arcsecond spectroscopy

["Glenn E. Allen" <gea@space.mit.edu>]

Hi Folks,

   The flood of Cas A data is spectacular.  
   
   Although I was not at the NRL meeting last month, I am analyzing the data
and want to inform you of what I and a few others here at MIT are trying to
do.  Our goal is to perform spatially-resolved spectroscopy on arcsecond
spatial scales.  Specifically, we have defined a square grid on Cas A.  For
each point separated by two pixels (i.e. 0.984 arcsec) from its immediate
neighbors:

   1. A spectrum is extracted.  (The spectra are not independent from their
      neighbors.)
   
   2. The spectrum is fit using an appropriate spatially-dependent ARF and
      RMF.
   
   3. The best-fit parameters are written to disk.
   
In the end, images of each fitted parameter (e.g. the silicon line flux) are
created.  Since the process is extremely CPU intensive, the fits are
performed on several computers simultaneously.  Nevertheless, we are still
restricted to spectral models that require little CPU.  The shock models,
for example, are too expensive to use.  The model being used at the moment 
includes a bremsstrahlung continuum and several Gaussian lines.  More
physical models may be tried later. 

   If all goes well, Michael Stage will present a poster of some preliminary
results at the COSPAR meeting in Paris.

   Happy analyzing,

Glenn





Glenn E. Allen
Massachusetts Institute of Technology
Center for Space Research
77 Massachusetts Avenue
NE80-6029
Cambridge, MA 02139-4307
Off: 617-253-3875
FAX: 617-253-8084
gea@space.mit.edu
http://space.mit.edu/~gea