[AAVSO-DIS] PEP

[Michael Newberry]

----- Original Message -----
From: "Brian Skiff" <bas at lowell.edu>
To: <mnewberry at axres.com>
Cc: <aavso-discussion at informer2.cis.McMaster.CA>
Sent: Friday, February 01, 2002 4:03 PM
Subject: Re: [AAVSO-DIS] PEP


> >> In general,
> >> well calibration CCD data will give you better results.
>
>      Assuming you meant to write "well-calibrated CCD data", my reaction
> is "baloney"---or "only at the faint limit of the instrument".

    Baloney is a strong word. Please read my message again, including its
caveats. I have done a lot of *both*, and I learned PEP from a
master---deVaucouleurs. So I know what I am talking about. :)  There is
nothing incorrect in my post.

Michael Newberry

> Using
> standard "public" photometric systems, about the best you can do in terms
> of real external errors in transforming data to the standard system is
> about 7 to 9 mmag rms.  There's quite a lot of single-channel photometry
> that good in the literature (though there's plenty worse), but you will
> have to look very hard to find CCD photometry that is as good as that.
> There's some by Alistair Walker, some by Arne Henden, and some by Peter
> Stetson...and then you have look for a long time before you see anything
> as good.  For instance, not a single photometric study using any HST
> instrument has real external errors better than about 0.02 mag., and
> usually it's in the 0.03 to 0.05 mag. range, a level where those used
> to p.e. data would say you did something wrong or it was cloudy (not
> possible with HST!).  The calibration problems with CCD data are
> substantial, and though well-studied are poorly understood by most
> observers and even more rarely dealt with.
>      There's also an assumption in this discussion that you can't or
aren't
> allowed to look at bright stars with a CCD.  Again, "baloney".  The first
> good demonstration that you could do high-accuracy photometry with a CCD
> that I know was by Walker using the SAAO 1-m telescope, looking at 6th mag
> E-region standards, in the early 80s.  (He showed about 4mmag rms.)
>      It's pretty hard to get better than 0.015 mag. external errors using
> a CCD, irrespective of magnitude, whereas that level of accuracy is pretty
> straightforward single-channel, but where the accuracy does have a strong
> trend with magnitude.
>      I would certainly not encourage anyone to _start_ doing
single-channel
> photometry these days, because the CCD is so much efficient and flexible
> in modes of observing.  But to claim that CCD data is better in accuracy
> when talking about results on a standard system is debatable at best.
>
> \Brian