the flickering star in Norma — what IS it?

I have been putting this off because I assumed somebody would tell me it was obvious, but four years in I have given up waiting.

There is a star in Norma, naked-eye, that sits most nights at about fifth magnitude and does not stay there. I have timed the swings repeatedly and there is no pattern I can find: sometimes it brightens to fourth magnitude over a week, sometimes it fades past sixth and disappears from naked-eye view for a fortnight, and then it is back at fifth as though nothing happened. It is there, in some state, every single clear night I have looked, going back to my first log entry in March 1995.

What I want to know is what kind of variable this is. Not "it's a variable star," which I had worked out. What is driving the swings? Why no period I can pin down after four years of trying?

-- dark_adapted
Blue Mountains NSW · 10" f/5 Dobsonian · naked eye for the flicker watch

I get this one too and I am fairly confident I know what it is. It is a nova, fading back down after an outburst you probably missed the peak of. The slow ragged decline is the giveaway — novae take months to years to settle, and the fading is never perfectly smooth.

Mystery solved I think! Worth keeping an eye on it as it keeps settling, since the decline rate itself is interesting.

-- nightfader
Adelaide · 8" Newtonian · binoculars for the wide stuff

It is not a nova. It is just a variable star. There are thousands cataloged. Nothing to see here.

-- old_dob
Newcastle · observing since 1961 · a Dobsonian older than you

Neither of those answers survives the actual numbers, and I do not think either of you has written the numbers down, so here they are.

> It is a nova, fading back down after an outburst you
> probably missed the peak of.

A nova decline is a one-way trip. It fades, sometimes with small wiggles superimposed, but it does not climb a full magnitude back up and then fade again, repeatedly, for four years running. What dark_adapted is describing — repeated brightenings and fadings, same rough range each time — is not a decline curve, it is a cycle. Wrong shape entirely.

> It is just a variable star. There are thousands cataloged.

"Variable star" is not an answer, it is the name of the whole category we are trying to narrow down. The classical pulsators — Cepheids, RR Lyraes, Miras — keep a period tight enough to set your watch by, some to within minutes over decades. Four years of logging with no repeatable period rules every one of them out. A two-magnitude swing (roughly a factor of six in brightness) on an irregular days-to-weeks timescale looks nothing like a pulsator.

What it does look like is an eruptive variable — something with an unstable accretion disk or a young star still settling down, throwing occasional outbursts with no fixed clock, because the driving mechanism is a physical instability, not an orbit or a pulsation mode. That is a real category, and it is the one that fits an irregular multi-year record with no period. It is not proof. It is the shape that survives contact with your log.

-- seeing_hermit
southern Tasmania · 12" Newtonian · keeps a logbook and an argument in equal measure

Fair enough, and thank you for explaining rather than just telling me I was wrong, which is what happened the last time I posted on a board like this.

The point about a nova decline being one-way is the bit that actually landed. I had been pattern-matching on "slow and irregular" without checking whether it went the right direction the whole time. It does not.

-- nightfader
Adelaide · 8" Newtonian · binoculars for the wide stuff

I dictate every naked-eye check onto a cassette recorder rather than writing it down, so I do not need a light and do not lose my dark adaption. Going back through four years of tapes to timestamp every brightening report on this thread, I noticed something nobody had mentioned: every single "it's brightening" log entry, mine included, falls within about five nights of new moon.

I got excited for about ten minutes thinking that was the periodicity we have been missing. It is not. It is a selection effect. A star swinging between roughly fourth and sixth magnitude is only visible to the naked eye at all against a dark sky — under even a gibbous moon it is lost in the glow regardless of what it is actually doing. We can only catch a brightening near new moon. That tells us nothing about when the brightenings actually happen, only about when we are capable of seeing one.

Which means the four years of tapes contain less information than I thought they did an hour ago, and the honest fix is binoculars logging on every clear night regardless of the moon, not just naked-eye checks near new moon.

-- tape_hiss
Ballarat · 6-inch Newtonian · a shoebox of cassette tapes and counting

This is the most useful thing anybody has posted here in a year, so I am going to summarise, because the thread is getting long enough that new readers will not wade through it.

What we think we know: a naked-eye star in Norma swinging roughly two magnitudes on no fixed period, running since at least early 1995. Shape of the variation rules out both a nova decline and the classical pulsators. Best-fitting category is an eruptive variable of some kind. The apparent "always brightens near new moon" pattern is not real — it is a selection effect from naked-eye-only logging, per tape_hiss.

What we do not know, and I want to be honest that this is still nearly everything: the actual period if one exists, the physical mechanism, and whether binocular logging through the moon will even change the picture. I have added it to the logbook as an unidentified variable and I intend to leave it there. Writing "unid" for four years feels like failure, but the alternative is writing down a guess and having somebody quote it back at me in 2003 as a fact.

-- dark_adapted
Blue Mountains NSW · 10" f/5 Dobsonian · naked eye for the flicker watch

Forty-one replies across this and the old board, and we cannot say what it actually is. I want to register that as a failure before anybody gets sentimental about the journey.

But I will grant the following. In 1995 we had a flicker. We now have a flicker with a documented amplitude, a ruled-out shape, a probable category and a diagnosed selection effect in our own methodology. Every one of those was produced by somebody sitting out late writing down what they actually observed, rather than what they expected. tape_hiss's point about the new-moon bias is the best observation made on this board and it came from cross-checking four years of dictated tapes nobody else had bothered to line up against a calendar.

I was dismissive of both nightfader and this star in this thread. Saying "it's just a variable" was not an answer, it was me not wanting to do the work seeing_hermit then did properly. Posting a lazy answer in public and then being shown up for it is how any of this gets done, and it is more than most people manage. My apologies.

Keep logging it as unid. Somebody will read this page in twenty years and know exactly what it was, and they will wonder why we could not see it. Fine. They will have four years of tapes, because we wrote it down.

-- old_dob
Newcastle · observing since 1961 · a Dobsonian older than you