Main Image Credit Atomic Weapons Establishment Burghfield site. Courtesy of Tony Austin.
The country’s nuclear projects require more political attention and more parliamentary accountability.
The recent National Audit Office (NAO) report on the management of infrastructure projects at nuclear-related sites makes grim reading. It provides a set of three detailed case studies to complement its 2018 ‘Landscape Review’ of the Defence Nuclear Enterprise, drawing a bead on the three MOD projects currently in the construction phase at nuclear regulated sites:
- a £240m submarine build facility at the BAE Systems dockyard at Barrow-in-Furness;
- a £474m production facility for submarine nuclear reactor cores at Rolls Royce in Raynesway;
- and a £1,806m facility at Aldermaston, known as MENSA, for the assembly and disassembly of current and future nuclear warheads.
All are delayed and overspent, but MENSA is the worst offender of the three, with schedule and cost overruns at around 100% and 148% respectively.
This facility is a critical element of the UK’s nuclear programme, designed to replace the UK’s elderly ‘Gravel Gerties’ – the bunkers designed to provide containment during nuclear warhead assembly and disassembly processes – at Atomic Weapons Establishment (AWE) Burghfield. When MENSA was conceived, these facilities were due to be decommissioned between 2015 and 2021 with the in-service date for MENSA set for 2017. The estimated in-service date is now set at 2023. The MOD and the AWE are therefore now reliant on a warhead assembly facility operating on life support: the Office of Nuclear Regulation – the independent regulator for the sector – has only certified its safety for operations for a short period, subject to ‘significant upgrades’ to those facilities and enhanced regulatory monitoring arrangements. A worthy target for NAO scrutiny, to be sure, but at least two other significant examples from the warhead programme display similar characteristics and could point to a serious underlying challenge to the long-term future of the UK’s nuclear deterrent.
The first of these cases is AWE’s Pegasus project, which deals with the construction of a new facility to handle and produce enriched uranium components for the UK’s nuclear warheads, essentially supplying inputs to the current and future assembly/disassembly facilities. The NAO report does not examine this project in detail, for example, noting only that the cases on which it is based are those currently in the construction phase, and that Pegasus has not yet reached that point. In fact, this £634 million project should be well beyond the construction phase by now with the MOD’s original estimates and planning application indicating that it should have been operating since 2016. But Pegasus was suspended with no obvious output yet realised as of 2018, with the MOD placing the ‘requirement under consideration’.
The planned Pegasus in-service date of 2016 is important because regulators in 2008 indicated that the existing enriched uranium facility, known as A45, would require replacement or satisfactory improvement works by that point if operations were to continue to be licensed. The previous year had seen the MOD conclude that refurbishment would not deliver a satisfactory future enriched uranium capability, leaving replacement as the only safe and effective option. Since then, the risks of overreliance on ageing facilities have been amply demonstrated: operations at A45 were suspended in 2012 following the discovery of corrosion in the building’s steel structure, and only restarted in 2015 once safety critical works had been completed. The suspension of this project in 2018 therefore leaves the future of the UK’s ability to produce enriched uranium components for current and future warheads uncertain, and there is no information yet in the public domain as to how much public money of the £634 million budget will ultimately be written off.
A second candidate for consideration could have been the project to deliver a new hydrodynamics facility. Hydrodynamic experiments are conducted to understand the behaviour of solid materials in the warhead during detonation, when they are subject to implosive forces so great that they behave as fluids. They are an essential component of the UK’s ability to underwrite the existing Trident Holbrook warhead and to design a future warhead should one be required, given that nuclear weapons testing is no longer permitted by the Comprehensive Nuclear Test Ban Treaty that the UK has been bound by since 1996. In 2002, AWE and MOD Chief Scientists, along with another senior MOD official, wrote that AWE’s capabilities at that time were ‘not capable of providing data of an accuracy sufficient to meet future programme needs’, and that a new facility was being planned.
This new facility, under project Hydrus, was developed until 2010, at which point work was terminated shortly after receiving planning permission, and costs of nearly £120 million were written off. Development of hydrodynamics capability continued instead under the UK-France TEUTATES programme enabled by the UK-France Treaty of 2010, with a ‘Technology Development Centre’ at Aldermaston in the UK and the main hydrodynamics facility, ‘TEUTATES EPURE’ at Valduc in France. Although we cannot evaluate whether or not the planned cost savings arising from this collaboration are likely to be realised – we know only that the costs have grown from a base that the government refuses to acknowledge – it is more straightforward to look at when the necessary capability will be available. Whereas Hydrus envisaged an enhanced hydrodynamic capability in service between 2010 and 2015, an equivalent capability is envisaged under this treaty to emerge only by 2022, a schedule that MOD last year assessed as ‘deliverable but challenging’. That is a long and possibly lengthening delay for data previously judged to be so critical to the warhead programme, especially given that the government needs to make an informed decision on the UK’s future nuclear warhead by the same point.
In each of these cases, although the costs are eye-watering, it is the schedule that is the critical issue. After all, it is the ability to meet schedule and performance thresholds that will dictate whether the UK can maintain the posture of Continuous At-Sea Deterrence (CASD) that Parliament voted to support in 2016, and whether it can maintain the level of warhead capability it has previously deemed necessary when considering alternative options. If either slip too far then clearly CASD as we currently know it would be at risk – perhaps from a safety-related shutdown to an ageing programme-critical facility, or from a lack of data or components to design, underwrite and produce new warheads, or from an inability to produce a system capable of meeting the high performance standards that the UK has set for itself.
For those who follow these issues, none of the NAO’s detailed and diligently-constructed report will have been surprising, least of all to the NAO itself. As early as 2008 it noted five key risk factors for the nuclear deterrent programme as a whole timetable; design decision; governance; financial control; and procurement practice. Over ten years later, this report evidences the realisation of each one of those risks in three case studies of a broader nuclear enterprise that the NAO two years ago indicated continued to exhibit many similar failings. And the examples above show that this latest NAO report has just scratched the surface of the warhead programme.
The NAO does not spell out solutions to the common issues it identifies, though it does note several common features – for example the challenges in acquiring sufficient personnel with suitable nuclear skills, a lack of sufficient regulatory engagement and costly requirement changes once projects were well underway – to all the projects, as well as a general failure to learn from relevant past examples. It also makes no clear assessment of whether the scope changes that lead to cost growth have resulted in additional capability for the money spent, or whether this too has diminished. In the case of MENSA, for example, it notes only that a requirements review between 2015 and 2017 resulted in a ‘reduced scope’, though this appears to be against the revised requirement set rather than the original and is in any case far from confirmatory either way. This also implies, but does not confirm, that premature commitment to cost and time targets before full project specification and de-risking – as was the case for the Queen Elizabeth class carriers – may be more to blame than inefficiency in subsequent management and delivery.
Whatever the answers to these questions, the UK’s nuclear programme clearly requires more urgent political attention than it has hitherto received, as key decision points for the programme approach and cost and schedule overruns begin to bite in earnest. As it would make sense for MPs to investigate the matter more closely, and to require greater transparency from the government and from the contractors that deliver key elements of the programme. The enhanced secrecy surrounding weapons development, production and maintenance hinders effective oversight and means that industry-wide issues with cost estimation in major infrastructure programmes are exacerbated.
A system of regular and increased accountability could include the establishment of a subcommittee to the House of Commons Defence Select Committee, tasked with oversight of the programme for the critical period we are now in. Existing independent review bodies, particularly the Nuclear Research Advisory Council, could also be given more resources, more teeth, and a reporting line outside the Ministry of Defence – which could link in to the Cabinet Office, 10 Downing Street or even to this new subcommittee in Parliament – to allow them to more effectively provide a challenge as well as advice in circumstances where they feel it necessary. These bodies could ensure that the key decisions that will be made in coming months and years are properly scrutinised. And, by providing mechanisms to raise issues that neither the contractor nor the MOD customer would otherwise have an interest in raising, could help to drive better performance and value for money from an enterprise that sorely needs it.
Tom Plant is Director of RUSI’s Proliferation and Nuclear Policy programme.
The views expressed in this Commentary are the author’s, and do not represent those of RUSI or any other institution.
Senior Associate Fellow