Re: change of airmass across can be important

[Chris Albertson <chris@216-178-250-236.lax1.phoenixdsl.net>]

Herbert R Johnson wrote:
> 

> Chris, can your controller software report a simple zenith value, and
> make it available for the image file header? By "simple" I mean a single
> value, rather than different values for each camera. These "airmass"
> effects are approximate anyway, so rather than try to guess where each
> camera is "really" pointing, choose an arbitrary point on the mount
> as reference.  You know that it's easier to ignore a line of FITS data
> in the image file, than to ADD it later on.

By request I could likely add any computable value to the header.  As
I know the aim point of the camera, the location of the camera and
the time, Zenith Angle is not hard to compute.  But then anyone could
compute this themselves.  Once the image is processed the aim point
will be exactly known and therefore also the zenith angle.  Getting
air mass from the Zenith angle is not trivial if you consider the
Earth is curved and add in refraction.

Back when I has in the (down looking) IR space business the experts I
worked with would actually divide up the IR pass band and look at 
atmospheric transmission and refraction wave length by wavelength.

I think my policy with headers will be to add anything someone asks for
as long as that person is actually processing data and will use the requested
header line.  But then I publish source code that anyone can do
what they want with

> 
> ....except, if the TIME STAMP of the image exposure is off, or if the location
> of the camera is changed, then it will not be exact. Also, all that info
> may not be in the FITS header. My attitude is, redundancy is a good thing,
> and after 8MB of image data, a few bytes more of header is not excessive.

I'll tell you what.  If someone can't figure out their Lat, Lon to
within a few feet I'll loan then my GPS unit via FedEX.  I doubt I'll have
to do this knowing the list of people Tom is sending cameras to.

Time should be good to well under 0.1 seconds.  This computer and my
Mk IV test bed stay "synched" to "true time" to within tens of milliseconds.
No one should be setting the system time by their wrist watch.

Yes I agree, A few bytes is not big deal.  I've made the code
very easy to modify in this area because I know it will be modified.

> 
> In the end *you* are the one
> implementing the controller software which generates the image data.
> So it's your call anyway. But I would not break a sweat trying to
> determine the EXACT zenith angle for EACH STAR, or even for EACH IMAGE.

Each image,I have to.  I treat every image the same.  If something
is done for one it's done for all images.  For each star, no.  Images
don't have starlists attached.

> The atmosphere is turbulent, there are low and high pressure zones, etc.
> Any of these effects overwhelm the simple assumption of calm, steady,
> uniform temperature air that Richmond's "airmass table" presumes.
> Our colleagues forget the Mark IV will be used in non-photometric
> conditions.

Again back at and old job, some people would get temperature and pressure
and winds aloft profiles from the ground up for the line of sight from the
national weather service.  Call these guys nuts but if you want to do 
something like wide baseline stereo imaging and you care about a few feet
of altitude....

All, said I think a single, simple geometric airmass number for the
frame center is all we need.  Let others do the esoteric stuff later
if they want.
> 
--
   Chris Albertson             home: chris@albertson-home.net
   Redondo Beach, California   work: calbertson@primeadvantage.com