Caught in the Act: The flickering accretion disk of a dwarf nova

My telescope under the stars at the Giovale Open Deck Observatory, Lowell Observatory, Flagstaff. The bright "star" near center is the International Space Station passing overhead.

On the evening of 9 July 2026, a small refractor no wider than a coffee mug pointed at an unremarkable speck in the constellation Cygnus and started taking pictures, one every half-minute, for six hours. The speck is called SS Cygni, and it is one of the most famous variable stars in the sky. I expected to catch it dozing. Instead, I caught it erupting.

A star that refuses to sit still

SS Cygni was discovered in 1896 by Louisa Wells, one of the women "computers" at the Harvard College Observatory who combed through glass photographic plates star by star. What she found was a star that brightens and fades roughly every six to eleven weeks, and it has been watched almost without interruption ever since. For well over a century, amateur astronomers around the world have logged its brightness night after night, making SS Cygni one of the best-studied stars in all of astronomy. Its record is a monument to what patient backyard observers can build.

It is also just one entry in a vast catalog. The map below plots all 10.3 million variable stars in the AAVSO's Variable Star Index (VSX): the glowing band is the Milky Way, and the two blobs at lower right are the Magellanic Clouds, our galaxy's satellite companions. SS Cygni is a single marked dot in that swarm; the art of this kind of project is choosing one target and studying it in depth.

Two features of the map reward a second look. The arrow marks the Kepler field, the primary patch of sky NASA's planet-hunting Kepler spacecraft stared at, without blinking, for four years. Kepler's job was to hunt for planets around other stars; but measuring those stars so relentlessly turned up a bounty of new variables as a bonus, so many that the region still stands out as a distinct knot of dots. (When the spacecraft later lost the stabilizers that held it on that field, its extended K2 mission hopped between a series of other patches along the sky, and a few of those show up as fainter knots elsewhere on the map.) The faint swirls and loops threading the whole map are not structures in the sky at all: they are the fingerprint of Europe's Gaia spacecraft, whose billion-star survey now feeds VSX. Gaia sweeps the heavens along a fixed, looping scanning pattern, logging slightly more variables along the tracks it has crossed most often. The map records not only where the variable stars are, but how we have gone looking for them.

The variable sky: 10.3 million known variable stars (author's own map, galactic coordinates), with SS Cygni (red) and the Kepler field (arrowed) marked. The looping texture is the scanning pattern of the Gaia spacecraft, whose catalog now feeds the VSX database: the sky as the survey saw it.

SS Cygni is the prototype of a class called dwarf novae, a type of "cataclysmic variable." Every several weeks it flares up by three to four magnitudes (roughly twenty-fold in brightness) over a day or two, holds there for a week or more, then slides back to quiescence. It has done this for as long as we have watched, but the interval between outbursts wanders unpredictably from one cycle to the next.

Six months of SS Cygni, straight from the AAVSO's public database: real data, not a sketch. Four eruptions, and the gaps between them (labeled) wander from one cycle to the next. The most recent eruption, at right, is the one I caught; the figures further on zoom into it, and then into a single six-hour night.

A telescope that changed its mind

The instrument that caught all this was never built for science. It is a small apochromatic refractor bought for astrophotography: the art of long, deep exposures that turn faint nebulae into wall-worthy color. The image below on the right is one such shot from this very rig: the Wizard Nebula, a cloud of glowing hydrogen and oxygen 8,000 light-years away.

What changed is the goal. Instead of chasing pretty pictures, I set out to make the telescope measure: to turn each exposure into precise brightness numbers for hundreds of stars at once. That meant building a photometry pipeline from scratch, done here in collaboration with the AI assistant Claude. The same rig that paints nebulae now does quantitative variable-star science; this article is one of its first results.

Left: the rig, a 60 mm apochromatic refractor on a German-equatorial mount. Right, top: the Wizard Nebula (NGC 7380); below it, the Rosette Nebula (NGC 2244), a stellar nursery in Monoceros. Both are deep-sky astrophotographs from the same telescope, which was bought for images like these and repurposed for measurement.

An engine built from two stars

The secret is that SS Cygni is not one star but two, locked in a furious 6.6-hour orbit, just a few times the Earth–Moon distance apart, almost close enough to touch. One is a white dwarf: the burnt-out, Earth-sized core of a star like the Sun, so dense a teaspoonful would weigh tonnes. Its companion is an ordinary cool red star. The white dwarf's gravity is so fierce that it peels gas off the companion in a continuous stream.

That gas cannot fall straight in; the orbital motion whips it into a swirling accretion disk around the white dwarf. Where the incoming stream slams into the disk's outer edge, it shock-heats a brilliant bright spot. Every so often the disk itself becomes unstable, dumps its stored gas onto the white dwarf all at once, and blazes up: the dwarf-nova outburst.

The anatomy of a dwarf nova like SS Cygni: a white dwarf pulls gas from its red-dwarf companion into a hot accretion disk, and the stream lights a flickering bright spot where it strikes the disk's rim — the whole system orbiting once every 6.6 hours.

Two of those processes leave marks I could measure in a single night. The disk is no smooth, placid ring: it is violently hot and turbulent, churned by magnetic forces that supply the friction gas needs to spiral inward at all. Because that infall is clumpy and chaotic — knots of gas arriving unevenly, the bright spot flaring and sputtering — the system's light never holds still. It flickers, second to second and minute to minute, and that flicker is a direct readout of the turbulent physics of accretion. Layered on top is a slower beat tied to the 6.6-hour orbit: the bright spot and the tidally lopsided disk are not symmetric, so as the pair wheels around we view them from continuously changing angles, and the combined glow swells and ebbs — much as a slowly turning, irregular object catches the light differently as it spins.

What I caught

My photometry placed SS Cygni more than a magnitude brighter than its quiescent level — it was plainly in outburst. Riding on that elevated glow were two things at once. A slow, smooth rise-and-fall spanning the whole night, on a rhythm close to the 6.6-hour orbit — the wheeling view of the bright spot and the lopsided disk. And, superimposed, a restless flicker: rapid, random twinkling of a few percent from minute to minute — the raw sound of gas crashing onto the disk, captured live.

But where in the eruption did my six hours fall? A single night, on its own, cannot say — and here is where the crowd comes in. SS Cygni is watched almost nightly by observers around the world, and their public measurements, pulled straight from the AAVSO database, let me rebuild the entire outburst and drop a pin on my night. The story is clean: SS Cygni climbed out of quiescence near magnitude 11.6 in about four days in late June, held near maximum (magnitude 8.4) for roughly two weeks, and was already fading when I observed it on July 9–10, about six days down the decline, halfway back toward quiescence. I caught it on the way down.

The same eruption up close — the last of the four in the six-month view above, now assembled in detail from the AAVSO's public database (the pooled work of observers worldwide). My own six-hour run (red line) is a single night far down the descending branch. No one telescope built this curve; the community did.

That wider context also settles what I saw within my night. Over six hours the outburst's overall fade amounts to only a few hundredths of a magnitude — too gentle to notice. So the clear dip-and-recover in my own data, about a sixth of a magnitude deep, is not the eruption declining: it is the 6.6-hour orbital wave, the disk and its bright spot turning through my line of sight.

Left: SS Cygni's brightness through the night — elevated (in outburst) and tracing a slow ~orbital wave (red line); the gap near the middle is the meridian flip (explained below). Right: strip that wave away and a rapid, chaotic flicker remains — the accretion disk's turbulence, recorded from a backyard-class telescope.

The gap in that light curve has a mundane, mechanical cause worth a word. My telescope rides a German equatorial mount, which cradles the tube out to one side of a central pier. As a star drifts across the meridian — the north–south line marking its highest point in the sky — the tube would eventually swing down into that pier. To avoid the collision the mount performs a meridian flip: it rotates to the other side and re-points at the target, which now sits upside-down in the frame. The maneuver costs a few minutes of downtime — that is the gap — and rotates the entire star field by 180°. My pipeline shrugs it off by matching stars in celestial coordinates rather than by their pixel positions, so the flip leaves no seam in the measurements, only the brief pause while the mount was moving.

One field, hundreds of variables

SS Cygni was the target, but a wide-field camera is not a sniper; it is a dragnet. Every 30-second exposure captured not just my dwarf nova but every star in a patch of sky two Moon-widths across. When the night was reduced, the pipeline had logged 84,440 individual brightness measurements of 370 catalogued variable stars, from that one pointing, in a single sitting. They span more than two dozen varieties: pulsating stars swelling and shrinking, eclipsing binaries winking as one star crosses in front of another, spotted stars carried around by their own rotation, slow red giants breathing over weeks. One target draws the telescope; the whole zoo comes along for free. It is a small illustration of why wide-field time-domain surveys have become one of astronomy's most productive tools, and why a backyard rig, pointed with intent, can add real data on many stars at once.

Why a faint speck matters

Because it is close and endlessly repeating, SS Cygni is a natural laboratory for one of astrophysics' hardest problems: how gas actually spirals inward and accretes: the same physics that feeds supermassive black holes and builds planets. Its century-long light curve, built largely by amateurs, is a record professionals cannot reproduce any other way.

And that amateur record does more than sit in an archive; it drives front-line research. The best example is SS Cygni's own distance. For years two measurements disagreed, and the discrepancy mattered: the Hubble Space Telescope placed the system far enough away that the reigning theory of dwarf-nova outbursts should have failed for it — an embarrassing crack in the model. To settle the question, astronomers turned to radio telescopes spread across the globe, which can pin a position with exquisite precision. But there was a catch: SS Cygni emits radio waves only briefly, during outburst. The only way to know when to point the dishes was the amateur network. AAVSO observers, watching night after night, sounded the alarm the instant an eruption began, and the professionals slewed within hours. The result, published in Science in 2013, placed SS Cygni about 370 light-years away, comfortably back in line with theory. A backyard alert had triggered a world-class radio campaign. That is what citizen science looks like when it works: not amateurs playing at research, but amateurs supplying the one thing the professionals cannot: eyes on the sky, every clear night, for decades. Anyone with a small telescope and a steady habit can join that record tonight.

All figures original, created for this article (free to share). Photometry: Apertura 60 mm refractor + ZWO ASI294MM Pro at the Lowell Observatory, reduced against the AAVSO VSX and Gaia DR3 catalogs with my own varstar pipeline. A technical companion report is available.