View source for Protoplanetary Disk Simulation (CASA 4.0)

{{Simulations Intro 4.0}}
[[Category: Simulations]]

* '''This is an advanced simulation tutorial.  New users are recommended to begin with the [[Simulation Guide for New Users (CASA 4.0)]].'''
* '''This guide is applicable to CASA version 4.0.  For older versions of CASA see [[PPdisk_simdata_(CASA_3.4)]].'''
* '''To create a script of the Python code on this page see [[Extracting scripts from these tutorials]].'''

== Data ==

For this CASA Guide we will use a protoplanetary disk model from S. Wolf.  Get the data [http://www.cv.nrao.edu/~awootten/mmaimcal/ppdisk672_GHz_50pc.fits here]. See the [http://casaguides.nrao.edu/index.php?title=Sim_Inputs Simulation Inputs] page for this and other sample images.  ''If you use this FITS data for anything more than learning CASA, please cite [http://adsabs.harvard.edu/abs/2005ApJ...619.1114W Wolf & D'Angelo 2005].''

== Script with Explanation ==

Set '''simobserve''' as the current task and reset all parameters.

<source lang="python">
# In CASA
default("simobserve")
</source>
   
Review the image coordinate system using task {{imhead}}.

<source lang="python">
# This reports image header parameters in the Log Messages window
imhead("ppdisk672_GHz_50pc.fits")
</source>  

We now use the '''ia''' (image analysis) and '''qa''' (units and quantities) tools from the CASA Toolkit to find the image center.  In comparison to tasks, tools are a more advanced way of manipulating data in CASA.  You can learn more about tools using the [http://casa.nrao.edu/docs/CasaRef/CasaRef.html tool reference manual].

When data are being manipulated with tools the data file must be explicitly opened and closed.

<source lang="python">
# In CASA
ia.open("ppdisk672_GHz_50pc.fits")
</source>

Next, get the right ascension and declination of the image center.  We get the number of pixels along each axis using '''ia.shape'''.  Then, we get the RA and Dec values for the center pixel using '''ia.toworld'''.

<source lang="python">
# In CASA
axesLength = ia.shape()
# Divide the first two elements of axesLength by 2.
center_pixel = [ x / 2.0 for x in axesLength[:2] ]
# Feed center_pixel to ia.toworld and and save the RA and Dec to ra_radians and dec_radians
(ra_radians, dec_radians) = ia.toworld( center_pixel )['numeric'][:2]
ia.close()
</source>

Use the '''qa''' tool to convert the image center from radians to sexagesimal coordinates.

<source lang="python">
ra_hms  = qa.formxxx(str(ra_radians)+"rad",format='hms',prec=5)
dec_dms = qa.formxxx(str(dec_radians)+"rad",format='dms',prec=5)
</source>

Let's call our project "psim2".  This defines the root prefix for any output files from simobserve.
<source lang="python">
project = "psim2"
</source>

We'll set '''skymodel''' to the FITS file downloaded, above, and leave all '''skymodel''' subparameters at their default values.  simobserve will create CASA image <tt>psim2.skymodel</tt>.
<source lang="python">
skymodel = "ppdisk672_GHz_50pc.fits"
</source>  
 
We will specify the sky position for the center of the observation and set the map size to the size of the model image.  Since the model image is 2/3 arcseconds across, we should only need one pointing.  In this case, '''pointingspacing''' and '''maptype''' can be left at their default values.

<source lang="python">
setpointings       =  True
direction          =  "J2000 18h00m00.031s -22d59m59.6s"
mapsize            =  "0.76arcsec"
</source> 

We do want to simulate an interferometric observation, so we set '''obsmode''' accordingly.  We'll set '''totaltime''' to a 20-minute snapshot observation.

<source lang="python">
obsmode            =  "int"
totaltime          =  "1200s"
</source>

We want to use the appropriate antenna configuration for the desired angular resolution.  Configuration 20, <tt>alma.out20.cfg</tt>, is the largest "compact" configuration.

<source lang="python">
antennalist        =  "alma.out20.cfg"
</source>

Now run simobserve.

<source lang="python">
simobserve()
</source>

Now that we've simulated the visibility measurements, we want to generate an image from the simulated data.  '''simanalyze''' makes this easy.  We begin by setting '''project''' to the same prefix used in simobserve and setting '''image''' to True.

<source lang="python">
default ("simanalyze")
project = "psim2"
image = True
</source>

We set '''modelimage''' to use the input FITS image when cleaning the simulated visibilities.  We set the image size to 192 pixels square.

<source lang="python">
modelimage = "ppdisk672_GHz_50pc.fits"
vis = project + ".alma.out20.ms"
imsize = [192, 192]  
</source>

Specify the number of iterations for cleaning, with proper threshold and weighting.

<source lang="python">
niter = 10000
threshold = "1e-7Jy"
weighting = "natural"    
</source>

We'd like to calculate a difference and fidelity image, but we don't need to see the uv coverage.

<source lang="python">
analyze = True  
showuv = False
showresidual = True  
showconvolved = True
</source>

Plot both to the screen and PNG files with lots of messages.

<source lang="python">
graphics = "both"
verbose = True
overwrite = True
</source>

Run simanalyze.

<source lang="python">
simanalyze()
</source>

==Simulation Output==

{| cellspacing=2
|Input:<br> [[File:psim2-4.0.alma.out20.skymodel.png|300px]]
|simobserve:<br> [[File:psim2-4.0.alma.out20.observe.png|300px]]
|-
|simanalyze image:<br> [[File:psim2-4.0.alma.out20.image.png|300px]]
|simanalyze output:<br> [[File:psim2-4.0.alma.out20.analysis.png|300px]]
|}

{{Simulations Intro 4.0}}
{{Checked 4.0.0}}