Hessdalen Lights: What 40 Years of Data Show

Sunday, February 12, 1984, 7:35 p.m. On a snowbound plateau in the Norwegian valley of Hessdalen stands a travel trailer full of borrowed instruments. Outside, at four below zero Fahrenheit, a man watches a light through binoculars while a colleague raises a laser at it. What happens next is written down, timed, and left uninterpreted — and that is the whole reason this valley matters. It is the only place on earth where a so-called UFO phenomenon has its own permanently staffed measuring station.

A valley, 150 people, December 1981

Hessdalen lies in central Norway, southeast of Trondheim, about nineteen miles northwest of the old mining town of Røros. The valley runs roughly seven miles, a river called the Hesja flows south to north through it, the mountains rise above 3,000 feet to the west and stay lower to the east. When the story begins, around 150 people live there.

In December 1981 lights appear that nobody can place. They hang still for over an hour, drift slowly through the valley, stop. Radar later measures one at something on the order of 19,000 miles per hour. They show up just above rooftops, just above the ground, high in the sky — but usually below the summits. In photographs they have shapes: a sphere, a projectile pointing down, an upside-down Christmas tree. Mostly white or yellow-white, sometimes with a small red light above or below. Up to four times a day, mostly in the evening, mostly in winter; in the bright Norwegian summer almost never.

Older accounts from the valley are said to reach back into the nineteenth century. That is tradition, not a measurement series, and the 1984 report treats it that way.

Five private citizens, a trailer, and 3,300 questionnaires

Here is the genuinely unusual part. No agency and no institute takes an interest. So on June 3, 1983, five private citizens found their own project, with a working committee and an advisory committee. On the advisory committee sits J. Allen Hynek — the Northwestern University astronomer who had served for two decades as the US Air Force's scientific consultant on UFO reports, who coined the phrase "close encounter," and who in 1985 would call this valley a "UFO laboratory." On the working committee sits Erling Strand, then an engineer, later a lecturer at Østfold University College, who shapes the project to this day.

In January 1984 they mail an information sheet with a simple reporting form to 3,300 households. On January 21 the field phase begins and runs to February 26, with three teams posted on ridges and plateaus and a trailer on the Aspåskjølen plateau holding most of the instruments.

The instrument list is where this parts company with everything else filed under "UFO": cameras with diffraction gratings to capture spectra, a seismograph installed the previous October to test the then-fresh "earth lights" hypothesis, a radar with a twenty-mile range, a spectrum analyzer sweeping continuously from 150 kilohertz to 1,250 megahertz, a fluxgate magnetometer, a laser, Geiger counters, an infrared viewer. Plus specialists you would not expect: the Norwegian Defence Research Establishment helped with planning and equipment, astrophysicists in Oslo evaluated the spectra, seismologists from Bergen ran the seismograph.

188 reports, 53 left over

Five weeks in the field produce 188 reports of lights. Strand then does something worth remembering, because it is the core of any honest anomaly research: he scores every report twice. A G-value from 1 to 10 says how well documented it is. An F-value from 1 to 10 says how likely a known light is behind it — F1 means "certainly an aircraft," F10 means "no explanation found." He notes in writing that the scoring is subjective and is therefore always done by two people.

The result: 135 reports land at F1 to F4, a known light. Fifty-three sit at F5 or higher — what the report calls "the Hessdalen phenomenon." Only six reach F9 or F10, and only four of those are also well documented. Strand draws the conclusion a less careful person would not have drawn: too few to derive anything common from the strongest cases.

Then the instruments. Radar delivers 36 echoes, only three of them to lights anyone saw at the same time; the rest stayed invisible. On Friday, January 27, at 10:58 p.m., a light travels south to north very fast, and radar shows two echoes 2.4 seconds apart, about thirteen miles from each other. Two echoes come back as strong as the reflections off the mountains. A radar specialist from the Defence Research Establishment judges, as the report records it: if that is not a solid body but only gas, the gas must be locally and strongly ionized, or there would be no reflection that strong.

The magnetometer registers a near-simultaneous geomagnetic pulsation for four of ten lights rated F5 or higher — Strand calls that a correlation of about forty percent and writes in the very next sentence that it may be coincidence, since there were many pulsations those days. The seismograph records not one local tremor in the entire period. The earth-lights hypothesis gets no support in Hessdalen in 1984.

And then the closing line: they had not found out what the phenomenon is, which was hardly to be expected — but they now knew that whatever it is, it can be measured. Strand signed that report on January 5, 1985.

The American comparison: Marfa and Brown Mountain

This is where an American reader should stop and look sideways, because the United States has at least two famous mystery-light locations and handled both very differently.

Three valleys, three approaches

Location	What was done	Where it stands
Brown Mountain, North Carolina	The USGS sent geologist George R. Mansfield in 1922. He set up observation stations, took repeated azimuth readings with a surveying telescope and plotted them against rail lines, roads and homesteads.	He attributed roughly half the lights to automobile headlights, about a third to locomotive headlights, the rest to fixed lights and fires, with refraction in the basin amplifying them. Published as USGS Circular 646.
Marfa, Texas	In May 2004 a Society of Physics Students group from the University of Texas at Dallas spent four nights with traffic monitoring equipment, video, binoculars and chase cars.	Light frequency correlated with traffic on US 67; motion followed the road; parking a car on US 67 and flashing its headlights reproduced a "Marfa light" at the viewing park.
Hessdalen, Norway	Five weeks of multi-instrument fieldwork in 1984, then a permanent automated station from 1998 onward — and counting.	A residue of unexplained reports, a documented internal dispute over the best-known case, and a peer-reviewed geological survey in 2024. Open.

Notice what the American cases have in common: a short, competent investigation found a mundane cause, published it, and the tourist economy carried on regardless. That is not a failure of science — it is science working quickly. Hessdalen's difference is not that its lights are stranger. It is that somebody kept measuring for forty years, which means the residue there has been squeezed much harder.

The honest summary for a reader in the States: if your local mystery light has never had a surveying telescope and a traffic counter pointed at it, the Brown Mountain and Marfa results are the most likely outcome waiting for it.

The Blue Box: August 1998

After 1984 the lights get rarer, and the project becomes what it still is: a permanent installation. In summer 1998, under Strand and his colleague Bjørn Gitle Hauge of Østfold University College, a container goes into service on the Aspåskjølen that everyone calls the Blue Box — by the project's account the world's first automatic facility for recording phenomena of this kind.

The technical description is unspectacular and that is exactly why it matters. Three CCD cameras sensitive to 0.00003 lux stream around the clock. An alarm system developed by Italian radio astronomers watches one mountain; if something appears, it saves a still and a video sequence beginning a few seconds before the alarm. Two more alarm cameras came from the University of Würzburg. Add a flight detector logging all air traffic, two weather stations and a radar on a thirty-foot mast. In 2024 the station was upgraded to 8-megapixel cameras, and each day's images now sit in a public folder with the cameras running live.

What the station delivers first is statistics. The lights occur more often in winter and preferentially between 10 p.m. and 1 a.m. No correlation with solar activity was found. And frequency falls: from up to twenty sightings a week in the early eighties to roughly twenty a year. An independent 2017 review of the documented reports shows the same downward trend — and adds that whether this is a decline in lights or only in reports remains speculation.

The night that had two truths: August 2002

This is the part that should decide how much weight you give the rest, so it does not get buried.

Through the early 2000s, teams from the Medicina radio telescope near Bologna ran a series of missions in the valley with radio platforms, telescopes, gratings and home-built pulse radar. The astrophysicist Massimo Teodorani summarized the results in 2004 in the Journal of Scientific Exploration, and one number from it has travelled through every video since: a light cluster photographed from about five and a half miles away, calculated at roughly 19 kilowatts of radiated power, plus or minus thirty percent — more than ten times a helicopter searchlight.

In August 2002, standing beside that team with a portable spotting scope, was Matteo Leone — scientific adviser to the Italian committee for the Hessdalen project, that is, the same organization co-funding the mission. When the blinking light appeared over a hill to the south for the second time, the team photographed it and took its spectrum. Leone looked at it at thirty times magnification and saw, as he wrote in 2003, two clearly separated car headlights.

His 29-page rebuttal is arguably the most important reading on Hessdalen precisely because it comes from inside. He does the arithmetic: at the assumed five and a half miles, two headlights a yard apart would have sat 0.19 degrees apart in the scope — he saw distinctly more than a full moon's width. At about a mile and a half it fits: 0.78 degrees, roughly one and a half moons. And at that distance, exactly on the light's bearing, runs a private toll road, nearly invisible in the trees. A car descending it would sweep the observers with its headlights, "blink" behind trunks, and slowly sink. That is precisely how the light behaved in the mission's own images. He had the famous three-peaked spectrum assessed independently by a scientist at a lamp manufacturer, whose verdict was an incandescent bulb, with the structure an artifact of the color film layers.

Teodorani replied in a footnote in 2004 that comparison shots of known continuum sources such as car headlights on the same film stock look entirely different. The dispute has never been settled by a measurement both sides accept.

What you should take from it is not a winner. It is this: on the same night, in the same place, two people with instruments looked at the same light and drew opposite conclusions — and the critic closes his own paper saying the witness accounts he collected in the valley still resist conventional explanation and deserve properly defined study. Hessdalen is not a place where believers face skeptics. It is a place where the same people do not know what they are seeing.

The models, and what their own authors say about them

Four explanations circulate. Every one of them is labeled provisional by the people who proposed it, and that labeling is the most useful thing about them.

• Burning mineral dust (2007). A spectrum with a dominant line suggested ionized air plus silicon, iron, titanium and scandium — dust from the valley. The author wrote underneath that this was a preliminary report, the data were not intensity-calibrated, resolution was low, and the intensity values should be treated "with the utmost caution."
• The natural battery (2013). Rock samples showed zinc and iron west of the river, copper east of it — the layout of a galvanic cell — with sulfuric acid from abandoned sulfur mines as the electrolyte. The paper itself uses the word "speculative." A Norwegian physicist objected publicly that the valley sides are far too far apart for a battery and that such a field could not supply the energy to make gas glow.
• Dusty plasma (2010–2021). Brazilian physicists proposed that the flat spectrum with steep edges could be bremsstrahlung from a dense ionized gas, later adding an electrically active inversion layer. Theoretical models without independent confirmation.
• Ball lightning. Presented at the first international workshop in 1994 and never demonstrated in the valley.

Sixty miles of survey lines: 2024

An international group — geophysicists from Thessaloniki, a volcano electromagnetics specialist from the French CNRS, and the Østfold team — spent six campaigns walking the terrain with VLF receivers, a reading roughly every twenty yards, about sixty miles of survey lines over some forty square miles, plus magnetic field and self-potential measurements. The result appeared in 2024 in the Journal of Applied Geophysics, a regular peer-reviewed journal — the first comprehensive geophysical survey of the valley ever done.

They found a structure: many conductive bodies, from ten to several hundred yards across, from the surface down past 300 feet, mostly sulfide ore. At the surface these trace a ring, an ellipse of roughly four by seven miles, matching the gabbro intrusion beneath the valley; the authors read it as the metamorphic aureole around that intrusion. Such deep-rooted conductive zones, they write, could become traps that concentrate electric charge, helped by the many lakes, streams and permanently wet ground.

Then, very carefully, they offer a "provisional hypothesis": the valley's particular conditions are necessary building blocks for short-lived local lights. Necessary, not sufficient. Their closing line asks for far more multidisciplinary study. Ten years after a critic demanded thorough documentation, it exists — and it leads not to an answer but to a sharper question: why does a valley with this geology glow, when the many other valleys with old copper mines do not?

Where this sits next to the US UAP debate

The American conversation since 2017 has been organized around military sensor data, congressional hearings, and since 2022 a Pentagon office — the All-domain Anomaly Resolution Office — tasked with collecting and resolving reports. A NASA-convened independent study team reported in 2023 with a recommendation that reads like Hessdalen's founding document: the field's central problem is not secrecy but the absence of consistent, calibrated data gathered with instruments built for the purpose.

That is the useful comparison. The United States has spent decades classifying reports, going back to Project Blue Book, which Hynek advised and which closed in 1969. Norway put one container on one plateau and let it run for twenty-seven years. Neither approach has produced an identification. But only one of them can tell you how often the thing happens, in which month, at what hour, and how much of the residue shrinks when you add a better camera. The answer, so far, is that it shrinks a lot.

What we do not promise

We do not claim that the Hessdalen lights are craft, probes or signs of intelligence. No measurement in forty years supports that, and the witness category "craft" on the project's own site is a reporting class, not a finding. We do not claim the kilowatt figures are established — they rest on a distance estimate disputed from within the same committee. We do not claim any of the four models is correct; their authors do not claim it either. And we do not claim the laser episode was a response. Three people recorded a pattern change nine times over; the report deliberately leaves the interpretation open, and so do we.

Practical: how to read a light story

• Ask for the distance first. Nearly every dramatic number in this field — power, size, speed — is a distance estimate wearing a disguise. Get the distance wrong by a factor of four and the power is wrong by a factor of sixteen.
• Ask who scored the report. The 1984 project used two independent raters precisely because it knew the scoring was subjective. Most accounts you will read used none.
• Check the road. In both American cases and in the most famous Norwegian one, the answer was a road nobody had plotted. Look at a map before looking at a theory.
• Prefer the author's own caveat. "Preliminary," "not calibrated," "speculative" — the researchers wrote those words themselves. Any retelling that drops them is less reliable than the paper it came from.
• Let unexplained mean unexplained. It is not a synonym for anything. Fifty-three reports with no known cause is an interesting number and not a conclusion.

Frequently asked questions

What are the Hessdalen lights?

Unexplained luminous phenomena in a valley in central Norway, reported in numbers since December 1981. Most reports resolve to known sources; a residue does not. What the residue is has not been established.

Is there really a permanent measuring station?

Yes. The automatic station known as the Blue Box has operated on the Aspåskjølen plateau since August 1998, with cameras running around the clock, an alarm system, weather stations, radar and a flight detector. It was upgraded with 8-megapixel cameras in 2024.

Did a light really respond to a laser?

On February 12, 1984, a blinking light changed to a double blink in eight of nine attempts while a laser was aimed at it, recorded by three people, with the observer at the binoculars unable to see where the laser pointed. The report logs this as an observation and explicitly leaves open whether it was a response.

How do the Hessdalen lights compare to the Marfa lights?

A 2004 University of Texas at Dallas student study attributed the lights observed southwest of the Marfa viewing park to headlights on US 67, and reproduced one by flashing a car's headlights. Hessdalen has a residue that four decades of instruments have not resolved — but its single most famous case was also attributed to car headlights, by a member of its own research committee.

Are the lights getting rarer?

Reports have fallen from as many as twenty a week in the early 1980s to roughly twenty a year. Whether that reflects fewer lights or fewer reports is not established.

Are they UFOs?

Nothing in forty years of measurement points to craft or intelligence. The valley is interesting for the opposite reason: it is the rare case where an anomaly has been instrumented rather than argued about.

Sources and further reading

• Erling Strand, Project Hessdalen 1984 — Final Technical Report, signed January 5, 1985 — chapters 3.1, 3.4, 3.7, 3.10, 4.5 and 5.
• Matteo Leone, the 2003 rebuttal for the Italian Committee for Project Hessdalen — the headlight analysis and distance arithmetic.
• Massimo Teodorani, Journal of Scientific Exploration 18 (2004) — the EMBLA results and the power estimate, with the author's own caveats.
• Vargemezis, Zlotnicki, Hauge, Kjøniksen and Strand, Journal of Applied Geophysics 226 (2024) — the geophysical survey and the "provisional hypothesis."
• George R. Mansfield, Origin of the Brown Mountain Light in North Carolina, US Geological Survey (1922; reprinted as Circular 646, 1971).
• Society of Physics Students, University of Texas at Dallas, field investigation of the Marfa lights, May 2004.
• NASA Independent Study Team on Unidentified Anomalous Phenomena, final report, 2023 — on data quality as the field's central problem.
• In the magazine: Alfred Watkins and the Ley Lines · Sacred Places and Power Spots · Shooting Stars and Meteor Showers

This article reports on a research history. It contains no health claim and no promise of results.

The valley's real achievement is not a discovery. It is forty years of somebody writing down what the instruments actually said. #TempleOfDesire