Plume visibility analysis and imaging exposure recommendations canopus56@yahoo.com 9-15-2009 (Rough draft v1) The following is an amateur note. Comments and criticisms on the same are welcomed. Jim on 9-12-2009 said to Rick Baldridge: > The relationship is the same whether we are talking about the Sun's > total brightness (visual magnitude) and irradiance > (the numbers just mentioned) or surface brightness (mpsas) > and radiance, since the latter numbers are just the former ones > divided by the same number of square arc-sec. The LCROSS Team webpage titled "Amateur Observations of LCROSS Impact: October 9, 2009" (Last updated: 9/06/09) states as of Sept 6, 2009: > "The brightness or magnitude of the event is of course dependant of what > part you are talking about, but we estimate magnitude 6-9 for the > best visible part and time." "http://lcross.arc.nasa.gov/observation/amateur.htm ---------------------- At the LCROSS Team press conference of 9-11-2009 one of the investigators stated: > "The impact will be between 4 and 5 magnitudes." - apparently refering to stellar magnitudes and not mpsas. http://www.youtube.com/watch?v=Y9RAWPBoi3I The above conclusions results from a misinterpretation of controlling photometry equations. A one-square-arcsec mpsas extended object is _not_ directly equivalent to the stellar magnitude of that object, except in very limited theoretical circumstance that almost never occurs for amateur observing. A 4.0 mpsas object is not always equal to a stellar magnitude 4.0 star. The LCROSS ejecta plume is equivalent in apparent brightness to 1.5 stellar magnitude - not 4 or 5 stellar magnitudes. The proposition that a 4 or 5 mpsas object is equivalent to a stellar magintude 5 object directly contradicts well-known physical observations - many years of lunar occultation data, by well-documented modeling of the telescopic limiting magnitude for lunar occultations and by simple naked-eye and telescopic observation of the Moon. Again, the equivocation of a 4 or 5 mpsas object to a 5 or 5 stellar magnitude star is a false logical equivocation and is inconsistent with basic physical observation of the Moon. With respect to the LCROSS impact plume, and as consquence of this misinterpertation, the suggested times on the LCROSS Citizen Science "About" page (calibration of exposure to 6 to 9 magnitude star), results in unusable exposures that will not capture the impact plume. If exposure calibrations are set to the correct equivalent stellar magnitude - which in this case is a small extended linear object 1.7 arcsec by 5.6 arcsec - an image of the ejecta cloud easily can be made. Photometric analysis of the changes in the plume over the first 30 seconds can be undertaken as a hobby science experiment. With respect to the LCROSS impact, I am recommending amateur calibration of their first raw images to a 2.5 stellar magnitude star. This reasons for this recommendation are discussed in more depth below. The foregoing conclusion is subject to further analysis of two items: (1) the target-observer angle of the Cabeus A1 target crater at the libration at the time of impact and (2) better detailed V-band photometry of shadowed area within Cabeus A1 and the surrounding surface terrain under libration and illumination conditions analogous to the impact. Amateurs can informally gather since photometry data for themselves in imaging dry runs on September 27 when the Moon is next at 71% (east) illuminated fraction. I have identified a USGS archive of some 85,000 lunar photographs taken from Flagstaff (the ROLO project) that have pre-reduced high quality V-band calibrations. Presumably, some images exist of the region around Caebus A1 at the correct illumination and libration. If the appropriate images can be retrieved and the photometry extracted from them, then amateurs will be will be better prepared to image the impact on Oct. 9, 2009. However, image scale is reported as 576 x 576 for the full disk. The resolution may be sufficient to extract regional surface photometry around Caebus A, but probably is not sufficient to determine the apparent brightness of the shadowed portion of the Caebus A1 crater relative to surrounding surface. The following discussion relates to Dr. Wooden's posting of a spectral irradiance chart - adjusted to unit arcsec size for three different sized of the same ejecta expanding ejecta plume - posted September 11, 2009 at url: Wooden file "ejecta_flux_predict_09sep02.pdf" posted to LCROSS Observation Google Group http://tinyurl.com/m6uqgj Dr. Wooden's chart is not a spectral radiance chart; it is a spectral irradiance chart showing the same object at three different points in time. Spectral radiance is expressed in units of W m-2 str-1 micron-1; spectral irradiance is expressed in units of W m-2 micron-1. I acknowledge that it also is possible to express spectral radiance in equivant arcsec units instead of steradians, but that is not the purpose that Dr. Wooden appears to have intended use the chart for. Rather, she is unitizing the expanding size of an ejecta curtain: "I am posting 3 images and an ejecta flux prediction (W/(m^2 um arcsec^2)) versus Wavelength (um) for different values of the grain column density (1E7 m^-2 represents 4-20 sec; 2E5 m^-2 represents 30-90 sec). The brighter part of the ejecta plume reaches about 1.5 to 3 arc sec in height and spreads to 15"-30" wide in 120-240 sec." Wooden 9-11-2009 post. http://groups.google.com/group/lcross_observation/msg/96ef20c316a23d27 http://tinyurl.com/ndfasu Note that she only provides one physical dimension for the size of the ejecta curtain during the 4-20 sec timeframe - the red line on the chart. Based on a grad student's post from an LCROSS [professional] Observation Campaign university, we can fill-in the dimensions for 4-20 sec cloud size: 10km wide by 5 km high or 5.6 arcsecs by 3 arcsec. See - Post by Chase Miller 9-4-2009 http://groups.google.com/group/lcross_observation/msg/7c14e92f172fe4e8 http://tinyurl.com/ndfasu With the obscuring Cabeus A1 crater rim, only 3km of the curtain sticks up into Earth view and has an equivalent arcsec size of about 1.7 arcsecs. Thus, the 4-20 sec cloud has an area of 1.7 x 5.6 arcsecs = 9.52 arcsecs. For the second green line - the 30-90 second curtain - she reports no dimensions. For the third blue line, it is unclear whether this is the 120-240 second curtain. If it is, Dr. Wooden provides us only with the width dimension but not the height dimension. It does not appear to be possible to infer the cloud dimensions in kilometers and corresponding arcsec sizes for the 30-90 second green line curtainn or the 120-240 second red line curtain from: Summy, D.; Goldstein, D. B.; Colaprete, A.; Varghese, P. L.; Trafton, L. M. 2009. LCROSS Impact: Dust and Gas Dynamics. 40th Lunar and Planetary Science Conference, (Lunar and Planetary Science XL), held March 23-27, 2009 in The Woodlands, Texas, id.2267. http://adsabs.harvard.edu/abs/2009LPI....40.2267S http://www.lpi.usra.edu/meetings/lpsc2009/pdf/2267.pdf Most amateurs are familar with the fundamental stellar magnitude equation: V_obj = 2.5 * log10( B_ref / B_obj) + V_ref [Eq. 1] - where B is the luminance of the object in any consistent system of units. The formula to determine the magnitude per square arcsecs of an extended object is a variant of the stellar magnitude equation: MPSAS_obj = 2.5 * log10( B_ref / ( B_obj / Object_Area arcsec^2 ) + V_ref [Eq. 2] - again where B is the luminance of the object in the same units for the reference and measured object. Here the measured brightness of the object is divided by its area expressed in square arcsecs. The stellar V magnitude systems currently in use are based the stellar magnitude of Vega (and can also use a zero-point based on solar brightness). The reference magnitude and brightness of Vega are known to a high degree of precsion: Vega V_ref = 0.0026 magnitudes Vega B_ref = 3.464E-8 W m-2 micron-1 at 555nm Allen's Astrophysical Quantities (4th 2000) at Table 12.7; Bohlin and Gilliland (2007). Doing a chart take-off from Dr. Wooden's chart for the red 4-20 sec cloud line, she provides unitized value (that is B_obj / Object_Area arcsec-2 ) for the spectral irradiance at 550nm of about 9E-10 W m-2 micron-1 arcsec-2. Thus we have all the values to find mpsas by applying Eq. 2, which works out to: 4.0 mpsas = 2.5 * log10( 3.464E-8 W m-2 micron-1 / 9E-10 W m-2 micron-1 arcsec-2 ) + 0.0026 However, 4.0 mpsas is _not equal_ to 4.0 V stellar magnitudes in the instant case. Mpsas is _only equal_ to V mags in the special case where the area of the extended object is 1.0 arcsec-2. In those cases: Mpsas_obj mpsas = 2.5 * log10( B_ref / ( B_obj / 1.0 arcsec^2 ) + V_ref [Eq. 3] which yields the same coefficient as: V_obj = 2.5 * log10( B_ref / B_obj) + V_ref [Eq. 1, repeated). Only in this special case, do stellar magnitudes equal mpsas. In the instant case, the extended area of the object for the red 4-20 second cloud is 9.52 arcsec+2; it is not 1.0 arcsec+2. The special case does not apply. The chart also plots a spectral irradiance curve for Vega. Vega is an A0V spectral class star. While this information is helpful (as discussed below), it unfortunately, reenforces the readers chances of misinterpreting the chart and falsely equivocating the apparent brightness of a 5.0 stellar magnitude star with a 4 or 5 mpsas extended object that is larger in area than 1.0 arcsec^2. The correct way to apply Eq. 1 to find the stellar magnitude of the cloud is find its spectral irradiance, or the product of its unitized spectral irradiance in the chart times its area in arcsecs^2, e.g. 8.59 W m-2 micron-1 = 9E-10 W m-2 micron-1 arcsec-2 * 9.52 arcsec^2 where 9E-10 W m-2 micron-1 arcsec^2 is the unitized spectral irradiance of the extended object - the 4-20 second red line cloud. This spectral irradiance (8.59 W m-2 micron-1) is then applied to Eq. 1 to find the equivalent stellar magnitude of the cloud at 4-20 secs: 1.5 V_mags = 2.5 * log10( 3.464E-8 W m-2 micron-1 / 8.59 W m-2 micron-1 ) + 0.0026 Because Dr. Wooden has provided the unitized spectral irradiance values for the red 4-20 sec, green 30-90 sec and blue 120-240 sec, we can determine apparent brigthness expressed in mpsas for the predicted LCROSS cloud at each of those time frames. The results are: Cloud at mpsas red 4-20 sec 4.0 green 30-90 sec 6.5 blue 120-240 sec 8.9 This is consistent with the LCROSS Citizen Science "About" page, that states: LCROSS - Citizen Science Portal About Page (Last updated: no update tag, downloaded 9-11-2009) "The expected plume average brightness for the Centaur (first) impact event is 4-6 visual magnitudes per arscec-squared (it's an extended object) in the first minute." http://apps.nasa.gov/lcross/about/ However, _nothing can be said_ about the equivalent stellar magnitude of the green 30-90 sec and blue 120-240 sec clouds _because we do not know the dimensions of the cloud_ at those times. Nor do we really need to know those values for amateur imaging, because for Earth-based amateurs, as the cloud expands and dims to below 6.0 mpsas - it will not be visible against the shadowed portion of Cabeus A1 or against the bright sunlit surface of the Moon. The LCROSS Citizen Science "About" page also suggests that exposure calibration can be done with respect to and 8 or 9 stellar magnitude star: LCROSS - Citizen Science Portal About Page (Last updated: no update tag, downloaded 9-11-2009) "If 8th to 9th magnitude stars can just begin to be imaged with a particular telescope/camera and exposure combination, that exposure should be able to image the plume." http://apps.nasa.gov/lcross/about/ It is unclear whether these values simply equivocate the green and blue mpsas values above with stellar magnitudes or whether there is some other computation that converts mpsas to stellar magnitude. I suspect the former is the case. This 8 or 9 setllar magnitude exposure calibration strategy (calibrate to a 9 stellar magnitude star) makes sense under a now deprecated impact scenario. Previously, the ejecta cloud was thought to rise against the background of the night sky - which was about a 15 mpsas apparent brightness. Under the new Caebus A1 impact scenario, the observable background is the shadowed portion of Caebus A1 crater, which is of unknown apparent brightness. For reasons discussed below, it is probably about 5 mpsas. This is far brighter than 15 mpsas, so a different exposure calibration value is warranted. The average apparent brightness for large areas of the lunar surface is well-known: 4 to 6 mpsas or an equivalent stellar magnitude value of about -2.9 to -12 with a maximum at the full Moon of 12.7 stellar magnitudes. The apparent brightness of the shadowed interior of craters is not well-reported in literature. The site specific apparent brightness in mpsas for the lunar surface around Cabeus A1 is not known. Assume that the surface around crater Cabeus A1 is about 4 mpsas. The shadowed crater is probably somewhere around 5 mpsas. This may explain the spectral irradiance curve for a "5 mag AOV" star included in Dr. Wooden's chart. As you can see, at 555nm, the take-off brightness value is spectral irradiance of 3.75E-10 W m-2 micron-1 arcsec-2. So let's model the ejecta cloud as it dims from a peak of 4.0 mpsas down to 5.0 mpsas. Plugging that value into Eq 1 and Eq 2 and using a 9.52 arcsec^2 sized cloud, we get: "Mag 5" Vega AOV class star mpsas: 5.0 V-mag: 2.5 Again, the 5.0 mpsas Vega type calibration simulated "ejecta cloud" _is not_ equivalent in apparent brightness to a 5.0 stellar magnitude star. If you want to calibrate the exposure of an image a 5.0 mpsas cloud with an area of 9.52 arcsec^2, then you use a 2.5 stellar magnitude calibration star, not a 5.0 V stellar magnitude star. By shear chance, I happened to take an image of the lunar south pole on Sept. 9, calibrating the exposure time to 2.8 stellar magnitude Alcyone. Some rough, and admittedly low-precision and probably inaccurate, diff erential photometry indicates that the median pixel value of the image was 2.5 stellar magnitudes. See frame-slide 5 at the bottom of: http://members.csolutions.net/fisherka/astronote/observed/LCROSS/20090909_1154UT_111W_41N_SP_photometry_panel_kaf.jpg http://tinyurl.com/r4x9fx By calibrating to a 2.8 stellar magnitude star, I ended up with a normal tone image that appears to capture the 4 to 6 mpsas tonal range of larger regions of the lunar surface. This image is observational evidence that the apparent brightness of a 5 stellar magnitude star _is not_ directly equivalent in apparent brightness to a 5.0 mpsas extended object in the instant case. Had I calibrated to a 5 mag star, the image would have been overexposed; it would have exposed as a flat white frame. Two other physical observations, done by many amateurs, also contradict the equivocation of a mpsas to stellar V magnitude when applying the data in Dr. Wooden's chart. First, is the experience of and well-documented algorithms to compute the telescopic limiting magnitude of occulting stars used by amateur lunar occultation observations. That was the topic of some recent messages in the LCROSS Observation Group K. Fisher post to LCROSS Observation Group 8-28-2009 "Using Photometry to Find Sky Brightness" http://groups.google.com/group/lcross_observation/msg/dd78a576fd7a6598 K. Fisher post to LCROSS Observation Group 8-28-2009 "Using Photometry to Find Sky Brightness" http://tinyurl.com/mr2j2v Derek Breit Post to LCROSS Observation Group 8-30-2009 "Using Photometry to Find Sky Brightness" http://groups.google.com/group/lcross_observation/msg/f1f5a57bdec3568a http://tinyurl.com/puyru6 Derek Breit Post to LCROSS Observation Group 9-1-2009 "Using Photometry to Find Sky Brightness" http://groups.google.com/group/lcross_observation/msg/7ac39d8eb15bc442 http://tinyurl.com/q5jc6j When told of the 9/11 LCROSS Team press conference statement that cloud brightness would be equivalent to 4 or 5 stellar magnitudes, one long-time member of the IOTA quickly commented in the LCROSS Observation group that: "There is zero chance I can image this with this being the target. . . . . All I would get would be a white screen. If I shortened the exposures to where I just began to see albedo features on the bright limb, then I couldn't see even a mag 2 point source." As experienced lunar occultation observers know, the telescopic limiting magnitude for stars about 2 cusp degrees on sunlit side of the terminator even at a reduced 71% lunar illumination, is about 4.0 magnitudes for a 10 inch scope and 5 stellar magnitudes for a 15 inch telescope. This is where the star has a background of the night sky. The notion of imaging a 5.0 stellar magnitude object against the bright limb correctly strikes even causal lunar observers as nonsensical. Anyone who has causally watched 2.0 magnitude stars disappear as they drift within a few arcminutes of - but never touch - the bright limb of a partially illluminated Moon know a 5.0 stellar magnitude star would never be seen on or tangent to the lunar bright limb. The physical evidence that the brightest magnitude stars (0.0 or 1.0 magnitude) stars can be imaged as they occult the limb, contradicts the notion that a 4.0-5.0 stellar magnitude star is equivalent in apparent brightness to a 4.0-5.0 mpsas extended object in the instant case. Second, is the simple naked-eye observation of the Moon itself. The Moon has a well-known range of apparent brightness: 4 to 6 mpsas or an equivalent stellar magnitude value of about -2.9 to -12 with a maximum at the full Moon of -12.7 stellar magnitudes. The Moon as a diameter of about 1/2 degree or 1800 arcsecs. Eq. 2 is simplifed by Covington and also Clark where only the V_mag and area of an extended object are known to: MPSAS_obj = 2.5 * log10( Object_Area arcsec^2 ) + V_obj [Eq. 4] See - Clark, R.N., 1990. Appendix E in Visual Astronomy of the Deep Sky, Cambridge University Press and Sky Publishing.http://www.clarkvision.com/visastro/appendix-e.html Covington, M. (1998, 2d). Appendix A, Astrophotography for Amateurs. pp. 259-261 http://www.covingtoninnovations.com/ This makes sense. You are comparing the reference object's apparent brightness to itself. Log10( B_ref / ( B_obj / ObjArea_arcsec^2 ) equals Log10(1 * ObjArea_arcsec^2 ). Thus - V_obj = MPSAS_obj - 2.5 * log10( Object_Area arcsec^2 ) [Eq. 5] For the full Moon - -12.7 = 4.0 - 2.5 * log10( pi/4 * 1800^2 arcsecs^2) - which is about the right published value for the integrated stellar magnitude of the full Moon's disk. Let's apply this pattern to TFOVs of decreasing size while holding 4 mpsas steady: Type| Area_arcsec_s^2 | mpsas | Vmag in image | Mpsas | V_mag Full Moon 1800 arcsec dia. 2544690 4 -12.01 CCD rectangle 4x5 arcmin 72000 4 -8.1 Gem-N image ~1x1 arcminute 3600 4 -4.9 Hypothetical 30 arcsec sq 900 4 -3.4 Hypothetical 10 arcsec sq 100 4 -1 Hypothetical 3 arcsec sq 10 4 1.5 One pixel covering 1 arcseq sq 1 4 4 As the area of the Moon seen in a TFOV is reduced, the corresponding apparent brightness of the Moon decreases. In otherwords, mpsas and the equivalent stellar magnitude are dependent on the area of the object, except for the special case of an "extended object" that is the same size as a stellar point. The claim that mpsas can be equivocated with stellar V magnitude of equal value in the instant case is inconsistent with modeling of the apparent brightness of objects of different sizes but the same mpsas. This effect can be seen in an office setting by cuting a hole in the bottom of a paper clip - much larger than your eye. Look at a flat ceiling panel above three meters away. Then look at it through the cup hole. The apparent brightness of the flat ceiling panel seen through the cup will be less than without viewing through the cup. Based on available information and the above discussion, the simplest statement that can be made is that the ejecta plume will be similar in brightness to the surrounding lunar surface. The shadowed portion of the crater from which the ejecta curtain will arise will be darker than the average surrounding surface brightness. The ejecta plume may or may not have any contrast against the dark shadowed portion of the crater from which it will rise. The more likely scenario is that as the curtain rises the curtain will increase in brightness to a level equal to the flat terrain that surrounds Cabeus A1. The ejecta plume is modeled to be between 4 and 6 mpsas within the V-band. This is equal in brightness to the average range of whole Moon's mpsas across a lunar cycle. As seen from Earth, the brightest portion of the ejecta curtain will be a tiny extended object framed against the dark shadow portion of Caebus A1. As the curtain is framed by any background that is illuminated by sunlight, it will be invisible. The LCROSS ejecta curtain may have the appearance of a rhomboid - an "inverted lampshade" - with parts of it sliced off. As previously noted, on Sept. 9 I attempted a rough low accuracy photometry test using a Meade DSI camera. The purpose of the test was to try to map surface V magnitudes in and around Caebus A1. See frame-slides 1 and 2 at the bottom of: http://members.csolutions.net/fisherka/astronote/observed/LCROSS/20090909_1154UT_111W_41N_SP_photometry_panel_kaf.jpg http://tinyurl.com/r4x9fx The test, although not scientifically accurate, indicated that the sunlit surface around Caebus A and A1 had a V stellar mag of 2.75-3.0, the light shadowed portion of Caebus A1 was 3.0-3.25 stellar mags and a small dark shadowed hole at the bottom Caebus A1 read as 3.25-3.50 V magnitudes. I was not able to convert these figures into mpsas. I have found no published data on the site specific brightness surrounding the LCROSS crater or of the shadowed portion of the crater Cabeus A1 or on the V magnitudes or mpsas of the shadowed portion of craters generally. There may be a USGS data image archive that contains images of lunar south pole in the same libration and with the same illumination as the impact. There is high precision calibration data to Vega's magnitude for each image in the archive but the image scale is only 576 x 576 pixels for the full disk. Keiffer, Hugh, H. and Stone, Thomas, C. (U.S. Geo. Survey, Flagstaff, AZ). 2005. The Spectral Irradiance of the Moon. AJ 129:2887-2901. http://adsabs.harvard.edu/abs/2005AJ....129.2887K http://www.iop.org/EJ/article/1538-3881/129/6/2887/204408.text.html Keiffer and Thomas (2005) describe how between 1996 and 2003, they operate a robotic lunar telescope that automatically imaged the Moon on any clear sky night. Photometry of the lunar disk was obtained in 23 wavelengths. Over seven years, they collected 67,505 images at almost the entire libration ranges. Id. Figure 2. Contact information for the ROLO archive is available online. The archive's website currently maintains 85,000 lunar images and is managed by Tom Stone. See url - ROLO Main Page http://www.moon-cal.org/database/image_archive.php Procedure to request images from the ROLO database http://www.moon-cal.org/database/image_archive.php Images in the database are 576 x 576 pixels for the lunar disk, which is too large for determining the mpsas or V band stellar magnitude of the shadowed portion of Cabeus A1. The image scale appears sufficient to obtain good whole disk images of the Moon at illumination and libration nearly identical to the LCROSS Oct. 9 impact and to determine regional mpsas and V-band magnitudes for the surface area that surrounds Cabeus A. I have replicated (Jim's) list of dates that match LCROSS illumination and libration for the years of operation covered by the ROLO image archive. This list can used as a search list for the ROLO image archive. I will not be directly pursuing a request for images from Dr. Stone. ----------------- Table - Cabeus A on Oct. 9, 2009 11:30 UT illumination characteristics from LTVT Cabeus A 34.15W 81.04S Sun angle 8.27 alt 331.33 az Illuminated fraction: 70.896 Colong: 158.158 Error: 0.5 degs Steps: 60 minutes Obs Long: -111.8 Obs Lat: 41.8 Obs Elv: 1300m Moon greater than 30 degs alt Matching illumination and libration dates 5/27/1997 11:09:18 (32 degs alt) 2/13/2001 07:28:41 12/23/2002 09:29:45 1/19/2003 05:50:49 2/17/2003 10:22:15 12/13/2003 07:36:14 ----------------- On Sept. 11, Rick Baldridge noted that: "NASA will provide professional and amateur observing groups more detail regarding plume size and visibility in the coming weeks. The plume will not extend above the lunar limb, and will not be situated against a dark background such as a shadowed region between craters. However, that does not mean the plume will not be visible. Video and photographic observations must now focus on bringing out the brightening caused by the eject plume in front of a lit lunar surface." September 11, 2009 LCROSS Science Team Announced Target Crater Posted by rickbaldridge on September 11th, 2009. NASA LCROSS Citizen Science Page Blog http://apps.nasa.gov/lcross/ Implications from the foregoing discussion for imaging the LCROSS impact are as follows. If the ejecta plume follows its predicted apparent brightness _and_ if the mpsas of the shadowed portion of crater Cabeus A1 is at least one magnitude lower than 4 mpsas, the impact can be easily imaged and observed using 5 inches of aperture or more. Earth based amateur observers will see essentially a surface contrast effect. As the brightest part of the 10km plume expands within and above the bowl of 17km Cabeus A1 and reaches a brightness of about 4.0 mpsas, the plume will obscure the shadowed portion of Cabeus A1 from Earth observers. The plane of the rim and/or just above the rim will take on a surface brightness equal the surface brightness of the surrounding surface terrain. That surface typically will have an mpsas between 4.0 and 6.0. Normally, Caebus A1 has a bowl shaped appearance caused by crater shadowing when viewed from Earth visually at 300x and/or when imaged, as illustrated in this high resolution amateur image taken in September by Stefan Lammel: http://tinyurl.com/qjex6e (I visually observed the same view a few hours after Stefan took his image using a Meade ETX 125 with a TMB 4mm planetary eyepiece at 300x. The level of detail that I was to observe visually easily exceed what Stefan was able to capture photographically. Stefan when posting his image accompanied the comment that the Moon was at a low altitude and that it was not his best work.) During the first 30 secs, the brightest part of the ejecta plume will create a contrast effect as the surface brightness of the brightest part of the ejecta plume removes the contrast between the crater shadow and surrounding terrain. For visual observers, the crater may see a blurry mesa effect. Caebus A1 will look like a terresterial mesa or will have a slight "raised muffin" appearance. Where the brightest part of the ejecta cloud crosses the sunlit portion of the lunar surface, it will not be visible due to lack of contrast between the cloud and the lunar surface. For low resolution imagers like myself, the crater will simply "disappear" due to this contrast effect beginning near the 30 second mark for a duration of about 30 seconds and then mysteriously reappear on AVI frames. See my 9-10-2009 image for an example of low-resolution image: http://tinyurl.com/qw32fy Although no images have been gathered that match the libration and illuminated fraction of the impact, the reduction in libration in latitude to -6.0 to -3.0 between now and the impact on October 9 will only increase this contrast effect. The Cabeus A1 crater shadow will be relatively thinner (in arscsecs) on the day of impact as compared to that shown in Stefan's image. Imaging of the LCROSS impact will be a fairly straight forward process for amateurs. High focal length imaging is preferred in order to minimize the percent of the sunlit lunar disk captured in a frame. High focal lengths dictate that large pixel DSLR cameras and CCD cameras are disfavored relative to small pixel sized fast moderate and high-end lunar imaging cameras. See Sinnott's Effective Focal Length to Pixel Size nomogram, url - http://media.skyandtelescope.com/images/Linked.gif - and the more detailed discussion in another message in the LCROSS Observation newsgroup - Post by K. Fisher 9-8-2009 LCROSS Observation Group Efls for imaging the LCROSS impact http://groups.google.com/group/lcross_observation/msg/764ceeede969207a http://tinyurl.com/owkdf7 Pre-impact image calibration is an easy three-step process. The goal of this process is to set the pixel value of the brightest edge of the rim of Cabeus A1 to 75% of your camera's well capacity in ADUs. This should assure that the full range of pixel values that can be captured on a line profile across the major axis of crater Cabeus A1 are recorded on images stored to your disk. First, focus your imaging gear on the target crater without concern for the exposure setting. Second, slew to 2.6 stellar magnitude theta Auriga. On the morning of the impact, the Moon will be between the horn stars of Taurus and just next to 1.7 mag bet Taurus (Alnath). 2.6 magnitude theta Auriga is one of the figure stars of Auriga and is about 11 degrees away. Nearby alternative stars for exposure calibration around 2.5 stellar magnitudes include: zeta Per 2.8mags B0.5V, delta Orion 2.2 mags O9.5II, gamma Gem 1.9 mags AOIV, beta Auriga 1.9 mags A2IV. Take some test images of theta Tau and adjust your exposure setting so theta Tau's brightness peaks at 50% of your well capacity. Note that the preview histogram in some image capture software _does not accurately_ represent what is stored in captured images on a disk. Open the test images stored on the disk and run a profile measurement or histogram on your image of theta Aur using your image processing software so you are sure your exposure setting captures the right amount of well ADUs. Keep this exposure setting and slew back onto the impact target Cabeus A1. Take some test frames and look at some of the raw images on your disk. Use the profile measuring line tool (e.g. one is available in AIP4WIN) and take a profile of the pixels that cross the major axis of Cabeus A1. Now adjust your exposure setting so that high pixel value of Cabeus A1 rim is at 75% of your well capacity. The minimum pixel value shown for the Cabeus A1 crater line profile should also be within the range of a histogram made of the entire test image. You can slightly adjust back from this exposure setting so the Moon is not overexposed _on images stored on your disk._ Again, _do not trust the preview image and histogram_ in your image capture software. Post-image processing will favor software packages that offer region masking like Photoshop. This way individual regions of the bright lunar surface can be supressed in brightness, but pixels that encompass the area within the Caebus A1 crater rim can be selectively stacked and gamma stretched. The impact will provide imagers interested in hobby science with an opportunity to study plume kinematics using photometry measurements from their images. The LCROSS ejecta plume will rise 5 kilometers (5000 meters) to its maximum brightness in about 30 seconds. The vertical plume speed is estimated at an average of 167 meters per second (5000/30). For the first 2 kilometers, the ejecta cloud will be masked from Earth view by Cabeus A1's crater rim. This trip above the crater rim will occur between about 18 seconds after impact through impact + 30 seconds. (30 seconds * 3000 meters / 5000 meters). For the final three kilometers, the plume will be sunlight and the total light from the cloud may have a changing photometric signature related to its vertical travel that will be recorded by amateur video imagers. Such recordings might be examined to extract a plot of the total brightness of the Earth visible ejecta curtain against time. The process for making such a recording and reducing it is generally described as follows. An LPI camera of video that records AVI files including both an audio track and a video track will be needed. For the time signal audio track, the video can be time stamped using a digital metronome as the timed audio source. An inexpensive $30 Ibanez model emits a good sharp tone at a maximum of 180 beats per minute with a different second identifier signal and is available at many local music and guitar stores. url: http://www.ibanez.com/electronics/product.aspx?m=MU40 . Alternatively, imagers can use a more expensive video time stamping rig favored by lunar occulation observers - the KIWI OSDI video-GPS timestamper. url: http://www.pfdsystems.com/ . The NIST WWW shortwave time broadcast is another option for a timing audio signal, but since there is no need to coordinate observations between observers and clear reception of the NIST shortwave signal is usually problematic, the Ibanez metronome may be the better inexpensive no-hassle option. Post imaging, the AVI file is reviewed using movie making software. The individual frames that contain identifable time beats and good images of Caebus A1 are separated. Then each time-stamp identified image is reviewed in image processing software. Most image processing software (such as AIP4WIN) contain a region measuring tool. These tools count the average value of pixels in an identified circular or rectangular area. Use the region measuring tool on each time-stamped image and surround all of Cabeus A1. It will be important to use the same relative pixel coordinates from the center of Cabeus A1 in each individual frame. Note the time stamp and the average pixel value for the uniform measuring region on each frame. Then plot the pixel values against time. Finally, compare your plot against the predicted increases in plume brightness that presumably will be provided by the LCROSS Team. In conclusion, the LCROSS impact can be easily imaged, assuming it reaches the brightness of 4.0 mpsas stated in LCROSS pre-impact modeling. The view may not be as dramatic as one might imagine, but it appears certainly worth trying for. The success of amateur imaging is dependent on the LCROSS Team gathering and publishing for amateur use, the apparent brightness of the surface area around Caebus A1 and the shadowed portion of Cabeus A1 in both mpsas and V stellar magnitudes. If the shadowed floor of Cabeus A1 is brighter than 4.0 mpsas, the impact cannot be observed or imaged by amateurs. By the NASA LCROSS Team calling for amateur imaging and by inducing, through press releases stating that the impact is observable, the public's attendence at private star parties, the LCROSS Team has undertaken the business ethical obligation to gather and publish such photometry data prior to September 27 and before Oct. 9. This ethical obligation is also incured by their dual role as scientists and public governmental employees. September 27 represents the last date in which the south lunar pole will be at 71% illuminated fraction and on which amateurs can make useful "dry runs" of their imaging gear. Advanced amateurs with photometric gear may wish to gather and share their hobbyist studies of the apparent brightness of the shadows of small craters on the opposite east side of the southern polar Moon on September 27. Again, this is an amateur note. Comments and criticisms on the same are welcomed. Clear Skies - Kurt Bohlin, R. C.; Gilliland, R. L. 2007. Hubble Space Telescope Absolute Spectrophotometry of Vega from the Far-Ultraviolet to the Infrared. AJ 127(6);3508-3515. http://adsabs.harvard.edu/abs/2004AJ....127.3508B Keiffer, Hugh, H. and Stone, Thomas, C. (U.S. Geo. Survey, Flagstaff, AZ). 2005. The Spectral Irradiance of the Moon. AJ 129:2887-2901. http://adsabs.harvard.edu/abs/2005AJ....129.2887K http://www.iop.org/EJ/article/1538-3881/129/6/2887/204408.text.html