Observations of the grazing occultation of tau Scorpii on March 11 - New 2007 Mar. 21, 20h UT

You can see Tom Campbell's interesting video recording of the graze below

Chris Stephan led an expedition to observe the graze of tau Scorpii 
by the last-quarter Moon from four stations east of Winter Haven in 
central Florida.  Chris and two others in the group visually timed 
multiple events during the graze, confirming the accuracy of the 
prediction, but several gradual fadings and brightenings of the star 
around the star around the times of the disappearances and 
reappearances led some to suspect that the star might be a close 
binary.  Tom Campbell video recorded the graze along with a Kiwi OSD 
time display; you can see his 24 megabyte .mpg file here.  Rather 
than just clicking on this, which may not work, I recommend that you 
right-click on it, and select from the menu "save as" to download it 
to your computer, and then try to play the file offline.  During 
grazes of close double stars, every, or almost every, event occurs 
in distinct steps, but I don't see such steps in this record, only 
gradual events, sometimes with some interesting structure, that I 
believe can be explained by Fresnel diffraction of the star's light 
at the edge of the Moon.  You can look at the video file and make 
your own decision.  Derek Breit analyzed an .avi digital version of 
the file with LiMovie and also concluded that the observations do 
not show clear evidence of duplicity; I'll post his plots and 
comments here soon.  Derek noted that step events, like he saw in 
his video of a graze of a bright double star, upsilon Geminorum, 
were absent or poorly defined in Campbell's tau Scorpii graze video.  
The observations don't completely rule out stellar duplicity.  For 
very close double stars, the diffraction curves are intertwined so 
that the events look gradual, but just a little more pronounced than 
if the star were single; then it's not possible to tell for sure, 
whether it's stellar duplicity or more gradual than usual lunar 
slopes where the events occurred.  Wayne Warren notes that the star 
has variable radial velocity, a hint that the star may be a close 
double with the companion at least 2.5 mags. fainter than the 
primary so that the companion's spectrum is overwhelmed by that of 
the primary, preventing resolution as a spectroscopic binary. 

Harold Povenmire also observed the graze visually from near 
Marianna, in the Florida panhandle, and observed spectacular 
multiple events, including one where the star faded, brightened, and 
faded out, all over a 4-second interval.  This and the other events 
Hal saw led him to believe that tau Scorpii is binary, which he 
suspected from observations of grazes of the star that he observed 
in 1970 and 1989.