Author: Arne A. Henden Date: 970516 Revision: #0 970516 Revision: #1 020918 Revision: #2 050922 Key Words: documentation, catalogs
Tom asked me to come up with a file naming convention for the TASS files. Here is my proposal. I don't consider it 'fixed in stone' yet, so if someone has an objection to any part, let me know and I'll try to accomodate them. It has to be a pretty strong objection, though, so be sure to explain yourself thoroughly.
TASS data files will have a name in the form:
abcdyyyy.xxx
where
a = site
b = camera
c = processing status
d = reserved
yyyy = julian date - 2450000
xxx = day fraction
At present, there will be no difference between upper and lower case ASCII. This is a similar format to tm3get to retain continuity. I also retained the 8.3 format for compatibility with DOS and the ISO9660 CDROM format.
The Julian Date is not heliocentric-corrected. Four digits gives 9999 days, and starting at 2450000 sets the initial date to October 10, 1995. The day fraction extension works for Mark III sites, as it gives 1.4minute granularity. For Mark IV systems, Tom envisions 100 second exposures with some dead time in resetting the drive -- so let's call the time between exposures to be 2 minutes. So the day fraction will still work in this mode. However, since a shutter is present, one could envision much shorter exposures to work with brighter stars, and there you might use a running file number (000-999) instead of day fraction. The above format should still work with this scheme.
The site character is A-Z, with the following sites currently defined:
A Norman Molhart B Glenn Gombert C Michael Richmond D Mike Gutzwiller E Paul Rybski F Ron Wickersham G Nick Beser H Tom Droege (TOM1) I Tom Droege (TOM2) J Tom Droege (TOM3) K Robert Creager L Tom Droege (TOM4) M Tom Droege (TOM5)This gives 26 possible sites. We can extend that by including numerals after the alphabet is used up.
The camera character is 0-9 initially, with the option of extending it into the alphabet if needed. For the Mark III sites, cameras are designated 0,1,2. For Mark IV sites, cameras are designated 3,4,5. If more than one camera system is present at a site, then use a number beyond 5. I just wanted to hardcode the two most common configurations.
All unused or reserved characters should be the underscore '_'. The basic file name has one reserved character. My gut feeling is that it will be needed to extend the JD to 4 fractional digits, but could have other uses as well.
The processing character is A-Z, and the following characters are predefined:
A extracted starlist using STAR B extracted starlist using SEXTRACTOR C collated starlist D collated starlist with differential photometry E extracted starlist using DAOPHOT F extracted starlist using Henden's EXTRACT G extracted starlist ala Richmond P processed (flatfielded/darksubtracted) FITS files R raw FITS files T transformed, collated starlist
The extracted starlists are in whatever format the individual programmer desires. I recommend the use of header lines to identify the program and program version used for the extraction (since file formats seem to change, and I need something inside of the file to identify that format for the collation process), and the headings for each column in the file. I highly recommend using a '#' character as the first character of any header line, and including the Julian Date for each star in the list.
My processing uses several intermediate steps, which is why there are so many collated starlist types listed above. This part of the naming convention is definitely open to discussion, though I recommend using some of my reductions as an initial schema prior to the AAS meeting, and then deciding whether to keep that processing style later.
The base collated starlist is a file similar to the one created by my COLLATE program, which takes input from all three cameras, identifies common stars from the three cameras and outputs a file where all data for each star is adjacent. The Julian Date for this file is the earliest JD of the files used to create the collated list.
The transformed, collated starlist is one where the photometric transformation into the standard Johnson system has been performed. This would be similar to my TRANSFORM program. Again, the Julian Date is for the earliest frame in the collated list.
The collated, differential photometry starlist is one that has been transformed into the standard system, and then each star is calibrated versus the master photometric starlist to place them on a consistent zero point. This removes any sky variations that might occur, and enables the use of lousy nights. This is the final output in my mind and would then be used to search for variables. The file format would be similar to my DIFCAL program.
I think this naming handles most of the common needs for TASS files. Some examples:
(1) a file from camera0 of the Dayton TASS triplet, taken on
JD 2450541.369, reducing using MG's STAR program:
b0a_0541.369
(2) A file from Tom Droege's triplet, collated and transformed
using Henden's software, with the first file in the collation
taken on JD 2450384.001:
h_t_0384.001
(3) A raw fits file from Michael Richmond's triplet:
C2R_0570.345
Back to Index of Technical Notes
Back to TASS home page