Information Management in Next Generation Anti-Submarine Warfare

The Future of Undersea Competition Topic Week

By Michael Glynn

The last decade has featured rapid advances in computing power, autonomous systems, data storage, and analytics. These tools are double-edged weapons, offering possible advantages to the U.S. while also opening the door to increased adversary capabilities. When combined with legacy systems and current doctrine, these technologies offer the U.S. Navy the chance to retain an advantage in the undersea contests of the future. The service must capitalize on these technologies. If they do not, they should realize that the low barrier of entry may drive potential opponents to do just that, eroding comparative advantage.

For the last 25 years, the Navy’s anti-submarine warfare (ASW) community has enjoyed the luxury of a permissive threat environment. What limited money was available to be spent on ASW was allocated to defensive measures to protect high value units in a close-in fight. The sensors and weapons that make up the stockpile are holdovers or incremental improvements of systems conceived in the late 1980s. The once dominant ASW task forces that tracked fleets of Soviet submarines have suffered from neglect, brain drain, and disuse in the last quarter century.

Despite these challenges, the U.S. currently retains a decisive advantage in the undersea domain. The service’s doctrine has been recently rewritten, and draws lessons from effective ASW campaigns of the past. Full-Spectrum ASW seeks to degrade the submarine threat as a whole.[i] It seeks to attack the adversary kill chain at every point, making damaging and sinking submarines only one piece of the ASW campaign.

Some observers have claimed that advancements in sensor systems and data analysis will strip stealth away from submarines.[ii][iii] This erosion of stealth will not happen unless the U.S. Navy solves three distinct challenges: gathering, analyzing, and disseminating environmental information, integrating operations analysis at the operational and tactical levels of war to maximize sensor and weapons effectiveness, and ensuring that ASW task forces are equipped with standardized equipment and highly effective training. Let’s discuss each of these challenges in detail.

Environmental Information

The ocean is an enormously complex and variable warfare domain. The properties of the ocean can change rapidly over small distances, just like weather ashore. Temperature, salinity, pressure variations, and the features of the ocean floor alter the way that sound energy moves through water. Characterizing the environment is critical to conducting effective ASW.

For decades, the Naval Meteorology and Oceanography Command (NMOC) has provided the service with oceanographic and bathymetric information. NMOC
a fleet of survey vessels, gliders, and sensors to gather information on the water-column.[iv] Computers ingest the information and build forecast ocean models.[v][vi]Operational planners and ASW operators use these products to model how sound energy will travel between their sensors and the submarine they are hunting. Without accurate ocean models, ASW operations are exercises in guesswork. Models are critical tools for effective ASW.

Paleobathymetry in the Southern Ocean. Photo: NOAA.

The Navy of tomorrow will need to make better use of the environmental data it collects and the models it produces. Many tactical platforms constantly collect data such as ambient noise or sound velocity profiles. Unfortunately, much of this raw data never makes it back to NMOC, due to communications limitations and process shortfalls. This hurts the quality of oceanographic models, and means the fleet will show up to the fight already at a disadvantage.

The undersea competition of the future will feature better dissemination and use of oceanographic models and bathymetric information. Ships and aircraft will automatically record environmental data and upload it to NMOC databases. When bandwidth makes it possible, ships, submarines, and aircraft should be constantly fed the most recent environmental model and use this information to drive radar and sonar performance predictions inside their combat systems. Fusion algorithms will automatically ingest real-time environmental measurements from sensors in the water to merge with the model and improve the accuracy of sonar performance predictions.

Operations Analysis

In the past, ASW planners have been able to degrade their adversary’s submarine force and maximize the effectiveness of a small number of ASW platforms by using operations analysis. In World War II, the British Submarine Tracking Room and U.S. ASW Operations Research Group used all-source intelligence to re-route convoys, assign aircraft to guard threatened ships, target submarine transit routes, and hunt down individual high-value submarines.

During the Cold War, the U.S. Navy used applied mathematics and computational modeling to predict the location of Soviet submarines. These search systems used track information of past patrols to build models of how Soviet commanders tended to operate. Cueing information was used to identify high probability search areas and recommended platform search plans.[vii] Real-time updates of positive and negative information during a search were fed to the computer to modify the search as it progressed.[viii] These computerized systems allowed planners to double their rate of successful searches compared to manual planning methods.[ix] Despite two decades of operational success, these planning systems were defunded and shut down after the collapse of the Soviet Union.

ASW forces of tomorrow will have to rediscover the value of operations analysis and apply these efforts at the operational and tactical levels. ASW task forces will be equipped with all-source intelligence fusion centers. Cueing information will flow from traditional means such as the Integrated Undersea Surveillance System, signals intelligence, and novel means assisted by big data analytics. Methods as unusual as monitoring the social media or Internet activity of adversary crew members and their families may provide indications that a submarine is getting underway.

A U.S. Navy P-8A Poseidon with Patrol Squadron 45, is at Clark Air Base, Philippines in support of Exercise Balikatan 2015, April 9. (U.S. Navy photo)
A U.S. Navy P-8A Poseidon with Patrol Squadron 45, is at Clark Air Base, Philippines in support of Exercise Balikatan 2015, April 9. Photo: U.S. Navy

Legacy computational search systems could only be run ashore due to the limits of processors of the day. Today’s hardware allows these systems to be run on a laptop. In the near future, tactical platforms will ingest cueing information and generate employment plans for themselves and assets nearby. A P-8A will generate optimized sonobuoy drop points, sonar dip points for two MH-60R’s flying nearby, and search plans for an ASW Continuous Trail Unmanned Vehicle and three unmanned underwater vehicles.[x][xi] The search plans and sensor points will automatically be broadcast via Link 16 and other future networks. The ability to direct multiple ASW platforms in today’s environment exceeds human capabilities, but tactical operations analysis systems will reverse this deficiency.

Optimized Task Force Training and Equipment

The final key to enabling next generation information management is revamping the equipment and training of the task forces who direct ASW at the Combatant Commander level. The increasing lethality of cruise missile armed submarines means focusing ASW planning at the Carrier Strike Group (CSG) level and fighting a close-in defensive battle is unacceptably risky. Future ASW campaigns will be won or lost at the theater level, with CSGs being only one piece of a multi-faceted approach. While 25 years of low budgets and disuse have blunted theater ASW (TASW) task forces, it is these commands that will direct the undersea battles of tomorrow.

Today, each TASW task force uses a hodgepodge of various systems and local information management procedures that have grown up to fit the unique challenges of the area. Lack of oversight means each task force uses its own training syllabus, communications procedures, and unique methods to maintain a common operating picture (COP). Despite this disunity, personnel are expected to flow from one task force to another in times of crisis and seamlessly master a system they have never trained with. This is not a recipe for success in an increasingly complicated information management environment.

The Navy should ensure each TASW task force is equipped with a standard suite of analysis and information management tools. The forces will adopt and master the Undersea Warfare Decision Support System and maintain a worldwide COP backed up at each task force. Standardized qualifications cards, methods for maintaining the COP, and disseminating information will allow personnel to rapidly surge and integrate with another task force. An open architecture construct will allow adjustments in managing relationships with regional allies, information release, and the unique nature of the adversary threat.

The aviation community uses the Naval Aviation Warfighting Development Center to develop and rigorously standardize tactics. The surface community has recognized that standardized employment and highly trained Weapons and Tactics Instructors are crucial for operating today’s exquisitely complex and capable weapon and sensor systems.[xii] The TASW community should adopt a similar focus on standardization of information management and search employment, just as their colleagues in the aviation and surface communities have. The Undersea Warfighting Development Center will take a much more central role in tactics development and employment standardization.


Operations analysis has proven itself a force multiplier in ASW. This will be critical as fleet size continues to shrink. In the information age, the problem is not too little ASW information, but rather how to properly ingest, analyze, and disseminate information. If the Navy capitalizes on the opportunities listed above, it will be well on its way to maintaining undersea superiority. If it does not, it should remain wary that the barrier for entry for other nations to build effective information management and operations analysis systems is low. The technology required is relatively cheap and has current commercial applications. There is extensive open source literature on the topic. Without having to contend with an entrenched defense bureaucracy and legacy programs of record that stifle innovation, these nations will certainly seek to rapidly capitalize on these concepts as a means to disrupt U.S. undersea superiority.

Lieutenant Glynn is an active-duty naval aviator. He most recently served as a member of the CNO’s Rapid Innovation Cell. The views expressed in this piece are entirely his own and do not represent the position of the Department of the Navy.

[i] William J. Toti, “The Hunt for Full-Spectrum ASW,” Proceedings, (June 2014),, (accessed May 22, 2016).

[ii] Bryan Clark, “The Emerging Era in Undersea Warfare,” (Washington, D.C.: Center for Strategic and Budgetary Analysis, January 22, 2015),, (accessed May 22, 2016).

[iii] James Holmes, “U.S. Navy’s Worst Nightmare: Submarines may no Longer be Stealthy,” The National Interest, (June 13, 2015),, (accessed May 22, 2016).

[iv] “Oceanographic Survey Ships – T-AGS,” (U.S. Navy, August 23, 2007),, (accessed May 23, 2016).

[v] “Naval Oceanographic Office Global Navy Coastal Ocean Model (NCOM),” (National Oceanographic and Atmospheric Administration),, (accessed May 22, 2016).

[vi] “Naval Oceanographic Office Global Hybrid Coordinate Ocean Model (HYCOM),” (National Oceanographic and Atmospheric Administration),, (accessed May 22, 2016).

[vii] Henry R Richardson, Lawrence D. Stone, W. Reynolds Monach, & Joseph Discenza, “Early Maritime Applications of Particle Filtering,” Proceedings of SPIE, Vol. 5204, 172-173.

[viii] Daniel H. Wagner, “Naval Tactical Decision Aids,” (Monterey: Naval Postgraduate School, September 1989), II-5.

[ix] J. R. Frost & L. D. Stone, “Review of Search Theory: Advances and Applications to Search and Rescue Decision Support,” (Washington, D.C.: U.S. Coast Guard, 2001), 3-4.

[x] “Anti-submarine Warfare (ASW) Continuous Trail Unmanned Vehicle (ACTUV),” (Arlington, VA: Defense Advanced Research Projects Agency),, (accessed May 22, 2016).

[xi] Michael Fabey, “ONR Seeks Long-Duration, Large-Diameter UUV’s,” Aviation Week, (October 29, 2012),, (accessed May 22, 2016).

[xii] Sam LaGrone, “Navy Stands up Development Center to Breed Elite Surface Warfare Officers,” USNI News, (June 9, 2015),, (Accessed May 22, 2016).

Featured Image: A P-8A Poseidon surveillance plane conducts flyovers above the Enterprise Carrier Strike Group on February 3, 2012. REUTERS/U.S. Navy/Mass Communication Specialist 3rd Class Daniel J. Meshel/Handout

2 thoughts on “Information Management in Next Generation Anti-Submarine Warfare”

  1. Good overview, but it seems that OA has always been a part of ASW since the days of destroyers and depth charges. I realize the article is about more than just the air component, but let’s look at that for a minute. The real problem of airborne ASW, specifically, is the conversion of tactical coordinator and navigator into fire controlman and radioman; the “people solution” to the massive amount of data has left only the sharpest TACCOs available (after they’re done poring over checklists) to really see the three-dimensional battlespace accurately and make good calls. (You really need an FC or FT and a radioman.)

    What I’m also curious about is the following: “…the increasing lethality of cruise missile armed submarines means focusing ASW planning at the Carrier Strike Group (CSG) level and fighting a close-in defensive battle is unacceptably risky. Future ASW campaigns will be won or lost at the theater level, with CSGs being only one piece of a multi-faceted approach.” Your experience with theater ASW probably involved a Wing – ever tried ASW without talking to a well-connected shore establishment? It’s possible – but the ship has to know what you know and what you can’t know. Not time for an OS3 to ask you to patrol between 500 and 1000nm ahead of force for x hours.

    I see where you’re going with theater-level, but could you agree that the CSG (or some variant thereof) is still the first target? Theater ASW independent of the CSG is important in a major war, but there’s a whole spectrum up to that point in which the CSG could play a role. The problem of standardization is already partially solved in the air wing – BT buoy to torpedo splash with no communication needed to land bases that might be smoking holes at that point anyway. Try ASW of any variety in a degraded long-range comms environment and the advantage of dispersed, standardized units becomes more clear – someone out there has downloaded the latest ocean data, and can talk you on to target.

    Why not solve both problems by treating the P-8 like an S-3 with longer legs – if there’s a CSG involved at all, make like you just took off from it, instead of from some major strategic land base. Actual launch and recovery? Plenty of long enough runways in the world, no need to have the theater ASW folks on-site. Information management is already practiced in the CVN, and line-of-sight is not a limitation – the “close-in defensive battle” is not the only role of a carrier or amphibious command ship. As a bonus, it can be replicated by any other ship if the CVN is incapacitated – not so with the theater ASW “office” on a contested land mass, which might be too busy putting out the (literal) fires to update you on eddies and layers.

    1. Ian, thank you for a thoughtful reply and pointed questions. We are in violent agreement of how central OA has been to ASW since it’s inception. Diving down the air ASW pipe is actually an excellent way to unpack where some of the shortfalls exist today. As early as the mid 1980’s, engineers were looking to integrate OA tools on airborne platforms to aid a TACCO (or TAC NAV for our Nimrod brethren) in the task you described. The Cold War ended before the utility of high speed computing could be brought to bear in making that battlespace more clear aboard aircraft. When P-7 and later Nimrod MRA4 died, their operations analysis suites died with them, and in the twenty years from 1991 to the present day, two generations of VS and VP aviators grew up not knowing these tools existed. P-8 had its integrated environmental analysis suite deleted during operational testing because testers couldn’t imagine how the capability to sense and model the environment inflight would be relevant. The sad truth is that today’s crews are operating against notional modern threats such as a Severodvinsk class SSGN with less capable OA tools than their parents used against a vastly less capable boat like a Yankee class SSBN. Requirements writers need to look at the recent past and study the successes of the integrated ASW force at its peak of operational effectiveness in the late 1980’s if they wish to write proper requirements and get the most out of their sensors.

      In discussing a focus on the CSG versus TASW, I wanted to drill in on two main points. The first being budgeting and programmatic priorities. Lack of funding for modernization of USN command and control programs for ASW has largely focused on building software to help address the unique challenges of defending a CSG. To the best of my knowledge, upgrades of the Undersea Warfare Decision Support System (USW-DSS) have gone first to the CSG’s and DESRON’s with less standard integration by TASW task forces. A CSG is a key part of an ASW campaign, but the coordination with joint commanders to enable attacks against the entire submarine kill chain will not be done at sea. A commander will require cyber and kinetic attacks against adversary ISR systems, mining from land-based bombers, and attacks against port facilities. Theater commanders will be more tightly tied to Combatant Commanders in this realm, so it makes sense to allocate the right amount of resources to sharpen these capabilities at the land-based task forces. CSG help defeat submarines of course, but breaking the back of the submarine threat will require action at a much higher level.

      As to the role of the CVN, I don’t agree with your premise that a CSG is naturally the first target in a submarine campaign. If I were an adversary submarine commander, I’d be just as happy to strike at vulnerable combat logistics ships in the rear than run a gauntlet of defenses to get in close to a CSG or other HVU. How long will the CVN be able to operate without resupply from MSC ships? With the current force structure, USN might be hard pressed to screen HVU’s and patrol areas, to say nothing of supply lanes in rear areas. I do agree, though, that the CSG makes for an extremely enticing target, not just for its military value, but also for the unique prestige and symbolic role that carriers hold as the embodiment of American power projection capabilities.

      As for utilizing the CVN itself as an information management hub, I certainly think there are advantages here. I wish we didn’t have to discuss a notional S-3, but with the airwing out of the mid-range ASW game, we have to solve problems as best we can. There would certainly be information exchange challenges. How do you see a CSG operating in EMCON being able to transmit data in a timely manner to other ASW platforms while evading adversary ISR assets?

      I think we are in agreement that with suitable systems to assist the human in sifting through data and using a computer to attack OA problems, much can be accomplished even in an airborne or shipboard environment. I think that a CVN or DDG or even a P-8 or MH-60R can accomplish quite a lot given sufficient computing resources and suitable analysis tools. The long pole in the tent, as you allude to, is training, so the proverbial OS3 can quickly recognize whether the course of action he or she tasks platforms with is reasonable or not. The right tools, with intuitive user interfaces are critical. Too often, we in the service buy the cheap and difficult to use systems. More communities should emulate the rapid prototyping initiatives that have been so successfully used by NAVSEA for submarine combat systems improvements.

      The vulnerability of land based facilities, especially forward ones, has to be addressed. This is why I wrote about standardizing a COP and having off-site (preferably CONUS based) backup so theater ballistic missiles or special operations direct action raids don’t completely dismantle any semblance of C2 on the first day of a conflict.

      Getting serious about solving the information management problem at the theater level and standardizing training will produce dividends for maritime patrol, rotary ASW, and the surface communities as well. If suitable OA systems can be built to solve theater level problems, they can be scaled to attack operational and tactical problems. What is needed is a combination of resource sponsorship to align funding and institutional pressure to rally the various communities together, rather than allowing each community to set their own agenda and focus in a parochial manner on their sensors and weapons while not taking a more integrated and holistic approach.

      I infer from your comments you have an aviation and possibly S-3 background. Care to offer your thoughts on how aviation assets can start to integrate with unmanned surface vehicles and unmanned underwater vehicles? Until now, submariners have largely had a monopoly on UUV’s. How do you see VP, HSM, and the DESRON utilizing unmanned assets in the not so distant future?

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.