Difference between revisions of "Simulation Recipes"

From CASA Guides
Jump to navigationJump to search
Line 14: Line 14:
 
* the GRS resolution of 40" at ~10kpc is 0.04" at 10Mpc, so we should be able to do a simulation of observing at ~0.1-0.2".  The resolution plot ([[File:Beamsummary.png|100px]]) indicates that for ALMA at 100GHz, configuration 20 is appropriate.
 
* the GRS resolution of 40" at ~10kpc is 0.04" at 10Mpc, so we should be able to do a simulation of observing at ~0.1-0.2".  The resolution plot ([[File:Beamsummary.png|100px]]) indicates that for ALMA at 100GHz, configuration 20 is appropriate.
 
* if we intend to set <tt>cell=0.04arcsec</tt> in <tt>simdata</tt>, then the cube needs to be multiplied by  
 
* if we intend to set <tt>cell=0.04arcsec</tt> in <tt>simdata</tt>, then the cube needs to be multiplied by  
2x10<sup>8</sup> * (.04/206265)<sup>2</sup> = 7.9x10<sup>-6</sup> to obtain Jy/pixel.  The cube peaks at ~3K, so we can perform the simulation with <tt>inbright=2.3e-5</tt>, which should yield a peak of ~0.3mJy/bm.
+
2x10<sup>8</sup> * (.04/206265)<sup>2</sup> = 7.9x10<sup>-6</sup> to obtain Jy/pixel.  The cube peaks at ~10K, so we can perform the simulation with <tt>inbright=8e-5</tt>, which should yield a peak of ~1mJy/bm.
  
 
will we be dominated by the noise in the input model?
 
will we be dominated by the noise in the input model?
* input noise ~150mK or S/N~20, so at our scaled intensity, ~0.015 mJy/bm. The [http://www.eso.org/sci/facilities/alma/observing/tools/etc/ exposure time calculator] says that ALMA will achieve 2.6mJy/bm in 2 hours if we keep the input 212m/s channel width (0.075MHz), so the noise in the input model should not affect our results.
+
* input noise ~150mK or S/N~20, so at our scaled intensity, ~0.05 mJy/bm. The [http://www.eso.org/sci/facilities/alma/observing/tools/etc/ exposure time calculator] says that ALMA will achieve 2.5mJy/bm in 2 hours for the input 212m/s channel width (0.075MHz), so the noise in the input model should not affect our results.
 +
* We do have a sensitivity issue though - if we decrease the spectral resolution by a factor of 6 (bin the input channels in some other program - simdata will know how to do that in the future but not yet), and plan for 3 8-hr tracks, then the sensitivity calculator suggests that we'll get <0.25mJy rms, or S/N>10 per beam.  Rather than simulate 3 days of observing, I'll increase inbright by sqrt(3) and simulate one 8 hour track. 
  
 
setup:
 
setup:
 
* the ALMA 12m primary beam is 50" so we'd space a mosaic by 25", but the model cube has 326x357 pixels, or 13 arcsec with our small pixels.  That's a lot smaller than the primary beam, so it doesn't matter much what output image size we ask for.   
 
* the ALMA 12m primary beam is 50" so we'd space a mosaic by 25", but the model cube has 326x357 pixels, or 13 arcsec with our small pixels.  That's a lot smaller than the primary beam, so it doesn't matter much what output image size we ask for.   
* there are 659 channels in the input cube, so we'll keep the same width and nchan.
+
* there are 659 channels in the input cube, but as noted above we want to bin those to 109 channels of 1.2 km/s each.
  
 
here're the simdata inputs : [[File:simdata.n891.txt]]
 
here're the simdata inputs : [[File:simdata.n891.txt]]

Revision as of 17:33, 23 December 2009

Simulating Observations in CASA

Massively Under Construction

(i.e. don't believe what you see here yet, or ask me when it doesn't make sense :)

Nearby edge-on spiral

Roughly modeled after NGC891

  • model data: Milky Way 13CO from the Galactic Ring Survey on the 14m FCRAO
  • units: K - first convert to flux surface brightness

[math]\displaystyle{ \frac{Jy}{Sr} = \frac{4\times 10^{23} k T}{\pi D^2 \Omega} }[/math], where [math]\displaystyle{ \Omega }[/math] is the beam solid angle. At 110GHz the factor is 2x108

now we need to decide if this model data will work at the desired pixel scale

  • the GRS resolution of 40" at ~10kpc is 0.04" at 10Mpc, so we should be able to do a simulation of observing at ~0.1-0.2". The resolution plot (Beamsummary.png) indicates that for ALMA at 100GHz, configuration 20 is appropriate.
  • if we intend to set cell=0.04arcsec in simdata, then the cube needs to be multiplied by

2x108 * (.04/206265)2 = 7.9x10-6 to obtain Jy/pixel. The cube peaks at ~10K, so we can perform the simulation with inbright=8e-5, which should yield a peak of ~1mJy/bm.

will we be dominated by the noise in the input model?

  • input noise ~150mK or S/N~20, so at our scaled intensity, ~0.05 mJy/bm. The exposure time calculator says that ALMA will achieve 2.5mJy/bm in 2 hours for the input 212m/s channel width (0.075MHz), so the noise in the input model should not affect our results.
  • We do have a sensitivity issue though - if we decrease the spectral resolution by a factor of 6 (bin the input channels in some other program - simdata will know how to do that in the future but not yet), and plan for 3 8-hr tracks, then the sensitivity calculator suggests that we'll get <0.25mJy rms, or S/N>10 per beam. Rather than simulate 3 days of observing, I'll increase inbright by sqrt(3) and simulate one 8 hour track.

setup:

  • the ALMA 12m primary beam is 50" so we'd space a mosaic by 25", but the model cube has 326x357 pixels, or 13 arcsec with our small pixels. That's a lot smaller than the primary beam, so it doesn't matter much what output image size we ask for.
  • there are 659 channels in the input cube, but as noted above we want to bin those to 109 channels of 1.2 km/s each.

here're the simdata inputs : File:Simdata.n891.txt

CASA <> execfile("simdata.ngc891.txt")
CASA <> go simdata

here's the cube with the simdata's scaling and World Coordinate System: N891.coord.png
and a spectral profile in the box marked in green
N891.grs-24-cube.coord18-59-59.976-40d00m01.972.png