I’ll admit to being surprised when Australia announced the termination of the deal with Naval Group to build the Attack-class submarines. Not because I thought that shouldn’t happen—I’m on the record as saying that it looked like a seriously risky project from the start—but because it’s so rare for Defence and governments to resist the sunk-cost fallacy. But if the Attack class couldn’t be made to work, then it would be scandalous to add to the estimated several billion dollars already spent on the program.

The Attack program timeline was also a concern, because the clock is ticking on the Collins class. Given the proposed level of complexity of the Attack class, further schedule overruns were likely. The Collins life-of-type extension—which is not without its own technical challenges—was proposed to buy time, but a disastrous capability gap between classes was a real prospect. There was a clear potential for a collapse in Australia’s submarine capability. We had already gone through that when transitioning from the Oberon to the Collins, and spent the next 15 years recovering.

Last week’s announcement does nothing to improve the prospective capability gap—and potentially makes it much worse. The design of a new class of nuclear attack submarine (SSN) isn’t an exercise that can be rushed, and there’s no established ability of Australian shipyards to build one, or of local industry to support one. No timeline has been specified as yet, other than the prime minister’s ‘hope’ that it will be before the end of the 2030s. But a delivery date somewhere on the other side of 2040 would seem more likely if we try to produce a bespoke Australian boat.

Given that the stated reason for the change of direction is our deteriorating strategic circumstances today, the disconnect is astonishing. If we’re serious about the role of a robust submarine capability in defending Australia’s interests, we need to find another way. I’ve had a couple of cracks at pitching a way ahead for the acquisition of the future submarines, with Sean Costello in 2009, and with Mark Thomson in 2014. Sean and I attracted some criticism for our ‘inflated’ estimate of a $36 billion price tag back then; now we look charmingly naive.

Those two papers—both titled ‘How to buy a submarine’—identified many of the challenges of risk identification and management in complex programs and pointed out the difficulties likely to ensue from the creation of a shipyard monopoly. Alas, the advice we proffered was trumped by the conspiracy of optimism that pervades the Department of Defence and by the politics of local shipyards.

In practice, Australia’s defence acquisition process has repeatedly shown itself to be very poor at recognising technical risk—see the technical and design issues surrounding the Hunter-class frigates and LAND 400 vehicles for the most recent examples. And successive governments have convinced themselves that supporting in-country naval shipbuilding makes it acceptable to slow down the delivery of defence capability, while paying more for it in the process. As a result, we have now lost a decade on the replacement for the Collins class and have the prospect of underemployed shipyards for the next several years.

But they say that the third time’s a charm, so let me suggest a couple of complementary approaches that might help sustain a national submarine capability and avoid the repeating of recent misadventures.

First, we need to accept now that the timelines mean that there’s no graceful transition from the Collins to a fleet of nuclear submarines. If Australia’s first SSN is launched in 2040, HMAS Collins will be 47 years old at the time, and already past a 10-year extension. Arriving at 2040 with few—or even no—operational submarines becomes a serious possibility.

So we are going to need a bridging capability. One option is leased nuclear submarines from the US or UK, though our dependence on the supplier for support and operation would reduce our ability to conduct independent operations. The other option is yet another acquisition, in the form of either an off-the-shelf boat or a Collins II design that draws on the engineering work done for the life-of-type extension. That would add more expense and consume more of Defence’s scarce engineering and project-management resources, but it would also allow the designing and building of the nuclear boats to be done in a measured fashion, and for the necessary Australian support infrastructure and skills base to be developed and matured.

Second, we need to take a hard-headed approach to the nuclear boats. We’re trying to become the first country in the world to operate SSNs without a domestic nuclear industry. That may be possible, but it suggests that we would be prudent to limit our ambitions for both innovation in design and industrial independence. It will make sense to maximise our leverage of the support capabilities of our UK and US partners in the enterprise and aim to acquire the most turn-key solution we can find, even if that results in less work for local industry.

If we learn nothing else from the Attack-class fiasco, we should accept that our collective ability to identify and manage project risk is poor. We’re about to embark on something that’s even more challenging in almost every respect and the risk of failure is high, so we should check our optimism at the door.

Prime Minister John Howard famously coined the term ‘barbecue stopper’ to refer to a political controversy so hot that it was likely to make backyard diners stop mid-shrimp-sizzle to debate the big issue of the day. If ever the specialist world of defence procurement has produced a national barbecue stopper, it would be over the question of Australia’s future submarine. Why are they so expensive? Why do we need 12 of them? Why build them here? Why not nuclear propulsion? Why a French design? Why not an American, German, Japanese or Swedish design? Aren’t submarines obsolete, to be replaced by drones? Won’t technology make the oceans transparent?

There are many questions and few, if any, easily accessible, plain-English explanations. A forthcoming ASPI special report attempts to answer the many questions that Australians pose when it comes to the design, acquisition, cost, operational service and strategic implications of submarines. Our writing team includes a vice admiral and former chief of navy, two rear admirals—one a distinguished submariner who has served with the Royal Navy and the Royal Australian Navy; the other, one of Australia’s leading naval historians. Other contributors include people steeped in strategy and capability development and with deep industry experience of defence production.

Our hope is that the volume will become the go-to guide for authoritative comment on all things to do with the present and future of Australian submarines. A draft of this work was provided to key individuals in Defence and from industry to advise on factual matters. As with all our publications, ASPI retains complete control of editorial matters.

Submarines: Your questions answered will be launched on 4 November. The following chapter by former admiral James Goldrick on why we need submarines provides a taste of what’s to come.

Peter Jennings

What, exactly, do submarines do?

Submarines are the apex predators of maritime conflict. They deploy a variety of highly lethal weapons—usually without warning. Submarines can lay mines around enemy ports and in shipping channels. They can sink surface ships with torpedoes or with anti-ship missiles. They can fire missiles against land targets. And they can insert and recover special forces for reconnaissance missions or small-scale raids ashore. They can also act as intelligence gatherers, sitting off an adversary’s coast to listen for enemy movements and monitor the electromagnetic spectrum for enemy signal and sensor transmissions.

Submarines operate alone but play important roles in other maritime operations. They can conduct reconnaissance before a surface force enters an area, using their own sensors to confirm that it’s clear of adversary units and safe for the intended operation. They can act as a barrier past which the enemy must go to attack its key targets. They can provide advance warning of enemy movements when stationed off the adversary’s coast—not only of enemy ships but also of aircraft. Properly timed, their own attacks can not only weaken or even halt enemy operations outright but force the diversion of resources away from the enemy’s offensive operations.

It’s their ability to operate covertly that allows submarines to do those things. Because submarines remain underwater and because the oceans are a hugely complex and dynamic environment, the boats are very difficult to detect and almost as difficult to track once detected. Modern submarines almost always sail submerged, surfacing only to enter harbour. They rarely transmit on radio and almost never use active radar or sonar, which might reveal their own position. They’re most effective when they can access intelligence and sensor information from other sources, allowing them to position themselves to locate and destroy their targets. This is particularly important for diesel–electric propelled boats.

Nuclear-powered submarines have practically unlimited high-speed endurance and can reposition very quickly. By comparison, diesel–electric boats are slow—although they can achieve high underwater speeds for short periods—and must recharge their batteries at intervals, even when they have other ‘air independent’ propulsion for low-speed patrols. If they’re to transit, they must use noisier diesel engines for much of the time. This means operating near the surface with a ‘snorkel’ or ‘snort’ mast above the water, making them more open to detection. Once on station, however, they’re much less vulnerable.

Although some emerging technologies may increase the probabilities of detection underwater, they have yet to threaten the ability of modern, stealthy submarines to operate covertly. Modern submarines are designed to have the smallest possible signature across the acoustic spectrum, to the point where they’re effectively ‘silent’ at slow speed. This is the reason the great powers all operate ballistic missile carrying submarines as key elements of their nuclear deterrent forces—and why new generations of those boats are in development. And although unmanned vehicles present a significant emerging threat, submarines’ ability to deploy and recover their own autonomous vehicles will give them new options at reduced risk to themselves. Their potential as underwater killers remains high.

James Goldrick

The Australian National Audit Office has just released its latest report on Australia’s $80 billion future submarine capability. It set itself the task of examining ‘the effectiveness of Defence’s administration of the Future Submarine Program to date’. The program’s supporters and detractors will both find some validation in this report.

We all are now calling an $80 billion spade an $80 billion spade, not a ‘greater than’ $50 billion spade or a $50 billion in-constant-dollars spade. Defence finally gave an $80 billion ‘out-turned’ number that takes real-world factors such as inflation into account at Senate estimates in November last year. That $80 billion figure is repeated in this report.

Defence says that this isn’t a cost increase but just a different way of expressing the same number. But once you add the future submarine’s $80 billion to the $35 billion future frigate and the $4 billion offshore patrol vessel, that increases the government’s $89 billion local shipbuilding program by $30 billion to $119 billion. $30 billion here, $30 billion there; pretty soon we’re talking real money.

One of the program’s key achievements in the past year has been the signing of the strategic partnering agreement between Defence and Naval Group to govern the design and construction of the submarine. It was originally meant to be signed in October 2017, but negotiations didn’t begin until the next month, and the government agreed to the negotiated outcome only in February 2019. The good news is that the ANAO reports that the government and Defence got everything they wanted, including intellectual property rights; transparency on costing; remedial measures and protections; and acceptable levels of Australian industry participation.

But there’s another narrative here. Many observers suggested that the lengthiness of the negotiations indicated fundamental cultural and commercial differences between the parties. The ANAO report notes that the Naval Shipbuilding Advisory Board, a panel of experts established to give the government independent advice on the shipbuilding program, had serious concerns.

Not only did the board recommend in September 2018 that Defence examine alternatives should the negotiations not succeed, but it also advised a year later that Defence should ‘consider if proceeding is in the national interest’ even if negotiations were successful. That’s serious stuff. The ANAO report doesn’t say what the advisory board’s concerns were. We can only hope that the final rounds of negotiations addressed them, particularly since the ANAO rightly assesses that the ‘success of the program is dependent on Defence establishing an effective long term partnership with Naval Group’. Since the government agreed to sign, it must have been confident that the issues were addressed.

The report also looks at aspects of the program’s schedule, which it’s been difficult to get a clear view of. There’s a high-level diagram in the 2017 Naval Shipbuilding Plan, and information trickles out, such as through Defence’s 9 May 2018 response to the estimates committee on foreign affairs, defence and trade. Moreover, terminology for key milestones has changed, which confuses things. I’d been hoping this report would provide the complete master schedule at different points in time so we could see which milestones had been delayed and where delays could be made up. That didn’t happen, but the report does provide some useful data points.

The report confirms, as foreshadowed at heated estimates hearings in October, that some key design milestones have slipped. The systems requirements review was originally meant to be completed by March 2019. After the design contract was signed, that milestone moved to October, and the review started only in December (and it’s not clear whether it’s been completed). Similarly, completion of the preliminary design review has moved from March 2020 to January 2021.

The report also reveals that Defence advised the government in February 2019 that Naval Group had proposed to extend the completion of the design phase by 15 months from July 2022 to September 2023. Ultimately, Defence agreed to a nine-month extension (to around March 2023). Construction is scheduled to start in 2022–23 (according to the high-level schedule provided to the Senate in May 2018 at least), so things are getting pretty tight if Defence wants to complete the design before starting construction. The ANAO hasn’t assessed the impact of the extension on the start of construction and the submarine’s entry into service, although at Senate estimates in November, Defence officials said that the 2022–23 construction date actually meant 2024 for the start of construction of the vessel itself.

Defence’s position is that getting the design right will save time and money in the long run by reducing rework. With construction scheduled to start in 2022–23 and sea trials scheduled to begin in 2031–32 (according to the high-level schedule provided to the Senate in May 2018), there’s potentially nine years to make up any lost ground. But, if the critical design review scheduled for June 2022 slips by the same amount as the earlier design reviews, we’re already eating into construction time.

The report also notes that Defence has advised that ‘a delay in the Future Submarine Program of more than three years will create a gap in Navy’s submarine capability’. What that means isn’t entirely clear. Based on my own analysis, I suspect that it’s saying that a three-year delay would result in the submarine force falling to fewer than six vessels, even if all six Collins boats go through a 10-year life-of-type extension to mitigate the risk of a capability gap. But what schedule baseline are those three years measured against? Have we already eaten into them? And with Defence now saying that the life-of-type extension program is seeking to replace virtually the entire Collins propulsion system (diesel generators, main motor and DC switchboard), that looks like a high-risk risk-mitigation proposition.

Overall, the report gives the impression that the program is being run according to solid project management principles and is putting integrity above expediency. It concludes encouragingly: ‘Defence has established the formal arrangements for the effective administration of the Future Submarine Program.’ But it also confirms that, under those arrangements, the first new submarine won’t be in service before 2034. That’s still a long way off.

When we expressed concern about the Royal Australian Navy’s plan to fit the first of its new Attack-class submarines with an outmoded lead–acid main battery system in over a decade’s time, former RAN submarine engineer Paul Greenfield rejected our key arguments.

There are a number of issues raised by Greenfield that we’d dispute. More importantly, it’s at the policy level that his critique should be rejected, and workable solutions should be developed to overcome the valid problems that he highlights.

Greenfield rejects the idea of fitting a light-metal main battery in the Attack class at this stage primarily because of his concerns about the danger to the crew of fire triggered by lithium-ion batteries. We believe those safety issues can be resolved.

He also raises the difficulty and risk involved in modifying a design already optimised for the older technology. Unfortunately, the speed of technological change precludes adherence to such a linear approach.

Four factors lead us to conclude that current planning is very likely to deliver a submarine that is obsolescent.

First, we aren’t discussing current levels of technology, practice and performance but those that are likely to exist or to be obtainable for conventional submarines 10 to 15 years from now.

Second, significant advances are now being planned by submarine builders around the world. While there are different levels of commitment and progress, this process is well established and is being funded to produce results.

Third, the technological advances for this transformation of submarine design aren’t dependent on military requirements or defence industry economics. They are being propelled by a transformation in the management of energy more broadly—and that research provides a range of options and alternatives to sustain progress in submarine development.

Finally, despite these likely advances, the intention is still to deliver HMAS Attack to the RAN towards the end of this period of transformation with completely outmoded and outperformed lead–acid batteries.

We acknowledge the absolute importance of safety, but we believe from the direction of technical development that the safety risks of light-metal batteries in submarines can be managed.

Historically, the effectiveness of lead–acid submarine batteries has been constrained by their limited performance spectrum. They frequently operate at the extremes of their capacity, suffering performance degradation through the process and hence reduced safety margins.

In contrast, the superior energy capacity and performance of light-metal batteries provide increased safety margins in all modes of operation. Naval Group regards lithium-ion battery technology as inherently safer than its lead–acid counterpart. This judgement is reinforced by South Korea’s decision, after 30 months of comprehensive study and evaluation, to use lithium-ion main batteries in the KSS-III batch 2 submarines. In East Asia, two designs of small, special-purpose submarines with light-metal power systems have recently been launched.

These vessels will be in the water for up to a decade before HMAS Attack, but data on their operation will be but a fraction of that gained from the increasing use of light-metal batteries in most areas of economic activity. Major automotive companies indicate that 50% of their models will be electric powered by 2030. Electricity grid stabilisation batteries are increasing rapidly in size and number.

Such developments, and the spread of domestic and small business storage, will provide both the operating experience and the market power to support investment in research and development that will accelerate the rate of evolution of battery technology significantly.

We have not attempted to second-guess the battery chemistry that might come to be preferred for submarine usage. The development of options will be driven by the demand of an expanding market. Automotive experience is relevant for the design of submarine power systems—the battery cells to be used for the KSS-III batch 2 submarines are those used in the latest version of the BMW i3 electric vehicle.

It seems, unfortunately, that HMAS Attack will be delivered with a lead–acid main battery. Designing the future submarine so that it incorporates nothing that hasn’t already been to sea in a submarine is one of the program’s primary risk-reduction strategies.

The already long development schedule for the first of class is driven by a determination to avoid what happened with HMAS Collins, whose construction began with detailed design still underway. These acquisitions strategies leave few options for innovation, even though completion of the design for the Attack is still some years away.

However, we agree with Greenfield that energy storage capacity and performance in Australian submarines must be improved. We believe that the best option for achieving this with the Attack class is to run a parallel design for a light-metal battery system to be fitted in the second of class—similar to the approach adopted by South Korea for the KSS-III batch 1 (lead–acid) and batch 2 (lithium-ion) submarines—with lead–acid and lithium-ion battery modules of the same weight fitted into the same overall volume.

Management and funding of this task need not be a responsibility of the submarine program, or even of the navy. There’s a joint-force requirement to understand the implications of emerging energy technologies if the Australian Defence Force is to remain operationally effective.

Deployed operations are enhanced by new systems requiring electrical energy. There will be an increasing need to recharge unmanned systems and for sufficient energy to sustain the ‘I’ in AI in a more complex and networked battle space.

Such developments seem ubiquitous and unstoppable, yet the ADF and Defence have no integrated mechanism to comprehend and respond to the energy demands of modern combat. A central agency—an ADF operational advanced energy centre—should be established to oversee the development of advanced energy capture and management, including identifying energy requirements and preparing options to solve problems such as the pathway for installing light-metal batteries in the Attack-class submarines.

Luke Gosling’s recent Strategist post, mooting Darwin as a forward operating location for Australia’s submarines, sparked a few thoughts.

First and foremost, it is refreshing to see Australian MPs sticking their necks out, thinking strategically and openly engaging in the defence debate, beyond the narrow confines of basing politics and ‘nimbyism’. Gosling’s proposal to make more of Darwin as a regional defence training and engagement hub is also worth considering.

Second, it serves as a reminder that Australia’s future submarine program is not simply behind schedule, with dim prospects for expanding the force beyond six boats until the late 2030s. It also suffers from a curious lack of attention to logistic enablers, such as submarine tenders and designated forward operating locations.

This is odd, given the prominence assigned by Canberra to the submarine program as the mainstay of the Australian Defence Force’s conventional deterrent capability.

However much the technology evolves, the distances that Australia’s conventionally powered submarines are required to travel from their home base, HMAS Stirling near Perth in Western Australia, to their patrol areas are unlikely to change. That is a drag on fuel and time on station, with no obvious prospect of being able to refuel or rearm in the theatre of operations. Fremantle is a long way to go to fill up and reload.

While Darwin offers a central, strategic location in the Top End, it’s not an ideal jumping-off point for submarines, for reasons that Gosling elucidates.

But Australia has other options that are arguably more suited for supporting forward submarine operations, from its eastern Indian Ocean territories, the Cocos (Keeling) Islands and Christmas Island. Christmas Island is better known as Australia’s front line against illegal migration, just 280 nautical miles from Jakarta. Such proximity imposes both political and operational constraints on its development as a military base. But the Cocos (Keeling) Islands, located 700 nautical miles from Indonesia’s capital, already play a significant enabling role in Australia’s defence, as a forward location for maritime surveillance aircraft. The enabling infrastructure on the islands, as foreshadowed in the 2016 defence white paper, is set for further expansion.

These far-flung Indian Ocean outposts are easily written off as liabilities. In How to defend Australia, Hugh White acknowledges ‘circumstances in which they would be critically valuable’, but concludes they are more trouble to defend than they’re worth. In testimony to parliament, Anthony Bergin and Sam Bateman have also highlighted obstacles to building up defence infrastructure on the islands, in terms of physical geography and challenging weather.

Still, I believe Australia’s Indian Ocean islands deserve serious attention as a force multiplier for Australia’s submarines. The Cocos (Keeling) Islands lie 2,000 nautical miles northwest of Perth, in a handy location for Australia’s northwest approaches or major straits that pass through the Indonesian archipelago. Christmas Island is even closer. As Australian territories, only the enemy and mother nature can deny their use.

Of course, the flip side to Cocos (Keeling)’s and Christmas’s value as prime strategic real estate is their vulnerability to blockade or direct attack. Yes, they would be much more vulnerable than Darwin, but still worth investment in hardened infrastructure, including fuel and weapons storage, as well as stable berthing locations. In peacetime or crisis conditions, the ability to sustain more of Australia’s submarines forward, for longer, could be very helpful indeed. Even one combat reload in theatre would be a significant multiplier for a small force in wartime.

The principle should be to make as many of the logistical enablers as mobile as possible, so that they can be flown in and flown out at short notice. Can Australia’s future submarines be designed in ways that assist resupply from such rugged, bare-bones locations?

We should not be too quick to discount the strategic value of Australia’s offshore Indian Ocean territories when they could help to stretch the legs and combat punch of the ADF’s most strategic and precious capability.

If history is a guide, the US submarine effort against Japan stumbled in its initial stages because it was too widely dispersed—between Hawaii, Brisbane and Perth—and because of a malfunctioning torpedo. Concentration matters, especially when the assets are so few.

But in a contemporary maritime conflict, Australia’s submarines are likely to require much greater intensity of effort in the opening stages, primarily directed at naval forces, instead of an attritional campaign against an adversary’s economic supply lines.

The pitiably small number of submarines that are likely to be available to Australia over the next two decades means that we must be creative about maximising the limited capacity available. Technology will surely evolve, sometimes in surprising ways. But the distances remain constant. This mandates more attention to logistical enablers, leveraging our outlying strategic geography, to enhance the ability of our submarines to operate and to fight forward.

Admittedly, this is an unabashed armchair admiral’s thought bubble. Others more in the know will spot flaws obvious in the argument. But that’s okay; at least the debate will have been joined. Well done to Luke Gosling for opening the valves.