[daip] Re: MOMNT questions

Eric Greisen egreisen at cv3.cv.nrao.edu
Mon May 5 15:52:29 EDT 2003


Veera Boonyasait writes:

I am unfamiliar with MOMNT in any detail - I did not write it and have
only reluctantly worked on it occasionally.  REading the code however,
I think I can answer your questions

 >    There are a few questions I have in regards to the task MOMNT 
 > in the original AIPS. I am trying to calculate the amount of 
 > RMS noise level contribution to the surface density map generated
 > by the task. Typically, I use the OPTYPE 'HG' with CELLSIZE= 3,2.
 > What I would like to ask is if MOMNT applies only the FLUX cutoff
 > method or does it also incorporate the window method as well? If
 > not, how does the task recognize one-channel noise spikes? The
 > EXPLAIN file does not describe whether a signal must be contiguous 
 > for a number of channels before being integrated. I see that the 

    There is no windowing - simply the flux cutoffs applied to the
smoothed image.  A one-channel spike would be summed into the
moments.

 > function moments in aips++ has the window method but I have been 
 > reluctant to use the program (aips++) because the version we have 
 > has been very reluctant in reading some of my FITS files which are 
 > easily recognized by AIPS. 

     The aips++ package has many problems and I would not want to
depend on it - particularly in an area that may never have been used
to actually produce science.

 > 
 > Lastly, the Gaussian kernel that is specified is a bit unclear to 
 > me. If the original beamsize of my data is 20", then does MOMNT 
 > convolve the data cube with a 40" beam to make the mask?
 > 
C                                 SPATIAL GAUSS
      NXY = MIN(((IROUND(GSIG)*2)/2)*2+1, 11)
      NXY2 = NXY/2+1
      Z = -4.0*LOG(2.0)/(GSIG*GSIG)
      S = 0.0
      DO 20 I=1,NXY
         X = NXY2-I
         KXY(I) = EXP(X*X*Z)
         S = S + KXY(I)
 20      CONTINUE

where I is in pixels.  The GSIG is the full width at half maximum in
pixels, not something related to arc seconds wrt the Clean beam size
or some such.

Eric Greisen



More information about the Daip mailing list