Документ взят из кэша поисковой машины. Адрес
оригинального документа
: http://www.naic.edu/~ages/gridzilla.html
Дата изменения: Mon Nov 29 01:58:49 2004
Дата индексирования: Mon Oct 1 21:46:08 2012
Кодировка:
Поисковые слова: universe
The GUI has changed slightly since the AUDS webpage was written. You
will notice two differences:
The addition of a new Data
Filtering section
The log is now attached to the rest of the GUI
The Data filtering section was
introduced to deal with massive RFI from Punta Salinas and Remy. It is
not necessary to apply these filters to our data, so set the values to
include all your data. IF YOU DO NOT YOU WILL FIND THAT ANY
SOURCE ABOVE 50 mJy WILL BE CLIPPED.
Values that seem to be working so far are Tsys range: 0-100000 and Data
range: -1:10.
The Search path is used to
locate all your LIVEDATA processed SDFITS files. As you can see you can
specify multiple paths.
Highlight all the files that
you want to grid by click and dragging over the files you want.
Pressing CTRL is useful for picking individual files out of sequence.
Image centre and extent
sets the pixel size in the final image. AUDS default sets the y-scale
to 5.2 arcmin, so be sure to change this to 1.
FITS spectral type can be set
to your choice. I have been gridding up in frequency but you can
equally choose to grid up the data in velocity (optical or radio).
Gridding Statistic should be
set to WGTMED.
Set the output directory to be /share/ages/drift_files/DATE (where DATE
is of the form YYYMMDD). If this
directory has not already been created, you must create it before
processing. Gridzilla does NOT perform a validity check on the
directory name, if it is not there Gridzilla will fail.
The result of this processing is a single FITS binary file containing a
3-D dataset called a datacube with axes: RA, dec, frequency/velocity.
If, while gridding, MOFONGO cannot hold all the data in memory at once,
then it produces a number of smaller datacubelets. These cubelets have
the same RA and dec ranges, but limited frequency/velocity ranges. When
these cubes are concatenated they will produce a single FITS binary
file, with the entire frequency/velocity range. There is a special
routine that can do this concatenation called build_cube.csh. To run it
change to the directory containing the datacubelets and type:
% build_cube.csh filebasename
(where filebasenameis the output filename you chose in Gridzilla
without the .fits extension). As well as creating a single FITS file
called filebasename.fits, a number
of *.mir directories are also created. This is because miriad
is used to do the concatenation. These directories can be safely
deleted once the final FITS cube is produced.
This webpage is maintained by Robbie Auld. Page last
updated on 28/11/2004
