(cba:news) Struggling with Olde Whiteface

[Joe Patterson]

A number of CBAers have been waging a good fight with 15th mag dwarf novae
this week.  But it has also been very hazy in the eastern USA through the
week (actually since March)... and I wanted to toss in some remarks about
photometry, in case you find them useful.

In theory, clouds equally cover the comparison star, the variable, and the
sky annulus.  Then the software can do the appropriate
subtraction/division to yield a delta-magnitude of reasonable quality.
Degraded by extra noise, but still reasonable.

We usually don't consider the possibility that clouds systematically
prefer the variable to the comp, or vice versa.  Too Oliver-Stone-ish for
normal science.

But sky background, now that's another story.  You can basically rely on
clouds to extinct stars pretty equally... but the effect of sky background
depends very greatly on how lit up the clouds happen to be.  Brightly lit
clouds *increase* the counts when you're observing a faint star, but
*decrease* counts when you're observing a bright star.  You might think,
"well, doesn't the delta-mag extraction take that into account?"  Yes,
sort of... in theory it does.  But we are afflicted with ambitious hopes
and modest scopes - so when moonlit clouds arrive in our data, sometimes
the counts are entirely dominated by noise in the rapidly varying sky
brightness, and the delta-mag has little to do with the target star.

You can expect this scenario to kick in when the stars (either star) are
faint and the sky background bright and varying.  Proximity to the Moon is
additionally deadly, since your general sky inspection may underestimate
the perils in the exact pointing direction.  This is a pretty good
description of V699 Oph in the last week, for many observers.

You can evaluate the peril by inspecting individual frames, to see if the
stars are actually swamped by sky counts.  Or, you can switch to a
brighter star.  All these problems are much ameliorated when you go to a
brighter star.

Speaking of which... anyone ever heard of V603 Aquilae?

joe