Astro-Mix 1: A VIRUS Display Tool.
  1. An amalgam of HET rendering tools.



An amalgam of rendering tools.

A combination of python, bash script, and OTW (bash+gfortran) tasks have been combined to provide a tool for visualizing the VIRUS components (i.e. IFU and fiber positions) on the sky. Much of this was developed under Python 2.7, and some problems were encountered when an mcs install was attempted (mcs is the HET machine used by resident astronomers for observing). The bottom line is that the argparse module is not available in Python 2.6 (currently installed on mcs). It is possible to add a 2.6 version for argparse, and Francesco Montesano has suggested this document that describes how to install a 2.6 argparse version.

The goal here has been to provide a versatile tool that allows the user to easily visualize the HET field on the sky. Such a tool may be useful during observing, but I am finding it extremely useful for developing software tools for VIRUS Astrometry Commissioning. A local (working) version of this document, can be reached HERE. As discussed in this plan, the tasks of going from sky-to-FP and FP-to-sky (FP = focal plane) are complicated slightly by characteristics of the HET (alt-az mount, partial rho motion, etc..) and the possibility of nonradial field distortion introduced by the WFC (wide field corrector). All of these operations will ultimately be handled by simple command-line calls, but I believe the ability to visualize the results expected from these tasks will be critical to our commissioning success.

I am primarily concerned here with the VIRUS IFU properties, but we can also over-plot a simple North-East legend or plate-scale legend (as well as parallactic angle direction), the acquisition camear (ACAM) field, and the guide probe annuli in the HET focal plane assembly (FPA). These capabilities will be useful in commissioning and future general work with the HET. One goal is to make the tool versatile with respect to the form of the WCS installed in the header of the astronomical image we are overplotting to. The tool should be able to use the old-style WCS (CROTA2, CDELT2, etc...) as well as the more newly adopted CD-matrix form. Finally, we must be able to make use of the new FITS conventions for XY distortion correction (SIP = Simple Image Polynomial). All of these concerns are being addressed in the current tool set.

The figure below demonstrates some of our capabilities as of July 2015.

The VIRUS IHMP outline, as well as the IFU positions and some fiber sets were overplotted above on an image (with FITS WCS) of NGC3379 using the two commands:
 
virus_ds9 n3379_B.fits 30
virus_mark_all 25.0 1000 900
An explanation of the command strings above is not warranted now, but is should be note that the virus_ds9 task simply takes the image name and a dynamic range value for the scaling (low for linear views close to sky, high for linear- of log- scaling in the areas of bright sources like galaxies). In the case above, a simple box-view is used to represent each IFU, but in the cases of IFUs 068, 024, and 075 we have plotted and labeled each of the 448 fibers.

This active ds9 display takes less than 1 second to render, and enables the user to interactively zoom into areas of interest (as in the figure below). For reference, it is useful to know that plotting the labeled fibers for all 78 VIRUS IFUs on this image requires about 12 seconds. The graphic below was made in this same ds9 run, where the user has zoomed to a particular IFU to a level that allows the individual fibers (and their identification numbers) to be viewed easily.

Here we have zoomed into a portion of IFU 068. The user can move through the field and identify fibers that coincide with sources of interest in the image. As an example, the bright source above should yield a bright Cure-extracted spectrum with fiber #261 in IFU068.


The envisioned final goal: The examples above use a WCS-calibrated CCD image from the McDonald PFC taken some time ago. We'll frequently use DSS or SDSS images for similar displays. However, by building a paradigm that uses well-established WCS and rendering approaches, our inputs should eventually be 2-d images from VIRUS itself. Images from Cure made with apimage (or "Son of" apimage) will contain the full WCS header including the SIP distortion terms derived for our VIRUS+WFC+HET system.




back to calling page