Electromagnetic Warfare: What Is Documented, What Is Myth, and What You Can Check Yourself

A US Air Force airman maintains a high-mobility antenna during Resolute Space 2025 in Guam. US Air Force photo by Senior Airman Jade M. Caldwell / DVIDS. The image depicts training, not a reported interference incident.

Electromagnetic warfare uses the electromagnetic spectrum to detect, disrupt or deceive an adversary and protect friendly operations. Ordinary military use of a radio or navigation signal is not automatically electronic warfare. ^[21] Much of it is engineering: antennas, power levels, frequencies, receivers that stop working. Some subjects attached to it attract a great deal of speculation, and that is where claims drift away from evidence.

This article starts with what governments, aviation regulators, shipping bodies and military analysts have put on the record. It then looks at three subjects often bundled with it: the HAARP research facility in Alaska, the "Havana syndrome" illnesses, and the electromagnetic pulse (EMP) threat. Where a claim rests on a news report rather than a primary document, the text says so.

The central question is what evidence can distinguish documented interference and its consequences from claims about its source or a particular weapon. Battlefield jamming and civil navigation are the core; EMP, HAARP and anomalous health incidents are separate cases used to test the limits of that evidence, not evidence of a single operation.

The Royal United Services Institute (RUSI), a London defence think tank, describes three functions of electronic warfare (EW) in a January 2025 paper on land operations.^[1] Electronic surveillance detects and tracks adversaries and collects and analyses their waveforms. Electronic attack can jam enemy communications and navigation, inflict permanent damage on enemy electronics, or deliver cyber payloads. Electronic defence disrupts enemy kill chains and interferes with enemy munitions to protect the force, for example by spoofing or otherwise disrupting a munition or targeting system.

Two GNSS-specific terms matter below:

• Jamming is described by the European Union Aviation Safety Agency (EASA) as intentional radio frequency interference that prevents navigation receivers from locking onto satellite signals, leaving the system ineffective or degraded.
• Spoofing is broadcasting counterfeit satellite signals to deceive receivers, causing incorrect position, navigation and timing data. EASA defines both terms on its public GNSS outage page and says all spoofing events should be reported.^[5]

The US government has a strategy for operations in the electromagnetic spectrum. In October 2020 the US Department of Defense issued an Electromagnetic Spectrum Superiority Strategy; according to the Government Accountability Office (GAO), the Department's stated intention was to bring together and expand earlier strategies from 2013 and 2017.^[11] The same GAO work found in a December 2020 report that the Department risked missing the strategy's goals because it had not taken key actions, such as developing an implementation plan.^[11] The strategy document itself could not be retrieved, so its contents are not summarised.

The practice predates satellites. NATO traces electronic warfare to the First World War and records its own electronic warfare policy from 1956. Modern dependence on shared civil and military satellite navigation changes the scale of possible spillover, but a modern map of disruption is not proof of a deliberate military operation. ^[21]

"GNSS" stands for global navigation satellite systems, the family that includes the American GPS and Europe's Galileo. Aircraft and ships rely on them for navigation, which is why interference with these signals reaches civilians and is recorded by aviation and maritime bodies.

Aviation

EASA publishes a Safety Information Bulletin on GNSS interference, first issued on 17 March 2022. The current version is Revision 4, dated 3 July 2026; the previous revision was 5 July 2024.^[3][4] Revision 4 follows analysis of recent jamming and spoofing occurrences, builds on a joint EASA–EUROCONTROL task force, and highlights pilot–controller phraseology, Electronic Flight Bag integration, operational and training measures, and air traffic control capacity. EASA encourages stakeholders to implement its recommended measures. The bulletin is complemented by a separate EASA web page listing affected regions, which EASA says is updated weekly; Revision 3 had already moved the impacted-airspace information there.^[3][4]

When EASA announced Revision 3 in July 2024, it said spoofing is riskier for air operations than jamming because it is harder to detect and may affect multiple systems. It cited pilots who reacted to false terrain-warning (TAWS pull-up) alerts with high-rate uncoordinated climbs.^[4]

EASA's GNSS outages page names the Mediterranean, the Black Sea, the Middle East, the Baltic Sea and the Arctic as sensitive areas, and says the problem has grown since February 2022. The page publishes changing rankings of affected Flight Information Regions.^[5] A regional ranking describes reported effects, not who caused an individual event.

The US Federal Aviation Administration (FAA) has noted the same trend. Safety Alert for Operators (SAFO) 24002, dated 25 January 2024 on the document and listed on the FAA's notice page on 2 February 2024 (the notice has since expired), says civil air operators worldwide report GPS and GNSS jamming and spoofing, which often occur in and around conflict zones, military operations areas and counter-drone protection areas, and calls this a potential safety-of-flight risk.^[9] It recommends that flightcrews monitor aircraft equipment closely, promptly tell air traffic control of any apparent GPS/GNSS degradation, and be ready to operate without GPS/GNSS navigation.^[9]

Shipping

In March 2025 the International Telecommunication Union (ITU), the International Civil Aviation Organization (ICAO) and the International Maritime Organization (IMO) jointly expressed grave concern about rising jamming and spoofing of satellite navigation. They asked states to prevent harmful interference, strengthen resilience, keep conventional navigation backups, and report interference to aviation and maritime authorities and the ITU Radiocommunication Bureau.^[7] Maritime Executive's report of the announcement says IMO Secretary-General Arsenio Dominguez warned of collisions and groundings, and recalls an earlier request from the IMO's Maritime Safety Committee that states limit interference from their territory or warn shipping.^[8]

A study of the Baltic Sea gives one measured picture at sea level. GPSPATRON and Gdynia Maritime University reported a six-month study using a monitoring system at the university in Gdynia; it recorded 84 hours of interference at that site. October showed the highest activity, with six major jamming events totalling 29 hours; the study also reports that some events across the period lasted more than seven hours. The authors report that most incidents were jamming rather than spoofing and that signal-pattern analysis points to a mobile source, which they think likely to be one ship or several vessels. They report that aircraft-based ADS-B monitoring did not register corresponding events. The author is associated with the company that developed the monitoring system; the result is one location rather than a survey of the entire sea.^[10]

Who is responsible?

Interference data shows an effect, not who operated the transmitter or why. Researchers have attributed some Eastern European events to Russian military action, but their attribution is an assessment, not proof of every incident. ^[6]

Drones depend on command links and navigation signals, and RUSI's papers show both are heavily contested in Ukraine. RUSI's January 2025 paper says that in Ukraine EW has become a company-level asset.^[1]

A RUSI report from February 2025 by Jack Watling and Nick Reynolds describes the conditions on the Ukrainian battlefield in the war's third year.^[2] According to it:

• RUSI says navigational jamming is "ubiquitous throughout the combat area" and that jamming of command frequencies is also widespread.
• Most vehicles entering the combat zone carry jammers, primarily aimed at command links or video feeds, and EW is effective in significantly reducing the reliability of unmanned aerial vehicles.
• It also causes widespread deconfliction and fratricide problems, so friendly drone operations are periodic rather than continuous.
• The report describes first-person-view (FPV) drones using wired links as able to operate through heavy radio jamming; operators reported trade-offs including approximately 10 kilometres of range for the designs discussed. This is not a guarantee against all forms of interference or a specification for every later design.

RUSI describes the wire-guided design as a way of reaching areas of heavy jamming that would otherwise make strikes impossible. Figures that circulate online, such as the average lifespan of a Ukrainian drone or the number lost each month, and specifications of named Russian systems, did not appear in the two RUSI papers read for this article, so they are left out.

A high-altitude nuclear detonation could produce an electromagnetic pulse (EMP). How much damage it would do depends on the fields produced, the equipment exposed and the ability to restore services.

The US EMP Commission's 2008 critical-infrastructure report warned that such an event could cause widespread, long-lasting disruption. It considered the power grid and the dependence of other infrastructure on electricity, including the difficulty of replacing damaged equipment.^[12]

In 2019, the Electric Power Research Institute (EPRI) studied the electric transmission system using laboratory tests and modelled pulse environments. Its assessment found that a regional blackout involving multiple states was possible, but its results did not support a blackout across the contiguous United States lasting months or years under the scenarios studied. EPRI warned that damage to protective relays and control or communication systems could hamper recovery. Its expectation of recovery comparable to historical large blackouts assumed robust protection and functioning supporting infrastructure.^[13]

These reports do not test exactly the same question. The Commission assessed critical infrastructure more broadly; EPRI's model focused on transmission and did not include some effects on generation and controls. Neither establishes the outcome of every possible attack. The size and duration of any disruption remain sensitive to the pulse, equipment and restoration conditions.^{[12] [13]}

HAARP

The High-frequency Active Auroral Research Program (HAARP) in Alaska is the subject of many popular claims. Its own published FAQ, from the University of Alaska Fairbanks (UAF) Geophysical Institute, gives these facts: responsibility for the facility and equipment formally passed from the military to UAF on 11 August 2015; the program had been run jointly by the US Air Force and Navy. The Ionospheric Research Instrument is a phased array of 180 high-frequency antennas capable of radiating 3.6 megawatts. The ionosphere starts at about 60 to 80 kilometres altitude, and artificially induced effects are no longer detectable after times ranging from less than a second to 10 minutes. The FAQ describes both active experiments and passive monitoring; its operational history may not reflect current campaigns.^[14]

On the popular claims, the FAQ is direct. It says HAARP cannot control or manipulate weather, because its radio waves do not interact with the troposphere or stratosphere, where weather forms. It also denies that HAARP is used for mind control, saying neuroscience is outside its research.^[14] A facility's description of itself is not independent verification. Nothing gathered for this article, though, supports the weather or mind-control claims. The FAQ does not address earthquake claims, and this article does not rule on them.

Havana syndrome

"Havana syndrome" is the popular name for what US agencies call Anomalous Health Incidents (AHIs). The evidence is published in layers, and the layers disagree.

The National Academies, December 2020. The National Academies of Sciences, Engineering, and Medicine (NASEM) produced a State Department–sponsored report, An Assessment of Illness in U.S. Government Employees and Their Families at Overseas Embassies. The committee found that many of the acute signs, especially the distinctive audio-vestibular ones, were consistent with directed, pulsed radiofrequency energy, and called that the most plausible mechanism among those it considered. It said the finding rests on circumstantial support from decades of open literature, and that other mechanisms, including persistent postural-perceptual dizziness and psychological factors, may contribute to chronic, nonspecific symptoms. Its limits: it was not in a position to assess how the cases arose, such as a possible source, much of the detail was classified or restricted, it was not able to assess specific scenarios involving malevolent actors, and the clinical cases varied extremely. It did not conclude that a weapon was used.^[15]

The intelligence community, January 2025. An assessment issued then judged that most participating intelligence components considered a foreign adversary very unlikely to be responsible; one judged it unlikely and one abstained on that question. Two components also assigned low-confidence, roughly even odds to different narrower propositions about a novel weapon being used or developed for a small subset of cases. The document explicitly recognised that affected people had real symptoms and that a small number of foreign-caused events could not be ruled out. These were historical assessments and were later rescinded. ^{[16] [20]}

The 2025 assessment cited 2024 NIH research. In the NIH cohort, investigators found no consistent MRI-detectable brain injury and no significant differences on most clinical measures compared with controls. NIH said this did not exclude an adverse event at the time, including one leaving no lasting detectable change. ^{[16] [23]}

June 2026: the assessments are rescinded. An ODNI congressional notification dated 11 June 2026 says the Director ordered the National Intelligence Council to rescind the 2023 and 2025 assessments. The notification alleges selective exclusion of evidence, mischaracterisation of sources, omitted source-quality information, reliance on an ethically flawed medical study without methodological critique, and limits on intelligence collection. These are the notification's allegations; it does not identify the medical study in the visible text, and the redacted document does not let readers independently test the allegations. ^[20]

A rescission withdraws two assessments. It is not a finding that a weapon exists, and it does not by itself supply a new conclusion. Whether any device caused any case has not been settled in public.

Flight data maps. GPSJam aggregates aircraft ADS-B navigation-accuracy reports over 24 hours. Its operator says green indicates more than 98 per cent of aircraft reported good accuracy; yellow and red indicate increasing shares reporting poor accuracy. The site subtracts one poor-reporting aircraft per map cell to reduce false positives, which can also hide a possible problem in sparsely sampled cells. ^[22]

• The map records aircraft-reported navigation quality; it does not identify a transmitter, its operator or intent.
• Red or yellow does not by itself prove jamming, or even that the relevant navigation source was GPS. Aircraft manoeuvres, equipment and other factors can also affect reported accuracy. ^[22]
• It describes sampled airspace rather than a comprehensive picture at the surface; the Gdynia monitor found local events not registered by aircraft-based monitoring. ^[10]

Regulator pages and primary reports. EASA's outage page and bulletin are public, and the page ranks the Flight Information Regions most affected by jamming and spoofing.^[3][5] The source list links to the RUSI papers, GAO testimony, National Academies report, Commission report and the historical intelligence assessment; access may vary by reader.^{[1][2][11][12][15][16]} Check claims against these rather than against summaries, this one included.

What you cannot check. This map cannot show the transmitters or establish by itself that a transmitter caused a particular degraded report. You cannot independently examine the classified material behind the 2026 rescission.

Electromagnetic warfare is documented military practice, and navigation interference is an established civilian safety problem. Individual interference maps do not establish an operator or intent. ^{[2] [5] [7]}

The possible scale of a high-altitude nuclear EMP remains disputed; the comparison depends on assumptions and systems studied. HAARP's operator rejects claims of weather or mind control. The Havana illnesses are real, but the public evidence does not establish a common cause or a weapon. The recall of an intelligence assessment does not establish one either. ^{[12] [13] [14] [15] [16] [20]}

Method and limits: This investigation compares dated regulator notices, military research and institutional reports with reporting. The EPRI assessment models the transmission system under specified scenarios and does not cover every consequence of an attack. Intelligence and classified evidence are not independently accessible. A documented assessment is not proof of its conclusions. Evidence reviewed through 4 November 2026.