eastward to the Atlantic through the North-West Passage or to westward around Point Barrow and through the Chukchi Sea and the Bering Strait to the Pacific and California. The tanker options had grave difficulties. The North-West Passage had been dreamed about for three centuries, but only a handful of ships have ever actually traversed it, and that only in summer. It is some 3500km from Prudhoe Bay to relatively open water in Davis Strait, all of it is covered with thick ice in winter, and some difficult sections ¬¬– such as M’Clure – have heavy ice all the year round. Going westward instead, the distance to Bering Strait is less than half at 1500km, but the area around Point Barrow is a notorious choke point, so that that area may be blocked by ice when the sea further east is open. However, that option was pursued as a demonstration project. The tanker Manhattan was ice-strengthened and reconstructed at a cost of $50 million, and in 1969 it was taken through the North-West Passage from Philadelphia to Prudhoe Bay and back. It brought out one symbolic barrel of oil, but it was concluded that year-round transportation of 2Bb of oil a day would require a pipeline. Reportedly, that decision had already been taken before the tanker demonstration had been completed.
Other options were looked at. The Russians had for many years moved large amounts of oil to the Russian Far East by rail, and some railroad enthusiasts proposed that Alaska Railroad be extended to the Arctic shore, or alternatively that a railroad might be built to join the existing Canadian railroad at Great Slave Lake. Those concepts were considered seriously at the time, though it was pointed out that so many tankcars would be needed that it might be simpler just to park them and weld them end-to-end to make a pipeline.
A pipeline to an ice-free port on the southern coast of Alaska was the obvious and more sensible choice. The Alaska pipeline was conceived, survived huge controversy, and constructed in the late 1970s. The controversy was extreme: conservationists argued that the pipeline would ‘ruin Alaska’, but even the people most hostile to the petroleum industry do not advance that argument today. Coates  has written an instructive history of the lengthy argument, thought-provoking to anyone involved in pipelines, and there are many other books.
Meanwhile, there was lots of gas, both in the Mackenzie Delta area and close to the oil at Prudhoe Bay. The Canadian Arctic Gas pipeline, an earlier version of the current Mackenzie Gas pipeline project, was planned and went to lengthy public hearings. The Berger Inquiry recommended a 10-year moratorium on construction, essentially not for technical reasons but on socio-political grounds and because First Nations’ land claims were not settled. Equally ambitious competing projects were put forward, founded on the notion that hydrocarbon prices would rise inexorably and that $10 gas was only a year or two away. One of them was the Polar Gas pipeline to bring gas from Melville Island  through the Arctic Archipelago, perhaps south to the west of Hudson Bay, or south-east to join up with a Mackenzie Valley line, or even east across the mouth of Hudson Bay and down through Quebec. There were several competing Alaskan gas pipeline projects, the most attractive following the Alaska Highway to Alberta.
Hardly anything more was done on the ground, except for a few short lines tied-in to the Alyeska system, and one experimental subsea flowline bundle in the Arctic Islands . In retrospect, it can be seen as providential that the big projects did not go ahead in the early 1980s: the investors would still be waiting to get their money back.
At the same time, and on the other side of the world, the Russians were developing the mammoth gas province of the Yamal peninsula, and building large-diameter pipelines, first to their sphere of influence in Eastern Europe and later both further westward to the EU and eastward towards China. They bought western pipe, but otherwise worked largely alone, and in the process developed new technologies such as flash-butt welding that remain largely unknown in the west [4, 5]. They did almost nothing under water.
In 1986 the mood collapsed with the oil price. Oil companies stopped design projects, eliminated most of their specialists, and lost their expertise. Much knowledge was lost, and some of it will have to be expensively rediscovered. Later, confidence slowly crept back. Prices recovered, political change made projects in the former Soviet Union practicable, and of course the hydrocarbons are still there. The northern Caspian is only 46ºN but shares many of the problems of the Arctic, such as ice-covered sea, extreme low winter temperatures, and ice gouging. The reserves in that region are enormous: the oil-in-place in the Kashagan field is estimated at between 30 and 50B barrels, and one estimate of the development cost puts it at $136 billion. Two comparatively small Arctic offshore pipeline projects were carried out at Northstar (once called Seal Island, a name abandoned for reasons of environmental tact) and at Ooogaruk [6, 7], both in shallow water.
In 2011 we are facing another collapse of confidence in ambitious Arctic projects, particularly under water. That has been brought about by two factors: the Macondo catastrophe in April 2010, and the unexpected development of alternative gas reserves in the north-eastern and western USA. It is hard to overestimate the impact of Macondo, comparable with the impact of Chernobyl on the nuclear power industry, as the writer has argued elsewhere. On 19 April, 2010, almost anyone in the offshore industry would have argued that the industry was on top of its game, and that though there might be occasional mishaps, competence had reached a level at which major pollution events were unlikely. Ongoing disasters like Lusi in Indonesia could be explained away. A week later, that argument would have been met with hollow laughter, and the inept response of the industry made matters worse. That catastrophe took place in early summer and as close as imaginable to the technical and commercial stronghold of the industry. Nonetheless it led to a five million barrel spill, took BP four months to fix, discredited the whole industry, and will lead to litigation that may well last for decades. In the Arctic context, it was pointed out that the situation would be far, far, worse if there should be a similar mishap in the Beaufort Sea, most of all if it were to happen when the sea ice was too thick to be easily broken by ships but too thin for vehicles. Some technology for Arctic oil clean-up exists, but experience is extremely limited.
Many of the technical difficulties remain. Progress has been disappointingly slow, and many of the issues identified 40 years ago have not been solved satisfactorily. On land, one issue is frozen ground and differential settlement: the amount of ice contained in frozen ground varies enormously, and there can be big variations within a few metres, both horizontally and vertically. It is difficult to follow the first principle of construction on permafrost – Tsytovich’s injunction  – to change the thermal regime as little as possible, and therefore to make the pipeline temperature coincide with the ground temperature. If the pipeline is warmer than the ground, the permafrost under and around the pipeline thaws, and the pipeline settles further where the ground had been ice-rich, and less far where there had been less ice. The pipeline bends in response, and the bending can overstress the pipe wall and cause it to buckle. If, on the other hand, the pipeline is colder than the ground, the soil beneath the pipe freezes progressively, and migration of water towards the freezing front causes the soil to heave, and again the pipeline is compelled to bend and may buckle locally. That bending can damagingly interact with other forms of buckling such as lateral and upheaval buckling. Much remains to be learned about how to carry out the enormous amount of geotechnical survey work required in permafrost areas, particularly in discontinuous permafrost. Yet another complication is the effect of climate change.
An alternative is to support the pipeline above the ground surface. The supports can be simple piles, or they can more sophisticated thermopiles that actively keep the ground frozen around the bases. Thermopiles can be subject to internal corrosion, and thermal surveys have suggested that they do not invariably operate satisfactorily. Over several hundred kilometres of the Alaska oil pipeline, those options were selected conservatively whenever it was uncertain if a buried line would be safe. They are expensive, they are visually intrusive, there have to be crossings for wildlife (though caribou turn out to be rather intelligent), an above-ground system may be adversely affected by earthquakes, and the pipeline remains vulnerable to malicious damage and to fools armed with rifles. Much remains to be learned about how to design and construct an above-ground system and to be sure that it functions correctly.
Turning to Arctic pipeline construction in the sea, one of the hazards is ice gouging. Floating ice runs aground in shallow water and scrapes along the seabed, driven by wind, current, and the pressure of other pieces of ice driven along behind it. The ice cuts into the seabed, and forms a dense network of gouges, a few of them very large indeed and, in an extreme cases, 50m broad, 5m deep, and hundreds of metres long [9, 10]. A back-of-the-envelope calculation shows that the force required to make such a gouge can reach several thousand tonnes, so that if the gouging ice mass should encounter a pipeline, the line would inevitably be damaged severely. Worse still, a pipeline below the level at which the ice might strike it is still not necessarily safe, because the ice drags along some of the soil beneath it, and would carry with it a buried pipeline and could bend it severely.
Gouging and subgouge deformation remain difficult and controversial issues, and it is not an accident that several of the papers in this issue of the Journal of Pipeline Engineering are devoted to it. At the end of the day, the engineers responsible for a project have to make a concrete decision and select a trenching depth, rather than hypothesising about encounter probabilities. A possibility that has received less attention has been that of proactively altering the design strategy by doing something more than just burying the pipeline. Our research indicates that at least the effects of subgouge deformation can be eliminated by interposing a weak layer than cannot transmit large forces downwards above the pipeline and below the maximum gouging depth. More needs to be done to look imaginatively at alternatives.
That remark applies to other geohazards that might be a threat to Arctic marine pipelines. One of them is ‘strudel’ scour [11, 12]. In the early summer, ice on the Arctic rivers melts first while the sea is still frozen. Fresh water flows out over the sea ice, finds holes and cracks in the sea ice, and flows downward, generating powerful rotating whirls (known as strudels). Under each strudel is a downward-point jet, and that jet erodes the seabed. If the scour hole it creates intersects a pipeline, the pipeline might be damaged by vortex-induced oscillation, the line might be overloaded, and the strudel might interact with other forms of sediment transport, such as the formation of sandwaves. Some research suggests that the probability of damage from that source is relatively low, but on the other hand it has been suggested that the presence of a pipeline might encourage the formation of a strudel immediately above, because heat from the pipeline would be convected upwards and thin the ice.
These and other technical issues will rightly be scrutinised with great care, before any decisions to build Arctic pipelines are reached, whether offshore or onshore. Handwaving and appeals to industry experience and competence will not be enough. The unfortunate experience of the past year will heighten awareness. There is much to do!
1. P.A.Coates, 1993. The Trans-Alaska Pipeline controversy: technology, conservation, and the frontier. University of Alaska Press, Fairbanks.
2. A.C.Palmer, R.J.Brown, J.P.Kenny, and O.M.Kaustinen, 1977. Construction of pipelines between the Canadian Arctic Islands. Proc. 4th International Conference on Port and Ocean Engineering under Arctic Conditions, St John’s, Newfoundland, 1, 395-404.
3. A.C.Palmer, D.J.Baudais, and D.M.Masterson, 1979. Design and installation of an offshore flowline for the Canadian Arctic Islands. Proc. 11th Annual Offshore Technology Conference, Houston, 2, 765-772.
4. Flash butt welding video, McDermott, Inc, 1990.
5. G.O.Andersson and M.Weidemann, 1985. Flash butt welding for S-lay barges. Proc. 17th Annual Offshore Technology Conference, Houston, 261-273, OTC 4869.
6. G.A.Lanan, J.O.Ennis, P.S.Egger, and K.E.Yockey, 2001. Northstar offshore Arctic pipeline design and construction. Proc. Offshore Technology Conference, Houston, OTC13133.
7. G.A.Lanan, T.G.Cowin, B.Hazen, D.H.McGuire, J.D.Hall, and C.Perry, 2008. Oooguruk offshore Arctic flowline design and construction. Proc. Offshore Technology Conference, Houston, OTC19353.
8. N.A.Tstyovich, 1975. Mekhanika merzlykh gruntov (The mechanics of frozen ground) English translation: Scripta Book Co, Washington.
9. A.C.Palmer, I.Konuk, G.Comfort, and K.Been, 1990. Ice gouging and the safety of marine pipelines. Proc. 22nd Offshore Technology Conference, Houston, 3, 235-244, OTC6371.
10. C.M.L.Woodworth-Lynas, J.D.Nixon, R.Phillips, and A.C.Palmer, 1996. Subgouge deformations and the security of Arctic marine pipelines. Proc. 28th Annual Offshore Technology Conference, Houston, 4, 657-664, OTC8222.
11. A.C.Palmer, 2000. Are we ready to construct submarine pipelines in the Arctic? Proc. 32nd Annual Offshore Technology Conference, Houston, OTC12183.
12. W.F.Weeks, 2010. On sea ice. University of Alaska Press, Fairbanks.
Professor Andrew Palmer
Centre for Offshore Research and Engineering
National University of Singapore