This ASKAP data release contains spectral-line simulations of a full observation with critically sampled beams. 

This was created in the following way:
* The SKADS S3-SAX simulation (Obreschkow et al 2009) was used as the source catalogue from which a 
  sky model image was created.
* This sky model was used as input into a telescope simulation. This simulation featured:
 o The 2km core of the ASKAP configuration (30 antennas only) 
 o The effect of the phased-array feed was simulated by using 32 idealised beams, spaced in a
   rectangular grid 0.5 degrees apart (the arrangement is a 6x6 square, without the corners). This 
   gives a smaller field of view than the first simulation, but with uniform noise
 o The spectral setup had 1024 channels of 92.5kHz width (5x the width of the ASKAP fine channels), 
   from 1327.39 to 1422.0175 MHz. 
 o The total integration time was 8 hrs, made up of 5sec integrations.
 o The simulating was done separately for each channel.
 o The total size of the measurement sets containing the visibilities is 5.5TB - we are not 
   making these available!
* The image cube was created with the prototype ASKAPsoft imager code. The cube presented is 
  the dirty image, without any preconditioning. The imaging is also done separately for each 
  channel, with the dirty images then being merged to form the complete cube.

A further two cubes are also presented. These have the noise reduced
considerably (the rms on the visiblities was 31.6x less) so that many
more sources are seen. These were produced for the purposes of testing
source-finders. Rather than being the full field, the inner 512x512
subsection of the full cube is presented - this means the noise is
fairly flat across the image. This also reduces the size, hopefully making them
more useful for source-finder testing.

The two cubes are made at different declinations: -45 degrees (to
match the full cube), and -3 degrees. This is to show the difference
in PSF.

Note that these cubes have not been cleaned, so sidelobes are clearly
visible for the bright sources. PSF images are available for both
declinations (along with FITS versions of the first channel for ease-of-use). 

There are also matching model cubes of the same pixel resolution as
the data cubes. Refer to the first simulation release for source
lists.

All processing was done on the NCI National Facility machine vayu.


The data is available on the local ATNF computer system, at 
giant:/DATA/DELPHINUS_2/whi550/ASKAPsimulations/SpectralLine/Apr2010.

We of course would welcome any feedback from your investigations:
please use the ASKAP redmine site or contact one of the ASKAP Project
Scientists.

Note 28/4/2010: An incorrect WCS was noticed on the matching model cubes. This has been fixed.


Contents of directory:

README                                            : This file
ASKAPsim_Spectralline_Apr2010                     : 1536x1536x1024 spectral cube over full field and full spectral range. (CASA)
ASKAPsim_Spectralline_Apr2010_fullfieldsubset     : 1536x1536x100 spectral cube, full spatial field, 100 channels. (FITS)
ASKAPsim_Spectralline_Apr2010_fullspectrumsubset  : 512x512x1024 spectral cube, full spectral range with 1.4deg FoV. (FITS)
ASKAPsim_Spectralline_Apr2010_theoreticalnoise    : 1536x1536x1024 spectral cube showing the theoretical noise (CASA)
ASKAPsim_Spectralline_Apr2010_noise               : 1536x1536x1024 spectral cube showing the noise from a blank field observation (CASA)

ASKAPsim_Spectralline_Apr2010_SF45deg             : 512x512x1024 spectral cube, reduced noise level, declination -45deg. (FITS)
ASKAPsim_Spectralline_Apr2010_SF03deg             : 512x512x1024 spectral cube, reduced noise level, declination -3deg. (FITS)

ASKAPsim_SpectrallineModel_Apr2010_matching       : 1536x1536x1024 sky model, same pixel scale as dirty cube, smoothed with 30arcsec beam. (CASA)
ASKAPsim_SpectrallineModel_Apr2010_matching_fullfieldsubset    : 1536x1536x100 subset of matching sky model. (FITS)
ASKAPsim_SpectrallineModel_Apr2010_matching_fullspectrumsubset : 512x512x1024 subset of matching sky model. (FITS)

ASKAPsim_SPectralline_Apr2010_PSF45deg            : 1536x1536x1024 spectral cube showing the PSF for declination=-45deg (CASA)
ASKAPsim_SPectralline_Apr2010_PSF45deg_ch1        : 1536x1536x1 image showing the PSF for declination=-45deg at the first channel (FITS)
ASKAPsim_SPectralline_Apr2010_PSF03deg            : 1536x1536x1024 spectral cube showing the PSF for declination=-3deg (CASA)
ASKAPsim_SPectralline_Apr2010_PSF03deg_ch1        : 1536x1536x1 image showing the PSF for declination=-3deg at the first channel (FITS)

---

Matthew Whiting
April 2010