One Simple Power Law Governs Eight Years of Solar X-Ray Flickering

The Sun’s outer atmosphere, the corona, burns at more than a million degrees while the visible surface below it simmers at a mere 5,800 kelvin. What keeps this improbable furnace hot has been one of the most stubborn puzzles in astrophysics for more than eight decades, and the leading suspects are countless small explosions called nanoflares. Now a new statistical analysis of eight years of soft X-ray measurements from the GOES satellite series suggests that the same physical process, whatever it is, operates across the entire dynamic range of solar activity, from the deepest lull of the solar minimum to the most flare-violent maximum. The finding, published in the journal Solar Physics by Hugh S. Hudson of the University of Glasgow and the Space Sciences Laboratory at the University of California, Berkeley, does not identify the heating mechanism directly, but it places a powerful new constraint on any theory that hopes to explain it.

Hudson’s approach is deceptively simple. Instead of cataloguing individual flares or fitting models to light curves, he asked a statistical question about the GOES 1 to 8 angstrom soft X-ray timeseries: how does the variance of the measured flux relate to its mean? This is the domain of Taylor’s law, an empirical scaling relation first described by the ecologist Lionel Taylor in 1961 to explain how the spatial variance of insect populations grows with their average density. In its astronomical and physical applications, the law takes the form var(S) = a times the mean flux raised to a power alpha, where the exponent alpha encodes the character of the underlying variability. When alpha equals one, the fluctuations follow Poisson statistics, the signature of independent, randomly occurring events such as photons arriving from a steady source. Values of alpha greater than one indicate clustering, coherence, or correlated structure in the variability, and the precise value can discriminate between competing physical pictures.

The result is striking. Across the full eight-year data set, spanning 2018 through 2025 and covering both the quiet rise phase and the active maximum of Solar Cycle 25, Hudson finds a single fitted exponent of alpha = 3.06, with an uncertainty of only 0.05. That value is substantially greater than one, meaning the soft X-ray emission is far from a random Poisson process at every flux level the instrument can measure. More remarkable still, the exponent shows no evidence of changing with time and no dependence on the base-level flux. The same power law that describes the variance during weeks of explosive flare activity also describes it during the stillest stretches of the solar minimum. One number, one scaling relation, one physics.

The implications reach into one of the longest-running debates in solar physics: the coronal heating problem. Since the 1940s, when spectroscopic observations first revealed the corona’s astonishing temperature, researchers have debated whether the heating is delivered by waves propagating up from the convective zone or by the repeated snapping and reconnection of magnetic field lines. In the magnetic picture, championed in its modern form by Eugene Parker in 1988, the corona is heated by a torrent of nanoflares, tiny reconnection events far below the detection threshold of any instrument. A key prediction of some versions of this scenario is that the quiet corona should be heated by a distinct population of small, random events, separate from the large flares we observe. Hudson’s result challenges that division. Because the same variance scaling applies to quiet times as well as active times, he argues, there is no need to invoke a different physics to explain the heating of the quiescent corona.

Crucially, the finding does not rule out episodic heating. The corona is almost certainly heated in discrete bursts rather than by a steady drip of energy. What the analysis shows is that those bursts do not occur randomly. If the nanoflares and microflares were independent Poisson events scattered across the solar disk, the variance would scale linearly with the mean, giving alpha equal to one. Instead, the exponent of roughly three implies strong correlations: heating events cluster in space and time, presumably because they are organized by the magnetic fields that thread the corona and are themselves shaped by the slow churn of the solar dynamo. The statistics of the smallest, invisible events, in this view, are continuous with the statistics of the largest, most spectacular flares, forming a single connected population.

This continuity echoes earlier work on flare statistics. Studies of decades of GOES observations, including a 2012 analysis by Markus Aschwanden and Steven Freeland covering 37 years of data, found evidence for self-organized criticality in the flare frequency distribution, an invariance that persisted across three solar cycles. Self-organized critical systems, like sandpiles poised at the angle of repose, produce avalanches of all sizes with scale-free statistics, and the Sun’s flare distribution has long been cited as a prime astrophysical example. Hudson’s variance analysis adds an independent statistical fingerprint to that picture: not only do flare sizes follow a power law, but the second moment of the flux distribution obeys a power law too, with an exponent that refuses to budge across an entire solar cycle phase.

The paper is also candid about the observational subtleties involved. The GOES X-ray sensor data used in the study are the standard public-domain products from NOAA, and at the lowest flux levels those products carry a complicated background. Hudson’s appendix dissects this contamination with care. During the quietest periods, the two GOES channels, 0.5 to 4 angstroms and 1 to 8 angstroms, register roughly equal signals, which is not what a solar X-ray spectrum would produce; a genuine solar spectrum would show far less flux in the harder channel. The culprit is not photons at all but particles, specifically electrons from the Van Allen radiation belts that pepper the detector even after the Level-2 correction, a correction that itself carries systematic discrepancies. A clear diurnal pattern in the quiet-day data betrays this non-solar contribution, since it tracks the spacecraft’s orbital geometry rather than the Sun.

Hudson writes out the full variance budget explicitly: the total variance of the measured signal contains contributions from the solar photon fluence, the particle fluence, the Poisson counting noise of both photons and electrons, and a term from the digitization of the data into discrete data numbers. This decomposition matters because it defines the frontier of what the standard data product can safely reveal. Below roughly the 0.1 A-class level, the possibly distinct emission of the truly quiet corona becomes entangled with instrumental backgrounds, and disentangling them would require the raw data and its ancillary information. Even the curious bunching of samples that superficially suggests an exponent below one is traced to digitization: the reported flux can hang up on a single data number, an effect that acts like negative feedback and biases against the detection of minor events. These caveats sharpen rather than weaken the main result, since the robust alpha of about three holds across the dynamic range where the solar signal genuinely dominates.

For the broader community, the analysis offers a compact diagnostic that any coronal heating model must now confront. Wave-heating scenarios, steady heating scenarios, and random nanoflare scenarios each predict different relationships between fluctuations and mean emission, and a measured exponent of three, stable across eight years and five orders of magnitude in flux, is a demanding target. It also connects solar physics to a much wider scientific landscape, since Taylor’s law and its generalizations appear in fields from ecology to finance, and theoretical work on Tweedie distributions has provided a mathematical basis linking such power laws to 1/f noise and multifractality. The Sun, it turns out, keeps statistical company with ecosystems and markets, a reminder that fluctuation scaling is one of nature’s recurring motifs.

The next steps are clear. Extending the variance analysis to raw GOES data, or to faster and more sensitive instruments such as the SphinX photometer flown on earlier missions, could probe the quiet-corona regime where the distinct signature of background heating, if any exists, should appear. Comparing the soft X-ray scaling with hard X-ray and extreme-ultraviolet variability would test whether the same correlated physics governs different temperature regimes of the corona. And applying the same Taylor-law analysis to stellar flares, where missions like Kepler have catalogued tens of thousands of events on other stars, could reveal whether the Sun’s exponent of three is a universal fingerprint of magnetically driven atmospheres. For now, the message of eight years of X-ray flickering is elegantly simple: from the faintest whisper of the quiet corona to the roar of the largest flares, the Sun’s heating events belong to one connected, correlated family, and any complete theory of the coronal heating problem must explain them all with a single set of rules.

Subject of Research: Statistical scaling of solar soft X-ray flux variance and its implications for coronal heating

Article Title: The Variance of Solar Soft X-ray Fluxes

Article References: Hudson, H. S. (2026). The Variance of Solar Soft X-ray Fluxes. Solar Physics, 301(10), Article 154. https://doi.org/10.1007/s11207-026-02748-5

Image Credits: AI Generated

DOI: 10.1007/s11207-026-02748-5

Keywords: solar physics, coronal heating, solar flares, nanoflares, GOES, soft X-rays, Taylor's law, variance scaling, self-organized criticality, solar cycle 25, Poisson statistics, space weather