What the Plasma Hypothesis Explains, and What It Doesn’t

There are currently five competing physical explanations for the Hessdalen lights, all with peer-reviewed literature behind them, none of which accounts for the complete observational record. This is not a failure of science. It is a description of where the science stands. I am going to set out each model, note what it explains well, and note where it runs out.

I am not in the habit of adjudicating between competing physical models in a publication aimed at general readers. But the Hessdalen literature is unusual: the researchers are engineers and physicists publishing in journals with referees, and they disagree with each other in ways that are informative. The disagreement is worth following.

Combustion of mining dust

The simplest hypothesis: airborne particulate matter from the valley’s ore workings — hydrogen, titanium, scandium — ignites in the atmosphere. The valley has been mined for at least two thousand years; the ore dust is real, and it is in the spectral record. Scandium and titanium in the chemical signature of the lights match what you would expect from the local geology.

What combustion does not explain: the radar returns, which persist after the optical emission ends and which track movement at speeds up to 8 km/s; the laser interaction result (eight of nine attempts to direct a laser at an active light doubled its flashing frequency); the ejection events, in which a light ball ejects smaller spheres at estimated velocities of 10,000 to 20,000 metres per second.

Burning dust is part of the answer. It is not a complete answer.

Piezoelectric discharge

The Hessdalen bedrock is rich in quartz. Quartz under mechanical strain produces charge. The piezoelectric hypothesis holds that tectonic micromovement in the crystalline rock generates intense localised charge density sufficient to ionise air and produce luminous plasma — essentially the same mechanism proposed for some ball lightning events.

This is a plausible mechanism with real physics behind it. It does not account for the sustained stationary hover events (the strain would be transient), nor for the ejection velocities, nor for the laser interaction.

The geological battery model (2014)

This is the model I find most mechanistically interesting. The valley sides are geochemically opposed: iron-zinc rock on the east, copper-bearing rock on the west. The river Hesja, carrying sulfuric water from the mine workings, acts as a conductive electrolyte between them. When rainfall, temperature change, or other conditions alter the conductivity of the electrolyte, current flows between the valley walls. The discharge produces plasma.

It is a battery. The valley is a battery.

The model has not been quantified to the point where it can generate testable predictions about individual events. It is a framework, not yet a theory in the full sense. But it uses the actual mineral chemistry of the valley as its material, which gives it more geological grounding than the alternatives.

Dusty plasma and Coulomb crystals (Paiva and Taft, 2010)

Paiva and Taft proposed that the lights are macroscopic Coulomb crystals — structured arrangements of charged dust particles in a plasma state — produced by ionisation of air and mining dust by alpha particles from radon decay in the valley rock. This is the most technically developed single model in the literature.

A Coulomb crystal in plasma has a continuous emission spectrum, which matches the spectral data. The surface temperature estimate (5,000 K) is compatible with the plasma parameters the model predicts. It accounts for the UV-dominant energy budget.

What it does not straightforwardly explain is the radar persistence after optical extinction, or the interaction with the laser.

Electrically active inversion layer (2021)

The most recent peer-reviewed model proposes that atmospheric temperature inversions over the valley floor create electrically active layers — trapped charge in the boundary between cold valley air and warmer air above — that discharge under the right humidity and gradient conditions. The 2021 Springer paper is the primary reference. It is recent and the validation dataset is limited. It may be part of the answer.

Where this leaves the field

Teodorani’s assessment in 2004 remains accurate: “a self-consistent definitive theory cannot be constructed yet quantitatively.” The 2024 VLF survey adds subsurface electromagnetic data that constrains the geological models — the valley floor has electrically active structures that the battery model predicts should be there — but does not close the question.

My own view is that the correct explanation will involve more than one of these mechanisms, operating at different scales and under different conditions. The variation in observed behaviour — from slow hover to 30,000 km/h radar track to stationary discharge — suggests that “the Hessdalen lights” may not be a single phenomenon but a family of related ones, sharing a geological cause but not a single physical process.

I am comfortable with this being unresolved. The data will not stop accumulating.

In this series

Sources