Visual photometry

Human eyes are not the same as electronic detectors.

Our astronomer was chagrined to find that, even though we’ve described aspects of his latest paper here and here, we haven’t fulfilled our promise to summarize what it says.  If you want to read it in detail, it can be found here; the following is a summary.

The subject is visual photometry.  That’s the technical name for judging how bright something is, just using your eyes.  In this particular application, it means using a chart of stars whose brightnesses have been measured by carefully-calibrated instruments, and using these standards to estimate how bright a variable star is.  It has a long history; the American Association of Variable Star Observers (AAVSO) was founded in 1911, and work in the area had gone on well before that.

Of course measuring a star’s brightness with electronic instruments is more precise and better-standardized, and such instruments (including things like digital cameras, if properly calibrated) are ubiquitous nowadays.  But not everyone has them and uses them; and there’s a great deal of historical visual data.  Not even the best CCD in 2023 can tell how bright Omicron Ceti was in 1923.  So there is some point in understanding how the human eye measures stars.

Well, it turns out, not as you’d expect.  Our astronomer analyzed ten stars of various types in the AAVSO database, looking in particular at the scatter around the averaged light curve (a plot of brightness versus time).  This is a measure of how precise visual photometry is, and so pretty important if you’re going to use it.  The scatter varied by more than a factor of two from star to star, which is quite a lot.  An electronic detector would have a very steady scatter, if properly employed.

This is actually not terribly surprising, especially if you’ve tried to estimate brightness yourself.  Some stars seem just harder to work with than others; so they should show a larger scatter.  But it’s not for any of the reasons we could think of.  Stars of quick variability, or slow; large variations, or small; pretty regular, or unpredictable; there seemed to be no pattern.  We really expected naked-eye variables to be hardest: similar-brightness comparison stars are generally far away in the sky, and there’s lot of distracting extraneous light.  But they turned out to have the least scatter.

We thought we could rely on color.  It’s well known that people have different responses to color.  The lens of the eye turns yellow as one ages, meaning older observers are less sensitive to blue light.  Obviously, comparing a highly-colored star to standards of normal hue should result in a much wider scatter among observers.  It doesn’t happen.

We’re left scratching our heads.  The difference in scatter is quite real, but it’s not due to any cause we’ve been able to think of, including some there’s not space to mention here.  People are strange.