Revision as of 19:57, 19 July 2010 by Teuben (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search



Unless you have a seperate install of [[1]NEMO], this package can be easily installed using a script in MIRIAD. First grab the CVS source

  cd $MIR/borrow
  cvs -Q co nemo

then compile

  mir.install nemo

There's no support in the new-style install. For those familiar with NEMO's normal install, this is basically identical, except it will borrow the pgplot installation from the one already installed in MIRIAD, instead of keeping a separate one in NEMO.

ASCII tables

ASCII tables are frequently produced by the output of MIRIAD programs. Sometimes non-tabular human readable and needs to be commented out before being piped into the table programs, e.g.

 tabcomment - | tabplot -

A useful program that can be used with CARMA's quality script is meanmed. If it is not installed yet, NEMO has a quick and dirty way that installs most programs using the mknemo command:

 mknemo meanmed
 meanmed --help

This meanmed program is also available within CARMA (at cedarflat, in /opt/rt/bin/meanmed), and although the programs were written independently, they should produce the same results.

FITS files: images

8-bit and 64-bit FITS files cannot be read by MIRIAD's fits program, but sometimes you can use the fitsccd/ccdfits pair to convert, and use fitsedit to edit the header. If that doesn't work, fitshead and scanfits can be used to separare the header and data, edit the header and paste them back. E.g.

  fitsccd moon8.fits - | ccdfits - moon.fits 

for a simple conversion if you don't care about the WCS. If you do, NEMO's ccd format is not rich enough (silly theoreticians) and some tricks are needed:

 fitshead moon8.fits > moon8.h
 scanfits moon.fits moon.d select=data
 $EDITOR moon8.h                                 # at least need to edit BITPIX to -32, the default in fitsccd

now catenate the two, but the header needs to be properly formatted in records of 80 char/line and recordsize a multiple of 2880

FITS files: uv data

Ever wanted to look inside a UV FITS file? If fv doesn't work for you, try fitstsab in NEMO:

  fits in=vis1 out=vis1.fits op=uvout
  fitstab in=vis1.fits                             this will proably show the antenna table
  fitstab in=vis1.fits random=t                    this will produce a lot of output, the Random Table, which are the visibilities
Personal tools