TN 0050: Analysis of Mark IV image in Virgo

Michael Richmond
Apr 4, 1999
Revision: #1 990404
Key Words: astrometry photometry methods

Tom Droege distributed a CD-Rom with many images taken through the prototype Mark IV camera. The images were collected on March 17, 1999, from the roof of his house. There are both V-band and I-band images, in the native Mark IV camera format.

Note that these images were taken before Tom fixed a problem with the RA drive. There is quite a noticeable elongation of the PSF East-West in the images described below. More recent data has a nicer PSF.

This Note contains a brief analysis of one particular pair of images, a field in Virgo.

Table of Contents

Gross characteristics of the image

The particular pair of images on which I concentrated are called

Each was a 100-second exposure with the Mark IV camera sometime during the night of March 17, 1999. Tom says that his cameras were pointed roughly 10 degrees to the west of the meridian. The pictures are centered roughly on RA = 12h 36m, which passed the meridian at about 45 minutes past midnight, Chicago time. Adding another 40 minutes to move 10 degrees west, I estimate the time of the exposures to have been 01:25 AM CST = 07:25 UT on March 18, 1999.

I performed no image processing: I did not subtract a dark image, nor did I use a flatfield frame. The raw images are shown below: the V-band image has a strong gradient in one corner, while the I-band image is relatively flat. I have rotated both images by 90 degrees clockwise from their natural orientation, so that North is up and East to the left. Each image is roughly 4.2 degrees wide by 4.2 degrees high.

Identifying the field is tough: the V-band image, which would match most sky charts more closely, is hard to use due to the gradient in one corner. The I-band image is much cleaner, but it turns out that the brightest star of in it (near the southern edge) is actually a pretty faint in V -- and therefore inconspicuous on charts. Tom Droege very kindly provided the field ID: it runs 

                RA  =  12h 26m   to   12h 46m
                Dec = +03d 40'   to  +07d 40'

Below is a copy of the I-band image with a number of stars marked with letters -- and several galaxies marked with circles. 

Star      Name        RA (2000)     Dec         V mag
----------------------------------------------------------
  A    31 Vir        12:41:57.1   +06:48:24      5.59
  B    HD 109914     12:38:29.95   06:59:19      7.70
  F    HD 108677     12:29:05.16   06:51:54.8    8.00
  G    HD 108561     12:28:18.45   04:23:47.8    6.70
  H    HD 109287     12:33:37.03   04:13:13.9    7.70
  I    HD 109840     12:37:51.84   04:17:19.3    6.90
  J    HD 110261     12:40:49.96   03:56:11.2    7.90
  K    Tycho ???     12:35:58.14   05:32:55.7   10.4

There are many galaxies in this area: it's the southern outskirts of the Virgo cluster. If one looks closely at the image, one can detect quite a few faint, fuzzy patches that are clearly non-stellar. Three of the most prominent are marked in the image above:

We just missed catching the very large, bright galaxies M49 and NGC 4535, which are off the field to the north. Rats.

Tom provided a large set of pictures from this night. Images N40 - N60 run in a long strip, from west to east, with quite a bit of overlap between images. From frame N40 to N41, the field moved eastwards by about 494 pixels, which is almost exactly one degree.

Note that asteroid 8 Flora should be near the southern edge of the next few images in this sequence (N41, N42, N43, N44). The asteroid moves westwards over the next few weeks -- it would be near the middle of the field of view of frame N40 around April 3. If Tom has additional images with the same camera placement in Dec, one might be able to find the asteroid and watch it move from night to night.

Defects of different sorts

There are several kinds of problems with the images. Some of them might be removed or alleviated with processing, but others won't. I provide a VERY brief list here -- these all deserve more attention.

  1. Distortion: The images in both V and I clearly show radial distortion. Stars near the edges and, especially, in the corners of the image, have a core surrounded by oval "wings". The "wings" point back towards the center of the frame. On brief inspection, the images look similar to those taken a few months ago (see Technical Note 47).

  2. Ice crystals(?) The I-band chip appears to have some damage near its center: eight of them are visible near the top of the closeup of the central section below. The V-band chip seems clear of these defects. (Note that this closeup has not been rotated, so that north is to left, east is down.)

  3. Horizontal pattern noise: The I-band image has a subtle pattern of horizontal lines with slightly less than the average number of counts. You can see it in the closeup above. The lines are spaced by roughly sixteen rows. The V-band image is free of these stripes, at least to my eyes. Look at the V-band closeup below and check for yourself. (Note that this closeup has not been rotated, so that north is to left, east is down.)

Photometry

I'll address only two short questions here: what is the stellar brightness at which the images saturate, and what is the limiting magnitude of the images? More detailed comparison of photometry vs. catalog values will have to wait for later (and for some image processing).

I looked at stars near the centers of both V-band and I-band images, where there was little distortion of the PSF. The results are

The V-band image yielded PSF with FWHM around 3.2 pixels at best. The I-band image had FWHM around 3.4 pixels.

When I compared raw aperture magnitudes (using an aperture of radius 5 pixels) of several bright stars in the image to their catalog V-band values, I found good agreement: 

  star         catalog V          raw aper mag         diff 
---------------------------------------------------------------
  HD 109129       8.9                11.13              2.23
  HD 109248       9.3                11.49              2.19
  PPM 158953     10.0                12.19              2.19

What's the limiting magnitude of each image? I printed a hardcopy of the central section of each image (same as the closeup pictures above), and compared the hardcopy with charts based on the HST Guide Star Catalog. Just a quick and dirty check. The results:

It is impossible to do any accurate photometry of these very faint smudges, of course. But one can probably measure positions with some reasonable accuracy.

A very red star

The brightest star visible in the I-band image leaves a bleed trail that runs for several hundred pixels. It is marked as star "P" in the finding chart near the beginning of this Technical Note. The star is much fainter in the V-band image. I was intrigued by this difference, so I did a little digging.

The star is known by many names; a few are

Using the SIMBAD meta-catalog, I was able to learn the following about it:

Some closeups of galaxies

As mentioned earlier, this field is near the southern outskirts of the Virgo cluster of galaxies. One can find many faint fuzzies in the images. Let me show closeups of just a couple, for fun. The closeups are taken from the I-band picture.

Above is a closeup of NGC 4570 (North is up, East to the left). For comparison, here's a closeup of the galaxy from the Digitized Sky Survey: this picture covers only the central portion of the Mark IV closeup above.

Now a picture of the galaxy NGC 4580 (a more face-on spiral), from the I-band Mark IV image ....

... and from the Digitized Sky Survey: