Astr384 Class Projects, Spring 2008

Colliding Galaxies UGC 816 and 813, Kathy Hoogeboom

Galaxy collisions and tidal interactions represent the largest and some of the most important fireworks in the universe. While they create many of the interesting shapes we observe in galaxies, they are also essential to describing galaxy evolution, bursts of star formation, and the triggering and fuelling of active galactic nuclei (Condon). Watching interactions take place can also teach much about the structure and stability of galaxies (Struck).

Unfortunately, galaxy collisions happen very slowly compared to human timescales, so observing collisions as they progress is out of the question. Instead we turn to studies of a broad sample of colliding galaxies, catching them in various stages of the process. Putting together the wide variety of observed phenomena in collisions then provides a sort of frame-by-frame video of the collisional evolution, albeit with a rather mixed up set of frames.

This colliding galaxy pair, UGC 813 and 816, presents a unique and particularly interesting case. It is referred to as a "taffy" pair because there is a bridge of continuum radio emission extending between them, thought to result from magnetic field lines stretched between the galaxies following entanglement by a collision (Jarrett). High-energy electrons torn out of the galaxies during the collision circle around the field lines, emitting synchrotron radiation. Condon et al. estimate that this collision happened not so long ago -- a mere 50 million years.

This phenomenon is not common; only one other taffy pair is currently known: UGC 12914/12915, first spotted in 1990 (Cowen). It requires very specific collisional circumstances. First the galaxies must collide head-on so as to achieve significant overlap of molecular disks; furthermore, just the right amount of time since collision must have passed before our observations -- too little and the galaxies will not yet be far enough apart to resolve a bridge between them, too much and the synchrotron losses will have made the bridge undetectably faint at centimeter wavelengths. The galaxies and the bridge must also have the right orientation to our line of sight to be observable (Condon). Condon et al. estimate that about one percent of all spiral galaxies should be part of "taffy" pairs, but only half are far enough apart to see the bridge, and half again are at the correct orientation to be adequately separated in the plane of the sky.

We observed this colliding galaxy pair both at optical wavelengths using Calvin College's remote-operated telescope in Rehoboth, N.M., and in radio using the VLA (Very Large Array), also located in New Mexico.

Radio observation

Comparing optical (left) with radio (right) data of precisely the same field often tells a more complete story than either alone. Here many objects that appear in optical are not present in radio and vice versa. Particularly important here is the bridge of radio emission extending between colliding galaxies UGC 816 (left in the galaxy pair) and 813. Also apparent is radio emission from the nearby galaxy to the right and a bright radio source in the top right which does not appear in the optical data.

Radio observation is quite different from optical. Many of the objects we are used to seeing in optical images, particularly stars, do not appear at radio wavelengths. Other objects only appear in radio, including the bridge between our colliding galaxies. Observation has fewer limits without a specific field of view or the necessity for dark, clear skies that optical observing requires. Calibration has some different requirements. We interleaved observations of a "phase calibrator" (a nearby point-like source) with observations of our object, and tacked observation of a "flux calibrator" (one of a few sources with very well-known flux, or brightness) at the end of our observing run. These are used by AIPS (Astronomy Image Processing Software) to calibrate the data before the program uses Fourier transforms of the interference patterns from the different antennae to create an image like that shown above. We can also use the data to create a contour plot of the field.

Cont peak flux = 3.7183E-02 JY/Beam; Levels = 6.015E-05 (2, 4, 8, 16, ... 512, 1024)

Here the "taffy" of radio contours extending between the colliding galaxies is quite apparent, a clear indication of a violent interaction between UGC 813 and 816. The radio contours may become more meaningful when overlayed on the optical image. This will highlight the radio emission coming from optically dark areas.

Note how the contours clearly show radio emission stretched between the colliding galaxies in a dark part of the optical image. They also indicate brighter radio emission on the lower edge of the radio bridge, just as shown in the image above. This shows 10 contours ranging from 2E-04 to 8E-3.

Optical observation

Optical images also have much to tell about violent galactic collisions. The long tidal tails extending from either side of UGC 816 (on the left) are often characteristic for the gravitational interaction of galaxies near collision. There is also some variation in colour across the central region of UGC 813 (on the right). The lower left part looks slightly redder than the upper right.

Since the redder region also seems to be the side from which the brightest radio emission is coming from, it is probably the galaxy's core, perhaps reddened by dust. Blue often indicates a burst of star formation, something which would be expected to accompany the disturbances caused by such violent gravitational interaction.

By comparison to a number of standard stars whose colour magnitudes are known, the colour variation can be quantified. From the NOMAD catalog of stars I found four stars near UGC 813/816 with known B and R magnitudes to use as standard stars.

For my set of blue images I defined the magnitude of the reference star, then used the photometry function in MaximDL to find the magnitudes in blue for the three check stars and for each side of UGC 813. The same could then be done for the set of red images. Because I'm looking at two different parts of an extended object, the aperture used by MaximDL to define the object had to be set very carefully so as to select out only a specific portion of the galaxy while avoiding any other part of the extended object or any background stars in the annulus Maxim uses to define the background. I used an aperture radius of 3, gap width of 20 and annulus thickness of 5.

I used the averages of the B and R magnitudes found by Maxim for each image to calculate the B-R colour for the two parts of UGC 813. Uncertainty was estimated using the standard deviation of the values. For the upper right I find a B-R colour of 2.26 +/- 0.20; for the lower left, 2.48 +/- 0.12. The lower number for the upper right does indicate that the region is bluer, as expected. However, these are not statistically different from each other -- the uncertainties are too high. Furthermore, the calculations by Maxim of the check stars' magnitudes did not match catalog values very well. This may be due in part to bright stars near the check stars leaking some light into the background annulus, something which was likely not a problem for the reference star from which the galaxy magnitudes are found.

References:
Condon, J.J., G. Helou and T.H. Jarrett. "A Second 'Taffy' Pair." The Astronomical Journal, vol. 123, p. 1881-1891. April 2002.

Cowen, R. "Heavenly taffy: galaxies in collision." BNET Science News. 11 May 2002.

Jarrett, T.H., G. Helou, D. Van Buren, E. Valjavec and J.J. Condon. "A Near- and Mid-infrared Study of the Interacting Galaxy Pair UGC 12914/12915: 'Taffy.'" The Astronomical Journal, vol. 118, p. 2132-2147. November 1999.

Struck, Curtis. "Galaxy Collisions." Physics Reports, vol. 321, p. 1-137. November 1999.