That’s a nice image Klaus, and a very interesting topic. Also, very interesting for Lars to locate that report from September 2000. Coincidentally, the author, David Kingsley, is a professor at Stanford in the Developmental Biology department, and was a member of my thesis committee while I was a graduate student in that department. Small world! Needless to say, I was not expecting that when clicking on the link. David, it was very nice to read your report all these years later, and to see you appear on the forum!
The beam in Klaus’s image looked wide enough to me that it should not be overly influenced by latitude changes in the subsolar point, as the axial tilt of the Moon is only approximately 1.5 degrees to the ecliptic. Therefore, I would have thought the beam might occur at every sunset on Maurolycus. The table provided by Dave Mitsky lists the events that occur within a 2 degree restriction of the Sun’s azimuth, and shows that only six such events occur in 2019. If that restriction in azimuth is relaxed to 4 degrees, however, the event does occur at every sunset on Maurolycus, as shown in the table below. The point chosen at which to calculate the solar illumination angle was approximately the western apex of the beam of light (as it occurred on September 19, 2000 at 08:10UT), and the times listed are within 10 minutes of those in Dave Mitsky’s table, but with the addition of six more events because of the relaxed azimuth restriction. Note that for any given observer, not all of the events will occur with the Moon above the horizon.
In order to check to see if this relaxed azimuth restriction has any effect on the perceived beam, I used the Lunar Terminator Visualization Tool (LTVT) software to simulate the illumination at each predicted event for 2019. The results are displayed below as an animated gif. Sorry for the poor quality, but the file needs to be compressed to under 500kb for posting. The view in this simulation is not the view from Earth, but rather from a fictitious observer located directly over Maurolycus, so that the orientation of the image does not change due to libration.
As can be seen in the animation, the beam does exist at each sunset on Maurolycus, although the changing azimuth of the Sun does affect the characteristics of the beam. The beam sweeps slightly from north to south as the subsolar point cycles between its position 1.5 degrees below or above the lunar equator. As the beam sweeps to the south (with increasing subsolar latitude), the beam intersects with more of the central peak of Maurolycus, and this restricts its width to the east of the peak. The red and blue lines mark the sunset terminator, with the width corresponding to the angular diameter of the Sun, and you can also see how the orientation of the terminator relative to Maurolycus changes slightly with changing subsolar latitude.
I looked through my files to see if I had any images of this event. I did not, although I have an image that occurs approximately four hours prior to the event. The image is from last year, and was taken on September 30, 2018 at 11:55UT.
Interestingly, this corresponds to a period of 18 years, 11 days, and ~4 hours after David’s observation from September 19, 2000 at 08:10UT. This is just 4 hours short of a Saros cycle, which is used in predicting eclipses because of the nearly identical Sun-Earth-Moon geometry, but also corresponds to nearly identical illumination, phase, and libration patterns on the Moon. If you locate and zoom in on Maurolycus along the southern terminator, you can see a set of opposing triangles, one illuminated on the west that will become the beam several hours later, and the shadow of the central peak to the east. Although the terminator position in this image is several hours earlier than in David’s observation, this image of the Moon from September 30, 2018 is nearly identical to how it would have appeared on September 19, 2000.
Edited by Tom Glenn, 31 July 2019 - 04:09 AM.