Demo Science

Preliminary Plan for making quality EPO images

To increase public awareness and appreciation of the EVLA, the NRAO will produce a series of high-quality EPO images that include (but are not limited to) EVLA data. Like the best Hubble Heritage images, they should be visually beautiful and inspiring to the general public, even before they read the explanatory captions. Secondary goals include (1) demonstrating unique capabilities of the EVLA (e.g., higher angular resolution than HST) or radio astronomy (e.g., detecting stars hidden behind dust) and (2) scientific results interesting to professional astronomers.

Having suitable images at the formal dedication of the EVLA in 2011 imposes time constraints on both the beginning and the end of this long-term project because the various proposed images benefit from data taken in all four EVLA configurations:

  • The necessary D-array observations must be made before the array move now scheduled for July 12. The target sources, observing frequencies, and mosaic patterns must be chosen and the scheduling blocks written. This can happen only if an expert observer is willing to take responsibility for each image and we provide sufficient observing time, possibly by extending the D configuration slightly beyond the currently scheduled July 12. We already have “volunteers” for most of the observations suggested below but need to get firm commitments. Fortunately, some of the projects suggested in Bob Dickman’s May 17 memo have been tentatively scheduled for “commissioning” time, and some compact objects may not need D-array data.
  • The EVLA configuration schedule and dedication date should be coordinated to ensure that the A-array data can be taken and reduced in time. Because most EVLA antennas are too distant to see from the center of the A array, it may be best to delay the dedication until the start of the next D configuration in early fall of 2011. Note that the weather at the VLA can be very unpleasant in spring (high winds and blowing dust) and summer (violent thunderstorms) but is reliably good in the fall. A fall dedication also provides some “contingency” time in case commissioning doesn’t go as fast as hoped.

The most effective radio images usually have (1) high spatial dynamic range (ratio of source size to angular resolution and/or source detail size), (2) fairly high intensity dynamic range (the dynamic range of photographic paper is about 100:1, computer screens are slightly better), and (3) multiple data types (e.g., several radio frequencies, a range of Doppler velocities, images having widely differing angular resolution, and layered IR/optical/X-ray images) that can yield multicolor instead of monochrome images. The images need not be confined to static 2-D photographic prints or computer screens. For example, spectroscopic data cubes showing concentric spherical shells blown off by the carbon star IR +10216 can be converted to a 3-D color image that can be rotated to reveal depth or shown directly on a 3-D television.

Unlike filled-aperture telescopes, the reconfigurable EVLA is able to trade off angular resolution for brightness sensitivity, so the best radio sources (those covering a large number of synthesized beams with the largest flux per synthesized beam) tend to be those having the highest total flux densities per primary beam, with total angular size being only a secondary parameter. Most such sources are well known and already have good images from other wavebands. The appendix gives a preliminary list of the best candidates, suggested observations, and potential observers. Asterisks indicate the “must do” (in my preliminary opinion) sources. That preliminary list is too long and should be pared down significantly to minimize disruption of the already-full EVLA schedule of OSRO, RSRO, ECSO, and engineering/test observations and not overload the already busy commissioning staff. On a time scale of two weeks or so, we need to get a firmer list of proposed D-array observations and the required observing times from the observers so that a subset can be approved and scheduled.

We note that radio astronomers have rarely used color well—coding brightness with color bands in AIPS is not adequate for the proposed EPO images—and have limited experience with the best software for combining image layers effectively. EPO experts from the NRAO and outside consultants from universities and the STScI will have to take astronomical images in FITS format and import them to specialized image-processing packages (e.g., Photoshop) that can layer images effectively. We will work together to bridge any communications gaps between astronomers and EPO image experts so that astronomers can choose the sources and observations most suitable for producing the best EPO images.

List of Projects (to be prioritized)

Sources needing D-array data, proposed observations, and responsible observers:

  • Cas A = 3C 461 supernova remnant. 103 Jy at 5 GHz, angular diameter 3 arcmin, beautiful array of shock structures indicating expansion. Rick Perley originally suggested making a 10 X 10 mosaic at 34 GHz mainly because he thought we would only have D array data before the EVLA dedication. Instead, we might complement the existing 1.4 GHz and 5 GHz images with an 18 GHz image based on data from the B,C, and D configurations only to make a pseudocolor image reflecting spectral changes in shocks. Thus Cas A could make a good image even without the A configuration. Unfortunately, it is not clear that the small spectral-index differences within Cas A would make a good color image (Larry Rudnick is skeptical); we will just have to try it and hope. Rick Perley has agreed to do this and has an observing plan for the D-array mosaic pattern.
  • Crab Nebula = 3C 144 supernova remnant: 700 Jy at 5 GHz, angular diameter 7 arcmin. Unlike Cas A, this plerion has beautiful X-ray jets and rings imaged by Chandra, a good HST optical image, likely radio spectral gradients, and a pulsar visible in low-frequency radio images. Rick Perley has agreed to make the D-array observations. Other Socorro supernova experts (e.g., Dale Frail, Miller Goss, visitor Dave Green, Sanjay, …) may be interested in taking part.
  • IC 443 (the “Jellyfish nebula) = 3C 157 supernova remnant. 100 Jy at 5 GHz. A beautiful HST image exists, and the flat radio spectrum makes high-frequency imaging easier. However, its large (50 arcmin) size requires mosaicing even at L band, so this source may have to be deferred because it will require too much work.
  • Tycho supernova remnant = 3C 10. Large (17 arcmin) shell worth considering.
  • Cyg A = 3C 405 radio galaxy. 400 Jy at 5 GHz, 2 arcmin lobe separation, jets, lots of structure in lobes. The existing 8 GHz A-array image could be complemented by a 43 GHz image based on D, C, and B array data. The strong spectral gradients in this source would make a good color image. Rick Perley and Chris Carilli will collaborate on this image.
  • Vir A = 3C 274 nearby radio galaxy. 70 Jy at 5 GHz. There is an excellent HST image of the host galaxy M87 and the optical synchrotron jet. The EVLA could map the jet at comparable resolution at 30 GHz with the D, C, and B configurations only. If A-array data can be added, the EVLA image of the jet will have higher resolution than the HST image, a good comparison for demonstrating the power of the EVLA. D- and B-array L-band images show the radio halo well and can be superimposed on the high-resolution image with Photoshop. There is a complementary Chandra image. Frazer Owen has agreed to do this galaxy and already has some of the low-resolution L-band data.
  • 3C 353 low-redshift radio galaxy. 20 Jy at 5 GHz, 5 arcmin total extent. Striking filamentary structure visible in Alan Bridle’s X-band A-array image at http://www.cv.nrao.edu/~abridle/3c353xhi.htm Rick Perley is willing to plan the D-array observations at least; Alan Bridle may take over.
  • Her A = 3C 348 low-redshift radio galaxy. 12 Jy at 5 GHz, 200 arcsec total extent. Unique radio rings (Dreher & Feigelson 1984, Nature, 308, 48). Rick Perley is will to plan the D-array observations at least; Alan Bridle may take over.
  • M82 = NGC 3034 starburst galaxy. 8 Jy at 1.4 GHz. There is an excellent HST image of its superwinds. Make a modest-resolution L-band continuum image of synchrotron radiation in the ISM and superwinds plus a high-resolution (X band A array) image of the optically obscured young supernova remnants in the starburst powering this galaxy. It and NGC 253 could be imaged in OH absorption as well. Joshua Marvil is a resident predoc doing this galaxy for his PhD thesis, so many of the observations are already approved. This galaxy and NGC 253 Josh should lead the M82 group that may include Juergen Ott, Adam Leroy, Jim Condon, and Brian Kent.
  • NGC 253 starburst galaxy with outflows. 6 Jy at 1.4 GHz. This large southern galaxy requires 3 L-band pointings in the DnC and CnB configurations. HI, OH, RRL, continuum. Why make new HI images, instead of just using HI from THINGS galaxies? Do RRLs show anything that more easily imaged free-free continuum does not? Proposed by Juergen Ott.
  • NGC 1068 starburst/Seyfert galaxy. 5 Jy at 1.4 GHz. Radio core and jets plus starburst. See the NASA image of the day X-ray/HST/VLA image at Error: can't fetch image from 'http://en.wikipedia.org/wiki/File:NGC_1068_Composite.jpg': 404 Not Found L- and C-band continuum images, starting with the D array. Observers: Ott, Leroy, Condon, Kent?
  • NGC 6946 face-on starburst “fireworks” galaxy. 1.4 Jy at 1.4 GHz. Make a low-resolution L-band image of the synchrotron radiation plus a high-resolution high-frequency (e.g., B array at 30 GHz) image of the HII regions locating star formation, superimpose on HST image. Ott, Leroy, Kent?
  • M51 = NGC 5194/5 1.5 Jy. Face-on spiral pair, similar to NGC 6946. One of these should be done; I slightly favor M51. Ott, Leroy, Kent?
  • The best continuum-weak (L band flux < 1 Jy) galaxies M74 = NGC 628, NGC 1097, NGC 1232, NGC 1300, NGC 2903, and NGC 4038/9 in HI, OH, and radio recombination lines? Is OH practical, or will it take too much telescope time? Why make new HI images, instead of just using HI from THINGS galaxies? Do RRLs show anything that more easily imaged free-free continuum does not? We should probably do at most one of these galaxies. Juergen Ott and Adam Leroy
  • Quasars: The radio/optical jet of 3C 273 is much less impressive than the M87 = 3C 274 jet. 3C 175 and 3C 334 have one-sided jets indicating relativistic motions; see http://www.cv.nrao.edu/~abridle/quasars.htm Map one or both of 3C 175 and 3C 334 at two frequencies with scaled arrays? Alan Bridle, Rick Perley?
  • Rosette Nebula = 3C 171. 300 Jy at 1 GHz, 1.5 deg diameter (so will need GBT zero-spacing data). Spectacular HII region, beautiful HST image. L-band continuum plus high-resolution 43 GHz images of stars? Claire Chandler will think about it. D array mosaic; too much work?
  • Eagle nebula. Overlay EVLA images on the famous HST “pillars of creation” image to reveal embedded stars. C band with possible K and Q band continuum plus RRLs and ammonia for any embedded/entrained protostars. Proposed by Joe McMullin.
  • IRC 10216 carbon-rich AGB star with expanding shells up to 1 arcmin in diameter. An EVLA spectral sweep would allow the construction of a 3-D image showing several concentric spherical shells that could be shown on a 3-D TV or by rotating a 2-D slice. D-array only. Potentially unique “image” definitely worth pursuing. Assign longer integration times for higher signal-to-noise ratios? Proposed by Mark Claussen.

Possible sources that don’t need any D-array data, so we don’t have to decide right away:
  • NGC 7027 planetary nebula. 5 Jy at 43 GHz, 7 arcsec diameter. HST infrared and optical images. Observe at 18, 30 and/or 43 GHz with D, C, and B arrays for matching resolution and detail. See Ziljstra, van Hoof, & Perley 2008 ApJ, 681, 1296.
  • SS 433 is the canonical precessing jet source with a corkscrew structure extending about 10 arcsec. The jet has a steep spectrum, so for maximum signal, we should observe in C-band where the integrated flux density (jet + core) is 0.5–0.8 Jy, with the core dominating the total flux (0.3–0.5 Jy/beam). A-configuration observations would be required for ~0.5 arcsec resolution. Possible overlays include the resolved X-ray emission downstream in the jets, or optical images. James Miller-Jones
  • An “EVLA deep field” in two colors. About five square degrees in the Spitzer First-Look Survey (FLS) area have been imaged in L-band continuum by the VLA B array with 5 arcsec resolution and 23 mJy/beam rms noise. The EVLA could make a mosaic covering about 30 arcmin ´30 arcmin at C band with the same resolution and 10 mJy/beam rms in 10 hours for matching sensitivity on normal-spectrum sources. The main visual feature here would be having matching deep images at two frequencies so a color image could be made.

Demo Science Tracking sheet

-- JosephMcMullin - 22 Jun 2010
Topic revision: r2 - 2010-12-02, JosephMcMullin
This site is powered by FoswikiCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding NRAO Public Wiki? Send feedback