What the United States Military Can Learn from the Nagorno-Karabakh War

By Nicole Thomas, LTC Matt Jamison, CAPT(P) Kendall Gomber, and Derek Walton

The views expressed in this article are those of the author and do not reflect the official policy or

position of the U.S. Army, the Department of Defense, or the U.S. Government.

Background

On September 27, 2020, intense fighting erupted between Azerbaijan and Armenia when the Azeri military went on the offensive. The Azeris’ objective was to recapture the territories lost to Armenia in 1994. But to understand the underlying reasons for the current conflict, one must look back to the root of hostilities and to the role of other powers in the region.

In the 1920’s the Soviet Union established the Nagorno-Karabakh as an autonomous region within the borders of Azerbaijan. Though the land was within Azerbaijan, it was home to nearly 95% ethnic Armenians.[1] The region remained relatively stable until the collapse of the Soviet Union, which opened the door for the inhabitants of the enclave to declare independence. This move resulted in a war between Azerbaijan and Armenia with Russia initially playing a role by providing weaponry and mercenaries to both sides.[2] Russia ultimately brokered a 1994 cease-fire to reduce instability in the region. This cease-fire allowed the Nagorno-Karabakh to achieve military and political independence despite being constrained within Azerbaijan’s borders. But after two decades of little movement toward reconciliation, Azerbaijan and Armenia would remain in a battle over the region, killing more than 20,000 people, displacing millions, and solidifying the ethnic Armenians’ hold within Azerbaijan.
 

Despite the 1994 cease-fire agreement, there have been 7,000 breaches. But it would be the April 2016, “Four Day War” along the “line of contact” that would mark one of the region’s deadliest. After decades as a frozen conflict, the “Four Day War” demonstrated to the Azeris that their strategic objectives could be achieved by force. And it would be this lesson, along with new military capability and a powerful ally, that would set the conditions for the September 2020 conflict.

This latest war lasted 44 days and left Azerbaijan in control of nearly one-third of the territory in the Nagorno-Karabakh. Unlike previous skirmishes and cease-fire violations, the warfare that erupted in September 2020 included post-modern characteristics and multi-domain combat operations. At only six days into the conflict, Azerbaijan already claimed to have destroyed 250 armored vehicles, a similar number of artillery pieces, and 39 air-defense systems, including a Russian S-300 surface-to-air missile system.[3] Armenian forces faced a persistent threat of unmanned aircraft systems (UAS) that could attrite traditional defenses and minimize their overall defensive capability.
 

One Nation, Two States: The Impact of the Turkey-Azerbaijan Relationship

A long, but simmering, territorial dispute set the conditions for the September 27 offensive, but it was the outsized role Turkey played in Azerbaijan that would ultimately tip the scales in the Nagorno-Karabakh conflict. Independently, cultural similarities, military alliances, or economic interests can create a strong bond between countries, but in the case of Turkey and Azerbaijan, these two countries share all three—making it one the most strategic partnerships in the region.
 

Underpinned by a shared language and culture, the sizeable Azeri diaspora within Turkey creates the sense that they are “one nation, two states.”[4] During the early years of Azerbaijan’s independence from the Soviet Union, the country’s foreign policy pursued a pan-Turkism agenda.[5] This ideological approach strengthened the connection between the two countries and ushered in a new phase in Azerbaijan’s political, economic, and military future.[6]

Azerbaijan’s new independence created the need for it to have allies in the region early in its independence, which would do two very important things. First, it would enter into a 1992 agreement with Turkey that provided military aid, training, and participation in joint exercises; and second, in 1999, the two nations would further cement the Ankara-Baku alliance with the development of shared economic goals. Together, Turkey, Azerbaijan, and Georgia established the Baku-Tbilisi-Ceyhan (BTC) oil pipeline, which connects the Caspian Sea to the Mediterranean Sea. This economic alliance provided Azerbaijan with an economic advantage over its neighbors as its per capita income reached its peak of $7,190 in 2011, compared to Armenia’s $3,526 and Georgia’s $4,022.[7] The new revenue stream enabled Azerbaijan to increase investment in its military. Between 2006 and 2019, Azerbaijan invested nearly $29 billion in its military compared to Armenia’s roughly $6 billion during that same period.[8] In the end, it would be a combination of Azerbaijan’s alliance with Turkey and its nearly fifteen years of economic growth as a result of the BTC pipeline, that would give it an advantage vis-à-vis Armenia by rapidly improving military training, readiness, and capability.

Systems Employed

Azerbaijan leveraged the economic windfall to field several different types of UAS in the conflict. Among the deadliest and most effective was the Turkish Bayraktar TB2 which carries four MAM (Smart Micro Munition) laser-guided missiles.[9] The Azeris developed an imposing UAS arsenal composed of Israeli loitering munitions, also known as “suicide” or “kamikaze” drones that included the Harop and SkyStriker. They also deployed a locally-made version of the Israeli Orbiter-1K small kamikaze drone and converted a number of their old Russian AN-2 biplanes into ISR or suicide UAS. By contrast, Armenia’s UAS fleet consisted of smaller, indigenous systems focused on reconnaissance missions and is generally recognized as less capable than Azerbaijan’s fleet of foreign UAS.

Tactics

The armed conflict between Azerbaijan and Armenia highlighted the continued use and effectiveness of unmanned platforms in low-intensity conflict and its ability to transform smaller, less-funded militaries into more lethal warfighting organizations. The use of UAS, particularly by Azerbaijan, included a range of intelligence, surveillance, and reconnaissance (ISR) operations, as well as unmanned aerial attack operations involving a variety of different platforms and munitions. The resounding success of UAS in the Nagorno-Karabakh War marks what many consider to be a turning point in modern warfare. For the first time in recorded history, nearly all battle damage was inflicted by unmanned platforms. The attrition of forces and equipment by UAS led to a decisive Azeri victory.

Suppression of Enemy Air Defenses (SEAD)

At the onset of the conflict, Azerbaijan leveraged Soviet-era AN-2 biplanes to deceive and expose Armenian air defenses. Though decades old and intended to serve as traditional manned aircraft, the biplanes’ conversion to unmanned decoys allowed Azerbaijan to conduct low altitude flights into the highly contested environment—and more importantly—into the weapons engagement zone (WEZ) of Armenian air defenses. These improvised UAS were repurposed as decoys and flown to the front lines to force air defenses to give away their location and enable targeting by TB2s.[10] When the Armenian air defenses targeted, engaged, and destroyed the perceived threats, they inadvertently broadcasted their positions to Azeri unmanned aerial attack platforms that flew at higher altitudes—enabling the Bayraktar TB2 and kamikaze drones to destroy higher-payoff targets like the Armenian air defense systems.
 

These tactics are reminiscent of Vietnam-era “Wild Weasel” or “Hunter-Killer” concepts, where a bait aircraft would fly at low altitude in an attempt to gain enemy contact or draw fire, and a separate trail aircraft maintained enough lateral and/or vertical separation to immediately engage enemy forces that exposed themselves. By leveraging these tactics, techniques, and procedures (TTPs) with an unmanned fleet, Azerbaijan was able to destroy the vast majority of Armenian air and missile defense equipment and establish tactical air superiority with minimal risk to force or mission. It is worth noting that traditional rotary wing assets were not used during these attacks. The high density of ADA systems across the battlefield presented too great a risk for more expensive manned aviation assets.

Shaan Shaikh and Wes Rumbaugh of the Center for Strategic and International Studies (CSIS) provided detailed analysis of how and why Armenian air defenses failed to counter Azeri UAS:

The bulk of Armenia’s air defenses consisted of obsolete Soviet-era systems, like the 2K11 Krug, 9K33 Osa, 2K12 Kub, and 9K35 Strela-10. TB2s flew too high for these systems to intercept even if they were able to detect these relatively small aircraft. Russian-supplied Polye-21 electronic warfare systems disrupted Azerbaijani drone operations but only for four days. Armenia’s Buk and Tor-M2KM air defenses likely downed a few drones, but they were deployed late in the conflict, limited in number, and vulnerable to attack themselves. Armenia’s larger air defenses like the S-300 are not designed for counter-UAV missions and were targeted early in the conflict by Azerbaijani loitering munitions.[11]

In addition to the apparent lack of Armenian counter-UAS (C-UAS) capability, the strikes clearly demonstrated the vulnerability of more advanced air defense systems when faced with an overwhelming UAS threat.

Fire and Maneuver

A number of reports indicate that Azeri special operations teams, also referred to as “saboteur groups” by both Baku and Yerevan, infiltrated Armenian territory and occupied vacant houses days before combat operations began. Ethnic Armenians in the local area verified these reports and highlighted that “strange men, not Armenians” had established a presence in the town.[12] After initial UAS strikes decimated Armenian positions and opened gaps in defensive lines, the small groups of Azeri operators were able to seize key terrain with minimal resistance. With the use of UAS, the Azeri saboteur groups were then able to call-for-fire, directing accurate rocket, artillery, and air-to-ground fire onto designated targets. It is currently unknown whether these saboteur groups leveraged any type of manned-unmanned-teaming platform (i.e., a One System Remote Video Terminal equivalent) to receive live video, identify targets, or conduct battle damage assessments. These tactics demonstrated a variety of similarities to NATO operations in Afghanistan, where U.S. Special Operations Forces use unmanned platforms and laser range finder/designators to direct laser-guided and precision munitions onto targets, or sparkle targets to aid in directing unguided munitions onto target.

In the highly mountainous terrain of Nagorno-Karabakh, where movement by dismounted troops is slow, cumbersome, and restrictive for large military equipment like tanks, UAS serve as an equalizer that limits the advantage provided by elevated terrain and the cover and concealment it offers. The inhospitality of the terrain is amplified when small areas of low ground within the rising terrain provide little vegetation for concealment. Outposts and fighting positions in mountainous terrain can be identified and destroyed by UAS outfitted with modern sensor payloads and organic weapons. This is particularly applicable to fighting positions without appropriate passive defense measures (i.e., camouflaging, target hardening, etc.). When UAS do not have organic munitions and another UAS or manned platform is unable to support a remote engagement, UAS can transmit highly accurate grid coordinates to artillery or multiple launch rocket systems, enabling immediate “fire-for-effect” capability that yields accuracy to within ten meters of the intended point of impact.

Non-Materiel Failures

Video footage demonstrated that forces from both sides lacked training and proficiency in the application of passive defense techniques. Forces were regularly observed operating in the open, remaining static or moving slowly, poorly camouflaged, and aggregating together in tight groups.[13] Robert Bateman, writing for Foreign Policy, asserts that while UAS are a force multiplier, many of the failures witnessed during the war can be attributed to poor training. “Neither [side] seems to have grasped the idea that even the most high-tech tank (or armored fighting vehicle) is only so much scrap metal if you do not have a trained and disciplined fighting force inside those vehicles.”[14] Whether a lack of training or tactical discipline/patience, the cost to Armenia was immense. Video also shows Azeri UAS processing multiple high-payoff and high-value targets and immediately engaging targets of opportunity (i.e., troops in the open and unarmored vehicles) after high-priority targets were destroyed. In an ammunition-constrained environment, attack platforms would traditionally pass on engaging small groups of personnel or non-critical mission equipment. This is an indication that Azeri forces had a robust supply of Turkish air-to-ground missiles and Israeli loitering munitions.

Turkish Role

All indications point to Azerbaijan having acquired a substantial portion, if not all, of its TB2s from Turkey just prior to the onset of the war. In the beginning of 2020 alone, Turkey reportedly sold Azerbaijan more than $120 million worth of military equipment.[15] Given the rapid buildup of equipment, subject matter experts have expressed doubts that Azeri forces could have received an appropriate level of training and proficiency on the new equipment in a timely enough manner to conduct such lethal strikes. A reasonable conclusion is that Turkey likely played an expanded role in the attacks on Armenian forces and equipment. Mark Episkopos, a reporter for National Interest, highlights Turkey’s potential role in the war:

There is a mounting body of evidence that the Azberbaijani war effort was planned, coordinated, and in large part executed by Turkey. Military aid from Ankara included, but was not limited to, Turkish Bayraktar TB2 drones, special forces commandos, and Turkish-affiliated Syrian mercenaries. Turkish control over Azerbaijan’s armed forces is so deeply embedded that there are reports of Azerbaijani military officers being fired at Ankara’s behest after criticizing the extent of Turkish involvement in the Nagorno-Karabakh conflict. Turkish president Recep Tayyip Erdoğan’s recent participation at a victory parade in Baku evoked a striking portrait: the two leaders stood side by side, behind them a row of Turkish and Azerbaijani flags arrayed in equal numbers. Indeed, Turkish flags were just as ubiquitous on the streets of Baku as their Azerbaijani counterparts.[16]

The regional implications of Turkish involvement in the South Caucasus are far more compelling than just its apparent “support” of—some may assert “control” over—Azerbaijani sovereignty. With Russian peace-keeping efforts underway for at least the next five years, it remains unclear who will actually control Nagorno-Karabakh and which nation’s influence will dominate the larger South Caucasus region.

General Lessons Identified

In assessing the lessons that we have identified in this conflict, it is important to consider the overall context in terms of making a realistic determination as to the applicability of these lessons to U.S. military planning. While the conflict took place between two state actors, they do not share the same capability as near-peer competitors to the United States. Therefore, the operational and strategic choices each country made could be vastly different from a conflict between the U.S. and a near-peer. Neither has an air force that could realistically compete with the United States Air Force. Peer competitors like Russia and China would not be expected to fight in the same way as Armenia and Azerbaijan.

Additionally, non-state actors such as terrorist groups will employ UAS in different ways than state actors. While it is safe to assume that they will lack the air power of a state actor, they can achieve tactical air superiority in limited circumstances.[17] State actors that lack the air capability and capacity of peer competitors will similarly look to identify weak points where UAS can be applied to gain asymmetric advantage. Russia and China can be expected to augment manned systems with ISR missions and targeting support while also engaging in manned-unmanned teaming. While the U.S. military may not face Russian or Chinese forces in direct conflict, proxies around the world should be expected to employ their systems and TTPs at a similar level of capability.

For the DoD to truly internalize the lessons that have been identified and move them to the lessons learned column, it is necessary to consider what these lessons mean for the United States in future conflict. Put another way, how will adversaries employ these UAS threats against the Joint Force? Both peer and non-peer adversaries will use UAS to seek asymmetric advantage against the Joint Force. They will leverage hard to detect nap-of-the-earth flight profiles while masking radar cross sections to exploit weaknesses such as non 360-degree sensor coverage and man-in-the-loop command and control (C2) systems. They will seek ways to multiply effects such as through SEAD. It is generally recognized that most advanced air and missile defense systems are not designed to defend themselves against the full spectrum of threats. This will remain an issue until short-range, point defense systems such as the Army’s M-SHORAD or Iron Dome are deployed within a layered defense. Additionally, peers will employ complex attacks. UAS may not be the threat in terms of an end unto itself; these systems will also be employed as one component of a threat that includes ballistic missiles, cruise missiles, and air-to-ground-missiles. They will combine manned and unmanned teaming and swarm tactics to clutter the operational environment and attempt to overwhelm operators with decisions and engagement sequences.

• The security of rear area operations can no longer be taken for granted. UAS can support deep attacks on supply lines and enable a degree of “control” in the defender’s rear area. Today’s battlefield is no longer linear. Rather than attacking head-on, adversaries will employ UAS to facilitate 360-degree attacks along multiple, unpredictable routes in order to probe weak points and create surprise. Furthermore, UAS can potentially provide uninhibited ISR from above defensive positions.
   
• Employment of UAS provides a cost-effective mechanism to augment air capability. An opponent whose air forces may be considered comparable to or weaker than its adversary will incorporate UAS to achieve an asymmetric advantage. This will include targeting support, employment of loitering munitions, and a range of reconnaissance and security operations. While an adversary may be unable to operate freely in the air domain across an entire area of operations, it may be able to achieve localized air superiority with the use of tactical UAS.
   
• Adversarial decisive operations will likely leverage electronic warfare to support UAS employment when achieving quantifiable destruction is critical. Adversaries will employ electronic warfare systems to jam radars in order to increase the effectiveness of UAS attacks, particularly when conducting SEAD missions.

Implications for DoD C-sUAS Strategy?

The Department of Defense’s first C-sUAS Strategy was approved in December 2020, establishing a framework for dealing with the C-sUAS challenge along a spectrum of hazard to threat in three distinct operating environments: homeland, host nations, and contingency locations.[18] To fully assess how the Department of Defense should respond to the lessons that have been identified from this conflict, it is important to consider whether and how the recently approved strategy addresses them in its current form.

As developed, the DoD strategy enables an array of potential technical solutions in order to address a wide variety of threats. These solutions include kinetic and non-kinetic systems ranging from directed energy to jammers to projectiles. Central to this is a C2 architecture that facilitates interoperability across the Joint Force as well as with international partners and other federal agencies.

Also underway – and stemming from the strategy – is a Joint Training Concept that covers the spectrum from understanding what sUAS are and how to deal with them to long-term development of the C-UAS Training Academy with the Army’s Fires Center of Excellence. Portable, online courses will augment existing programs of instruction within the Military Services.

The DoD strategy also facilitates a three-pronged approach to keeping pace with a threat that is clearly evolving rapidly. First, it directs proactive coordination with the intelligence community in order to identify intelligence requirements so that the Joint Force can stay ahead of the threat before it is demonstrated on the battlefield. Second, the Department will leverage existing science and technology investment to expedite the development of innovative solutions through partnerships with industry, academia, and research facilities. Third, the operational requirements that dictate the key performance parameters that future C-sUAS capabilities must meet will be updated every 18-24 months to pace the threat. All of this is centered on an enterprise approach – across the Department and beyond – with the Joint C-sUAS Office in position to unify the overall effort.

The DoD strategy builds the train, but the implementation plan drives it. The importance of the strategy lies much more in the Department’s ability to flesh out the tasks that fall under its three lines of effort (Ready the Force, Defend the Force, and Build the Team) than in the text of the strategy itself. This implementation plan and assorted, supporting efforts provide the depth and flexibility which will be paramount in enabling the Department to defeat UAS threats.

Recommendations

Going forward, what steps should the Department take in response to ensure that it has a trained Joint Force with a suite of solutions that can protect personnel, facilities, assets, and missions from both current and future UAS threats?

1. DoD must integrate conventional air defense systems with C-UAS defenses. Specifically, the Department should continue to emphasize the development of layered, “system of systems” defenses. This includes the incorporation of C-UAS into all critical asset defenses and linkage with concepts such as Joint All Domain Command and Control (JADC2) and network-centric systems such as the Army’s Integrated Air and Missile Defense Battle Command System (IBCS).

2. DoD must incorporate machine learning and autonomy into C2 systems. Individual operators lack the bandwidth to deal with UAS swarms, so automatable decision support tools that provide “any sensor, best shooter” solutions are critical.

3. DoD must develop comprehensive defense concepts that integrate offensive and defensive fires. The Joint Force must be capable of integrating offensive fires with both active and passive defensive measures. Capacity will always be a challenge with defensive systems, so it is necessary to leverage attack operations that can potentially reduce threats and divert adversary resources. Passive defenses that apply cover and concealment will be instrumental in preventing successful ISR without revealing operating positions.
    
4. DoD must prepare the Joint Force to operate in a communications-degraded and/or denied environment. Adversaries will jam radars and radio signals and will seek to suppress C-UAS and air defense systems. Joint C-UAS training must incorporate the use of analog systems and redundant communications. Further enhancements to the resiliency of communications and situational awareness tools are also needed.
    
5. DoD must enable creativity and adaptability in C-UAS forces. This includes developing training and enabling flexible concepts of employment that support organizational flattening, seeking efficiencies in the modification of TTPs, and facilitating quick decisions at all levels. This is essentially an extreme example of mission command and is conceptually similar to how Special Operations Forces operate.
    
6. DoD must enable enhanced information sharing with our allies and partners and support the creation of interoperable solutions. Our allies and partners are key enablers. To more effectively protect U.S. forces abroad, the DoD must make collaborative research, development, test, and evaluation efforts and integrated systems a priority.

7. DoD must increase investment in non-materiel solutions. Materiel solutions are not enough by themselves. Non-materiel solutions (e.g., doctrine, training, etc.) must be synchronized with materiel to maximize their effectiveness. The right system with the wrong TTPs for employment will be ineffective. 

Conclusion

The improved capability of UAS combined with decreasing price points has lowered the barrier for entry into conflict in the air domain. The relatively low cost and high capability of these systems provides smaller states and non-state actors the ability to change the status quo. Countries such as Azerbaijan now have improved access to air power, which in turn has upended existing notions regarding traditional air dominance. Azerbaijan’s preparation for and execution of the recent conflict in the Nagorno-Karabakh provides a blueprint for state and non-state actors alike to easily modernize their military forces and rapidly augment air operations. The decreased personal risk and stand-off capability have altered the calculus for conflict, potentially increasing its likelihood and opening the door to the renewal of previously frozen conflicts. The United States military needs to heed the warning of Nagorno-Karabakh in order to effectively prepare for the battlefield of the future. However, the DoD must not simply look at tactical lessons learned from this conflict. The Department must also consider the wider strategic implications it suggests such as the likelihood for other frozen conflicts to heat up amid the clear benefits this low cost, low risk capability can provide to countries looking to change their fate.