Antarctic Engineering: Halley VI
Last week I attended the IABSE Annual Lecture (2014, just in case you thought I was part of the Dolorian drivers club) by Peter Ayres about his work on Halley VI. Taking that this is the third time I’ve mentioned an IABSE event, I should probably start by giving them a plug– they run some interesting events, and you don’t have to be a member to attend most of them.
Peter Ayres is a project director at AECOM who, in his own words, “loves doing unusual projects that break the rules.” He is one of the few engineers in the world who can say they’ve worked on every continent across the planet.
“Never good with numbers, [but] good with ideas,” he’s an interesting presenter; and although I’d already heard from him about the Antarctic research station before, for this talk he had mixed it up a little, and added a lot more detail.
Even without reading this article, I suspect you know that engineering for Antarctic conditions is difficult; so perhaps the first question is “why bother“. The answer is simple: “For Science!” The remote nature of the continent means that there’s an excellent view of space, we can detect things like holes in the Ozone, and use ice cores, animal population, and air composition to measure the effects of people on the planet.
Halley I (The Polar Menace) was built about as far south as we could make it in the 1950’s. Although we can get a damn slight further now, Halley VI (The Return of the Snow) is pretty much in the same place- why? Because the closer it is to Halley V (The Penguins Strike Back), the more relevant our 50 years of already collated data is. Unfortunately, however, we couldn’t do anything as simple as knock down No. V and build something at exactly the same site, because it’s standing on the wrong side of a huge ice cantilever that predictions say will separate from the mainland during the design life.
The remoteness that makes the Antarctic so good for science, creates one of the biggest constraints. To get to the site of Halley VI you need to take a boat, and then drag your payload for miles over meter thick sheet-ice with snowmobiles; limiting any prefabrication to hull-sized units weighing less than 6 tonnes. The shear effort to transport anything there means that the effective price of fuel is 10 to 20 times the final usage. Structural elements aside, the Antarctic provides no subsistence to people, which means that, with food, safe water, and heat, maintaining the life of a single person is an enormous expense; necessitating high efficiencies not only in operation, but in construction-crew as well.
These days the effort-balance between staying alive and doing science in the Antarctic has tipped.
Oh, and there’s only three months every year when the conditions are ‘mild’ enough to actually allow construction. For the other nine months the lack of sunlight and huge expanse of seasonal polar ice surrounding the continent make it impossible to access. On the mainland wind and snow are (perhaps obviously) a problem; the streams of wind running from the high, cold, ice to the ‘warm’ sea means that eddies form behind any obstruction, depositing phenomenal mounds of snow. The temperature differential also leads to glacial effects, which means the ‘ground’ beneath the station slowly flows out to the sea.
In the beginning people used pyramid tents to form a base for their science; and it’s interesting to note that these are still used for expeditions! For more ‘permanent’ structures, however, the next step was a traditional timber frame. The first of which, Scott’s hut, still exists; however this is only because its was built on a rocky outcrop in (relatively) mild conditions. In the 1970’s the trend was for structures that were designed to be buried by the snow; however eventually these got crushed under 20m+ flows or else were swept out to sea by the glacial movement. As a reaction, more recent structures have tended to elevated platforms that reduce the eddying effect, with the legs dug-out periodically. Large and inflexible, however, they didn’t address their inexorable movement out to sea.
Finding A Solution
By breaking the previous monoliths into a series of fully relocatable modules, the structure could now be periodically dragged back into place.
This really is a situation where the constraints almost define the brief. Whatever was proposed had to be easy deliver, build, operate and maintain; but it was just as important that it was a great science lab. These days the effort-balance between staying alive and doing science in the Antarctic has tipped; the challenge isn’t “how do we keep people alive in the Antarctic for a year”, but “how do we enable them to do the best science in remote conditions”. Peter’s team saw the importance in providing the best space possible, considering aspects such as comfort and personal interaction; discarding the traditional isolated corridors and eighties-sci-fi-esq “function-first” form, and throwing in a social module that looks nicer than any hotel I’ve every stayed in! Arguably it’s the first ever piece of polar architecture; making Halley VI for the best scientists- not just the best survivalists!
No one owns the Antarctic; and Halley VI is the product of the BAS, who Peter praised heavily for their pragmatic engagement. Starting out with a design competition, they identified the three designs they liked the most and then paid each team to to do a proper, costed, rigorously checked design (they even took them to the Antarctic!). At an estimate, this courageous exercise probably cost them £1m in fees alone, but without a doubt it was this procedure that led to them realising the best design.
AECOM’s concept focused on tackling one of the final vestiges of Antarctic construction challenges; the ice flow. By breaking the previous monoliths into a series of fully relocatable modules, the structure could now be periodically dragged back into place. They opted for a steel frame that exercised a little shell action to allow the modules to balance on two legs when over uneven ground. Utilisations were kept elastic, and given the small number of movements fatigue was not a consideration; brittle fracture was, however, and especially ductile steel was required to prevent it.
Design – Build – Operate
Halley VI obviously represents an un-codified design situation; and the huge penalties involved from failure meant that substantial testing was done during design. From making scale model units (as no one trusts CFD, it seems) to playing tug-of-war with Caterpillar’s to ascertain the maximum on-ice towing capacity (and thus module weight). The cladding, as an innovate use of FRP material, was of particular concern; with the first tests showing cracking in resin rich areas- leading to a simplification of the design shape. To reduce the risk they used proven train doors (they have trains in sub-Arctic Russia!) to link the pods, accepting the tapering shape created to accomodate them.
Antarctic exploration may well lead the design of structures in space, where agile, light-weight, modular approaches are also required.
Using Halley V as a base for construction, Halley VI was then towed to it’s semi-final location. The project took 3 years (9 months) of construction time, over four years (1 year was taken off to perfect the design); with everything being test-built prior to dispatch. Considerations such as ‘how do you build this in Antarctic PPE‘ led the design; focusing of modularity- with the a lot of the components simply ‘clipping-on’ and all wet-trades avoided. The design was completed under an NEC3 contract (with key risks, e.g. cladding, identified up-front and the burden shared), at an official cost of £26m. Considering the assets already employed by the BAS, however, it was probably closer to £50m.
Given the massive effort required to bring fuel to the Antarctic, a lot of the design focused on operational efficiency. Specialist FRP closed-cell insulated cladding was employed, with robust jointing to stop snow getting in (which, like sand, get’s everywhere and damages everything) as well as reducing heat wastage. Similarly water-waste services employed aeroplane-esq vacuum and treatment solutions that halved the per-person consumption and made it one of the first bases to leave only treated water behind. Learning from a tragic accident in a Brazilian station the design also allows half the base to fail (though fire) and still remain operational.
The attention to detail in optimising for effort can be clearly seen in the relocation routine. When the station needs to be moved back into place, the legs are dug out, the modules dropped to a lower position; braced, and given a shunt to break the ice. Then the modules are towed back from their drift. Where pods on six legs might be the more obvious from a stability prospective, four legs and temporary jacking towers were used instead. Why? Because testing and discussion showed that digging out each leg prior to moving was the critical ‘effort’ path.
Peter sees the final challenge as making these structures fully sustainable (in a much more literal sense than usual) with renewable energy. Currently, however, the technology is not reliable enough for these life critical situations; although “summer only” bases have achieved zero-emissions. More excitingly, to me, is the comparison of Antarctic exploration to engineering for space science-bases, where agile, light-weight, modular structures will also be required.
To the moon…