Some Talks and Demos of Interest at Astronomical Data Analysis Software & Systems XIV, Pasadena, CA, Oct 17-21, 2004

Talks

• Blind Deconvolution, Tony Chan (UCLA)
  Both the image and psf are recovered from a blurred and noisy image.

• Pan-STARRS, Eugene Magnier, IfA
  • Killer asteroid finder
  • 4 x 1.8 m telescopes, each with 3 deg fov and 1G pixel CCD.
  • "All-Sky" (30,000 sq deg) survey every 7 nights down to 24 mag (5 sigma)
  • Online c. 2008-2009; PS1 prototype Jan 2006
  • Sloan-like filter selection

• GOODS: The Great Observatories Origins Deep Survey
  • HST, Chandra, Spitzer + XMM-Newton
  • Useful for finding galaxies and QSOs for z > 4 and type Ia SNe for 0.2 < z < 2.

• Mars Exploration Rover

• Science with Virtual Observatory tools Paolo Padovani, ESO
  "Doing real science without doing real observations."
  • The next logical step after online catalogs (NED , Simbad), online databases (HEASARC) and central access points (CADC)
  • 1 Tb of new data/night (!?)
  • VO standards and protocols IVOA
  • in situ data analysis (e.g., Aladin)
  • Exercise: Multiwavelength analysis of GOODS data (HDF, CDF + ground-based spectroscopy) to identify QSO 2s Padovani et al.

• Aladin and the VO FAQ
  • AVO interface for displaying and analyzing multiwavelength data, specifically correlating catalog data with astronomical images.
  • Implemented in Java

• The Development of DS9
  • Interesting mostly for mention of need for multiprocessor support.

• SExtractor
  • Source identification tool
  • Has its own optimal filtering algorithm.
  • Used at Saclay for LAT source catalog generation.

"Focus" Demos

• NVO
  • How to make astronomical data and catalogs available using standard VO protocols.
  • Registry Services support publishing of data, querying of specific databases, and harvesting wherein different registries allow information to be passed between them so that a query at one registry may find data at a different one.

• Python Tools for Analysis and Visualization of Astronomical Data
  
  • Tools from Space Telescope Science Institute:
    
    • numarray A replacement for the Numeric package; offers similar array manipulation functionality as matlab or IDL.
    • PyFITS A nice interface for accessing data in FITS files.
    • numdisplay Steer image display using ds9 or ximtool from Python.
    • PyRAF Python replacement for the IRAF command language.
      
  • matplotlib A default plotting package for Python?\\\\
    An example plot:

    from matplotlib.matlab import * def join(x, sign=1): x1 = list(x) x2 = list(sign*x) x1.reverse() x1.extend(x2) return array(x1) func = "x**3 - x + 1" f = eval("lambda x: " + func) x = arange(-2, 8, 0.01) y2 = f(x) indx = where(y2 > 0) x = x[indx] y = sqrt(y2[indx]) x = join(x) y = join(y, -1) figure(1, figsize=(6, 6)) clf() plot(x, y, 'k') xymax = 20 plot([0, 0], [-xymax, xymax], ':b', [-xymax, xymax], [0, 0], ':g') xx = arange(-1, 8) yy = sqrt(f(xx)) indx = where((yy % 1) == 0) plot(xx[indx], yy[indx], "D") xlabel(r'$x$') ylabel(r'$y$') eq = '$y^2 = ' + func.replace('**', '^').replace('*', '') + '$' legend((eq, '$x=0$', '$y=0$')) xlim(-1.5, 2) ylim(-2, 2)

    
    Here is the result:\\\\
    \\\\
    An example of image display using this package:\\\\

    import pyfits from matplotlib.matlab import * dc1_map = pyfits.open("dc1_counts_map.fits") dc1_map.info() im = dc1_map[0].data.copy() clf() figure(1, figsize=(8, 4)) imshow(log10(im[0] + 1e-5))

    
    and the resulting image:\\\\