Anti-submarine warfare: A scalable approach

The growing and increasingly-quiet submarine fleets of potential adversaries present Western navies with a resource-intensive challenge using traditional ASW approaches. This article examines a number of more cost-effective and scalable alternative methods that may allow navies to do more with less.

The prosecution of anti-submarine warfare (ASW) represents one of the most complex aspects of combat at sea. Quiet nuclear-powered submarines and diesel-electric boats present navies with elusive targets which require multiple assets to track a single contact. This is, however, an approach which is difficult to scale against multiple contacts. In theatres such as the Indo-Pacific where China’s People’s Liberation Army Navy (PLAN) fields 60 diesel electric and nuclear attack submarines, or Europe’s high North, where the Russian Northern fleet is regenerating itself after a post-Cold War nadir in strength, these challenges will be keenly felt.^[1] The challenge of mass will be compounded by another factor, namely that newer submarines fielded by likely Western adversaries are increasingly quiet. For example, the Russian Yasen/Yasen-M class submarine is comparable in quietness to the most modern western variants. Quietness, in turn, compels the allocation of even more assets to track an elusive contact over a wide area. The purpose of this article is to describe alternative approaches to ASW which are more scalable than current models.

The challenge with ASW

Presently, the process of tracking a submarine follows a series of well-defined steps. First, an initial contact is provided by an undersea hydrophone network like the US’ Integrated Undersea Surveillance System (IUSS), formerly known as the Sound Surveillance System (SOSUS). Although these networks are occasionally capable of generating relatively high-fidelity returns (at the peak of its performance SOSUS was able to pinpoint submarines to within a 97 km (60 mi) radius) they primarily serve to cue other means of detection.^[2] When a fixed array generates a contact, a maritime patrol aircraft will typically be dispatched in order to localise a target. This can be achieved either with sonobuoys or with sensors such as the aircraft’s on bord magnetic anomaly detector (MAD), though the latter only operates over short distances.^[3] An aircraft might attempt to prosecute a target with its own on-board lightweight torpedoes such as the US’ Mk 54, or the contact may be passed to another tracking platform such as a frigate or submarine.

The challenge that this model entails is twofold. First, each contact ties up multiple tracking assets in a way that makes scaling difficult. Take maritime patrol aircraft (MPA), for example. To provide continuous coverage at the GIUK gap, roughly 12-14 P-8 MPA must be kept on station.^[4] Between them, the UK and Norway field 14 P-8s, and it should be presumed that not all aircraft will be at readiness at all times. Furthermore, this approach is inherently sensitive to disruption – small losses due to, for example, attacks on airfields with missiles, or aircraft being downed by ship-based surface-to-air missile (SAM) systems, can have a disproportionate impact.

Secondly, the approach described depends on assets which can patrol freely, such as MPA. Yet if a submarine threat needs to be contained near contested airspace, this becomes inviable. This is certainly the case in the Indo-Pacific, where the containment of Chinese submarines at chokepoints such as the Bashi Channel would require US and Allied MPA to operate in airspace which the People’s Liberation Army Air Force (PLAAF) may very well dominate in the opening stages of a conflict.^[5] In Europe, similarly, the ability to contain Russian nuclear-powered guided missile submarines (SSGNs), such as the Yasen-M class at the Bear Island–Svalbard gap will become increasingly critical if these submarines are to be prevented from launching cruise missiles at key military and civil installations in northern Europe. This, in turn, requires both aircraft and surface vessels to be placed close to Russian bastions, where a range of threats from the Tu-22M3 backfire to the SSGNs’ own missiles will likely pose a challenge. This is not to say that vessels cannot operate under these conditions, but they will likely incur losses in a context where a limited number of ship losses (for example) can sink a resource-intensive ASW concept.

There are a number of things that navies can do in order to prosecute submarine contacts in a more cost-effective way. Many of these will, however, require considerable departures from both past practice and the metrics by which success is deemed to have been achieved.

The importance of an active regime

The task of ASW has often been likened to a cat-and-mouse game between surface vessels built for quietness and submarines, with the risks of emitting high for both.^[6] In certain respects, this risk is greater under contemporary circumstances, given that submarines equipped with heavyweight torpedoes and cruise missiles often outrange the surface vessels hunting them.

Despite this, however, an active detection regime may be unavoidable if the number of vessels needed to cover an area is to be reduced. Modelling conducted by the author on the agent-based modelling software NETLOGO suggests that in an area such as the Barents Sea, frigates operating low-frequency active (LFA) sonar on an active basis can systematically defeat the Russian Northern Fleet even with minimal MPA coverage – albeit at the cost of a much higher attrition rate than when operating sonar only in passive mode. For many Alliances such as NATO, frigates are not a capability in short supply (although not all are equipped with LFA sonar), and to some extent the choice may simply be one of accepting a higher loss rate in the pursuit of an operational end.

Improvements in processing power are making it increasingly viable to employ LFA sonar, which has relatively low propagation losses, enabling detection of submarines with accuracy over long distances. The effectiveness of LFA sonar can be enhanced if operated as part of a multistatic array with uncrewed systems. For example, an uncrewed surface vehicle (USV) equipped with a low-frequency passive towed array receiver, and deployed 50 km ahead of a frigate equipped with an LFA, would effectively double the range of the sonar (in this case, in the forward direction relative to the USV), by reducing the transmission losses of the sound wave, when compared to mounting both transmitter and receiver on the same vessel. Moreover, the more nodes there are in a multistatic system, the more difficult it is to track the emitter, which in turn makes going active somewhat safer.^[7]

The uncrewed systems which can be equipped with towed array sensors need not be particularly costly and in many instances commercial-grade systems can serve this role. For example, at least one Chinese firm has advertised the conversion of an uncrewed underwater vehicle (UUV) originally built for oil and gas sector surveillance to the PLAN.^[8]

In principle, USVs can also be used as active emitters, but this imposes size and cost requirements given the power consumption of active sonar (with the detection of a submarine type target at 10 km requiring 500,000 W of power, for example).^[9] Here, the use of less-bespoke maritime platforms may make more sense. The containerisation of ASW solutions, such as towed array processing modules can just as easily be applied to active solutions, and can allow for auxiliary vessels taken up from the civilian market to support these functions as well.^[10] To be sure, such vessels would be highly vulnerable and the requirements of networking them with a dedicated naval asset are considerable. However, a submarine crew would also have to consider the risks inherent to sinking a relatively cheap ship at the cost of creating a flaming datum – the last known location of the submarine – for more dedicated ASW assets to prosecute. It should also be noted that the integration of containerised systems is never a simple systems engineering task, with requirements for cooling and integration with a vessel’s power plant often representing a highly complex activity. Navies would thus likely have to procure and adapt any auxiliaries they intend to use in this capacity in peacetime. The core point, however, is that the quiet ASW frigate, while still the most irreplaceable part of a network, need not be its only surface-based component.

Notably, standoff sensors are most useful if the data they generate can be usefully employed quickly. Despite this, many navies have not invested in long-range anti-submarine rockets (ASROC), with key exceptions including China’s PLAN, which has fielded the 50 km range YU-8, and the JMSDF which has fielded the 30 km range Type 07. While it is unsurprising that this has not previously been a priority given the short ranges at which a target can be classified as target, as localisation and classification becomes viable at longer ranges, the ability to exploit this will depend on standoff effectors much as it does in surface warfare.

SIGINT

The use of signals intelligence (SIGINT) to track submarine activity is not new. However, it may gain a new salience due to the growing importance of cruise missiles to submarine activity. Very few modern submarines are incapable of at launching cruise missiles, and in some navies (such as the Russian Navy) older Project 971 Shchuka-B (NATO reporting name: Akula) class submarines are being refitted to carry missiles such as the 3M-54 Kalibr (the anti-ship variant of the Kalibr cruise missile family).^[11]

Tactically, however, this introduced a greater requirement for communication with offboard sources. Submarines can identify the range and bearing of moving targets from long distances, but for target-grade data they require offboard cueing. During the Cold War, for example, the USSR’s Project 949/949A Granit/Antey (NATO reporting name: Oscar) class nuclear attack submarines (SSNs) were to be cued by the Soviet Tselina electronic intelligence (ELINT) satellites. Land attack missions, while simpler, also require coordination with other missile launchers and in some cases the communication of mission programming data for missiles (though this may be pre-loaded before a submarine departs). This need for offboard communications, coupled with the increasing convergence of SSNs and SSGNs therefore creates new opportunities for ELINT gathering.

In this context, platforms such as medium-altitude long endurance (MALE) uncrewed aerial vehicles (UAVs) can become increasingly useful as a means of gathering data regarding emissions. While the use of UAVs as substitutes for MPA has been mooted, their employment in a SIGINT role is more mature as illustrated by use cases in the land domain.^[12] Ground- and space-based SIGINT platforms can also provide wide area surveillance, as illustrated by the ongoing conflict in Ukraine, where the US XVIII Division was able to provide the Ukrainians with a SIGINT and ELINT soak 32 times a day.^[13] As submarines increasingly become a deep strike capability, they will have many of the same dependencies and vulnerabilities as other strike platforms.

A whole-of-society approach

There are a number of sources of data within the ocean which states do not own. For example, the earliest adopters of UUVs have been the oil and gas sector – a point which became apparent to the Norwegian Navy after the severing of a key cable near Lofoten, which prompted it to leverage the 600 or so commercial UUVs operated by the oil and gas sector. Similarly, the Italian Navy recently signed a deal with that nation’s largest internet provider to leverage data from undersea cables which are sensitive to, among other things, changes in background pressure.

The data held by private sector actors is proprietary and understandably sensitive, however it is possible to manage these concerns through a number of avenues. One approach, for example, is applying the ‘publish-subscribe’ model to data fusion. This does not require the underlying source data from a system to be shared; the only information which is critical is the fundamental structure of the data, which allows for the creation of translation layers which can merge it into a shared format. Equally, it should be noted that governments have been willing to compel the private sector to share data in the service of counterterrorism, arguably a less-pressing concern than the states which pose credible submarine threats. As such, coercive options cannot be entirely foreclosed, although some companies which represent data sources can fall across multiple jurisdictions, and so would require the coordination of policy across Alliances and coalitions. More incentive-based approaches might be another avenue. In its recent effort to ensure that US company SubCOM was able to participate in a consortium building the SEA-ME-WE cable between Asia and Europe, the US government secured the company credit on favourable terms through the Export-Import Bank of the United States (EXIM).^[14] Favourable access to credit could also be used by nations to incentivise the sharing of data, and if a supranational organization (for instance, the EU) led this process, data secured could be disseminated more easily among members.

Suppression as an alternative to destruction

Many of the most cost effective means of disrupting a submarine’s activity may depend on suppression rather than destruction. Take, for example, many UAVs such as the MQ-9 and the UK’s Protector which have been considered for ASW roles.^[15] While it is an open question whether these craft can in the near term be equipped with the processing power needed to play a role comparable to dedicated MPA such as the P-8, their persistence is a virtue in of itself. For example, a UAV laying fields of decoy sonobuoys which mimicked the emissions of active sonobuoys, such as the US’ AN/SSQ-62 would, at a minimum, force more evasive activity on the part of submariners than would normally be the case. In this vein, a UAV sonobuoy laying capability was recently demonstrated by an MQ-9B SeaGuardian during tests by manufacturer GA-ASI in January 2025.

In a similar manner, minelaying represents a cheap means of complicating submarine activity even when it imposes limited attrition – though it should be noted that mines can be highly lethal at chokepoints, such as those in the Pacific first island chain. This is a major assumption for the PLAN, which intends to use its large fleet of maritime militia vessels to seed minefields near the first island chain as a low-cost means of complicating transit near the first island chain.^[16] More sophisticated mines such as the US’ Quickstrike series can be cued by the magnetic signature of a nearby boat, as well as by other sensors. It is noteworthy that in the Second World War, mines accounted for the majority of submarine kills on both sides, and even when submarines evaded them, their presence limited the time which a vessel could dedicate to useful activity.

No panacea, but many effective solutions

There are no straightforward ways of tracking submarines, which remain among the most complex targets in modern warfare. However, efforts to exploit advances in data processing as well as a more risk-acceptant active-sensor-centric approach to employing vessels can limit the challenge posed by submarines. In addition, it must be recognised that to defeat the threat posed by submarines, they need only be kept inactive or focused on evasion, rather than destroyed, even if the latter is optimal. A range of tools, many of which exist in the civilian sphere, can be better-leveraged to allow navies to achieve this.

Dr Sidharth Kaushal

 

[1] Aaron Tu. Report Casts Light on China’s Submarine Fleet. Taipei Times. Jan 08 2024. https://www.taipeitimes.com/News/taiwan/archives/2024/01/08/2003811792; Sidharth Kaushal et al. The Balance of Power Between NATO and Russia in the Arctic and High North. (London:RUSI, 2022)

[2] On SOSUS see Tom Stefanick. Strategic Antisubmarine Warfare p.39

[3] Tom Stefanick. Strategic Antisubmarine Warfare and Naval Strategy. (New York: New Free Press, 1987)

[4] Bryan Clark, Seth Cropsey, Timothy Walton. Sustaining the Undersea Advantage: Disrupting Anti-Submarine Warfare Using Autonomous Systems. (Washington DC: Hudson Institute, 2021)

[5] Sidharth Kaushal and Juliana Suess. The Impact of a Taiwan Strait Crisis on Deterrence in Europe. (RUSI:London, 2024)

[6] For example see George Allison. The Type 26 Frigate – where even the plumbing is quiet. UK Defence Journal. https://ukdefencejournal.org.uk/the-type-26-frigate-where-even-the-plumbing-is-quiet-2/

[7] Robert Been et al. Multistatic Sonar: A Road to A Maritime Network Enabled Capability. NATO.

[8] Ryan Fedasiuk et al. Harnessed Lightning: How the Chinese Military is Adopting Artificial Intelligence.(Washington DC:CSNET, 2021)

[9] Norman Polmar, ‘The US Navy: Sonars, Part 2’, USNI Proceedings (Vol. 107/9/943, September 1981), https://www.usni.org/magazines/proceedings/1981/september/us-navy-sonars-part-2, accessed 22 November 2024.

[10] Atlas Elektronik, https://www.atlas-elektronik.com/solutions/anti-submarine-systems/actas

[11] Sidharth Kaushal. Optimizing the Readiness of the UK’s Astute Fleet. RUSI. https://rusi.org/explore-our-research/publications/commentary/optimising-readiness-uk-astute-fleet

[12] Thomas Withington. Master and Servant.Armada. September 2022. https://www.armadainternational.com/2022/09/russian-drone-operations/

[13] Jack Watling, ‘Long-Range Precision Fires in the Russo-Ukrainian War’, in Dag Henriksen and Justin Bronk (eds), The Air War in Ukraine: The First Year of Conflict (New York, NY and Abingdon: Routledge, 2023), p. 74.

[14] Robert Clark. US Hardball Drove Chinese Vendor From Huge Project: Reports. https://www.lightreading.com/cable-technology/us-hardball-drove-chinese-vendor-from-huge-cable-project-reports

[15] General Atomics Show off MQ-9 Sea Guardian.https://ukdefencejournal.org.uk/general-atomics-show-off-enhanced-mq-9b-seaguardian/

[16] Sidharth Kaushal. Chinese Anti-Submarine Warfare Capabilities: A Blunt but Evolving Tool