Based with some useful tips received elsewhere, I've decided I need to withdraw this and start from the beginning.  Perhaps the best that can extracted from the images is differential photometry relative to the mean of a large area of the bright lunar disk shown in the image.  But it is a start. - Kurt ----- Original Message ---- From: Canopus56 <canopus56@yahoo.com> To: Utah Astronomy List Serv <utah-astronomy@mailman.xmission.com> Sent: Saturday, September 12, 2009 6:51:59 PM Subject: Amateur lunar photometry - first test using a Meade DSI This note reports my test to use a Meade DSI to derive the rough magnitudes per square arcsec  (mpsas) photometry for the night sky above and dark carter floors on a 71% illuminated Moon.  In general the idea crashed and burned, but here is what I was able to salvage. In addition to other technical problems that I need to work through, I used the wrong procedure and the signal ratio was too low to be able to derive confirmable good photometric magnitudes from images of the south pole and Alcyone taken on the morning of Sept. 9.  This note is to extract what data and conclusions that can be recovered from the effort. While I am not going to vouch for the scientific accuracy of the magnitudes in the above presentation, it at least provides some differential magnitude information for the lunar south pole at 77% illuminated fraction.  It also illustrates how other amateurs, more experienced in photometry than I, might develop techniques for amateur absolute photometry of the Moon. This note discusses a photometric colorgram of the south lunar pole contained in a summary info panel of five slides at: http://members.csolutions.net/fisherka/astronote/observed/LCROSS/20090909_11... http://tinyurl.com/r4x9fx On Sept. 9, I took normal tone images of the south lunar pole through a Johnson V filter at a high f-ratio more suitable for topographic resolution, see: Sept. 9 image in info panel format http://members.csolutions.net/fisherka/astronote/observed/LCROSS/20090909_11... http://tinyurl.com/nlf58f Imaging particulars are: ETX 125mm, MeadeDSI Pro, Johnson V filter, 0.0625 secs, approx. 4500mm efl. by negative projection; dark frame substraction only.  Then I swung over, without changing exposure settings and imaged part of the Alcyone multiple.  Alcyone was only a few degrees lower in altitude than the Moon.  The three applicable stars imaged were: Star Johnson V*  Pixel_value mpsas Alcyone 2.9         12223       16.5  HD23629 6.3         596         21.3 HD23607 8.7         81          n/a * Per Hauck (1997) Analysis done in AIP4WIN indicated good sigmas for the three stars based on 70 individual images and the FWHM pixel value that corresponds to each reference object was obtained.  Sigma values looked good: 0.019, 0.0021, 0.09, respectively. The S/N ratio was not.  Mag. 8.7 HD23607 was just barely visible in the raw images - having only a few pixel values above background. I was unable to build a consistent single coefficient transform to relate the three reference magnitudes (due to poor signal to noise ratios). I was unable to relate the reference pixel value for Alcyone and the pixel values in a raw south pole image to mpsas within a reasonable range of normal values for the lunar surface by converting them to mpsas. See Cavadore for an algorithm to convert a reference stellar magnitude and pixel values to magnitudes per square arcsec.  http://www.astrosurf.com/cavadore/articles/pollution/index.html Nonetheless the raw pixel values for Alcyone and HD23629 can be used to extract some provisional differential photometry information from the corresponding raw image of the south lunar pole.  It is a reasonable assumption that pixels in the south lunar polar image that have the same pixel values as Alcyone and HD23629 have the same V magnitudes. A raw south polar image (file: author's file Moon50001-0008.fts) - a crisp lunar image - was selected and image was dark flat subtracted.   First, with respect to 6.3 V magnitude HD23629, there are no pixels in raw image below a pixel value of 596.  The minimum pixel value for the image is 611. The implication of this is that single point stars of less than 5.8 magnitudes cannot be seen in the sky immediately above the dark limb due to insufficient contrast.  This is consistent with Schaefer's articles and Qbasic program for the limiting magnitude of a grazing or occulting star.  Recently, we discussed here the September 9 graze of ZC387, a V magnitude 6.9 star, whose ground path ran through Utah to California. The IOTA's Occult 4.0 program returned the following limiting magnitudes by aperature and distance from the terminator's intersection with the lunar (negative cusp angles are on the sunlit side): Occult 4.0 limiting magnitude report excerpt for ZC387 graze on 9-9-2009 --------------------- Librations  Long -5.58   Lat -6.16 Illumination of moon  77%- At longitude -112.50: Limiting Magnitudes for various telescope apertures (in mm) CA\Tdia 250   300   350 -1.1    5.8   6.1   6.3 0.9     5.8   6.1   6.3 2.9     7.6   7.9   8.1 4.9     7.9   8.2   8.4 6.9     8.2   8.5   8.7 --------------------- Even having 14 inches of aperature, as compared to 10 inches, gets you only a small additional limiting magnitude reach.  The lack of contrast between the background sky and the object controls. This 1000 to 2000 pixel value clip (5.75 to 5.0 V mags) picks up the dark sky above the lunar limb at Faustini. Using some basic differential photometry, the following range of pixel values and magnitudes were defined: Pixel Value  V mag  596          6.3 1000         5.75 2000         5.0 5000         4.0 7000         3.5 9000         3.23 ~3.25 11000        3.0 12223        2.9 14000        2.75 For pixel value ranges less than 2000, 6.3 V mag HD23629 was used as the reference star. For pixel value ranges above 5000, 2.9 V mag Alcyone was used as the reference star. For discussion purposes, these magnitudes are divided into the following ranges that are color coded into the summary information panel in an image panel at (url: http://tinyurl.com/r4x9fx ): V mag rng  mpsas rng Color    Pixel value 5.75-5.0   20.4-19.3 Red-brwn 1000-2000 5.0-4.0    19.3-18.0 Magenta  2000-5000 3.5-3.25   17.5-17.2 Aqua     7000-9000 3.25-3.0   17.2-16.9 n/a      9000-11000 3.0-2.75   16.9-16.6 Yellow   11000-14000 Although I was unable to use direct computation using pixel values and the pixel size in arcsecs to reasonable values of apparent brightness in mpsas, Schaefer provides an algorithm to prepare a "cheat" indirect computation.  Schaefer's algorithm provides an method to convert V steller magnitudes to mpsas for stars above magnitude 6.6. The results of that conversion are included in the table above. See discussion url:  http://members.csolutions.net/fisherka/astronote/plan/tlmnelm/html/NELM2BCal... Using AIP4WIN's MinMix math function, a black/white fits image was created by replacing all pixels below 1000 with 0 and all pixels above 2000 (5.75 to 5.0 V mags) with 0.  The next MinMax threashold clipped image is for pixel values bewteen 2000 and 5000 - corresponding (I hope) to predicted V magnitudes 5.0 to 4.0. This pixel value clip picks up the shadowed area in the Malapert dark hole, the interior shadowed area of craters Caebus, part of the shadowed area in Haworth, but not any shadowed portion of Caebus A and B. The next pixel value clip is for 7,000 to 9,000 pixel values ( 3.5 to 3.23 V mags).  The clip picks up the shadowed floors in Caebus A and B.  The final pixel value clip is for 11,000 to 14,000 ( 3.0 Vmags to 2.75 V mags) which encompasses the 12,223 pixel value for Alcyone (2.9 V mags). Next, the RGB channel combine function of the image processing software AIP4WIN was used to create a three colorgram of these limited magnitude ranges.  A three-color colorgram is the maximum that you can create using AIP4WIN.  A second piece of freeware available on the internet - iMerge (url: http://www.geocities.com/jgroveuk/iMerge.html ) - was used to add a fourth color layer using two bmp images.  The assembly of multiple color layers is, of course, trivial for owners of the expensive Photoshop application. Finally, a second more detailed color gram was prepared of the Cabeus region. See the info panel (url: http://tinyurl.com/r4x9fx ).  An additional range clip of pixels was added into to cover the ADU pixel value ranges found in Cabeus A.  The detailed colorgram suggests that the shadowed portion of Cabeus A at 77% illuminated fraction was an apparent brightness equal to between 3.5 to 3.25 magnitudes. But again, this is one of my first photometry tests, so the result may be way off.  The next opportunity to experiment with this technique at illuminated fractions similar to a 71% illuminated fraction is will be 9-28-2009 3:00UT (68% illumination).  This will be an east illumination at very low horizon altitude of about 30 degrees, considering the Moon's position in the ecliptic. This amateur work product.  Comments and criticisms of the above are welcomed. Clear Skies - Kurt P.S. A zipped file of the dark subtracted raw image of the Moon and Alycone and the individual clipped magnitude channels in fits format is available online at url: http://members.csolutions.net/fisherka/astronote/observed/LCROSS/20090909Pho...