This occultation would test my ability to detect small mag drops. The predicted drop was 0.24 magnitudes. Since the combined image was fairly bright, I figured the point to point scatter would be less than a factor of two and averaged over a few points, would be fairly straightforward to see as long as the event was at least 1 second. The predicted central duration was 10 seconds, but I was on the edge of the southern 1-sigma zone. I used MovieMaker, then VirtualDub to get an uncompressed .avi file, and ran LiMovie and then Occular.
This is a capture of the image. Athamantis is at upper left and a brighter star is in the red/yellow circles at right. |
Below, in LiMovie I used 7-10-25 radii and had it track with Sych-APT and Link Tracking. I had Occular search for events lasting at least 3 and less than 12 seconds. The center of the predicted event time is 10:35:17 UT which corresponds to frame 958. 10 seconds is 300 frames. The odds of a 10 second central event giving only a 1 second graze is very low and the data is too noisy to detect such an event.
Occular plot of the light curve (blue) and figure of merit (black). There is a peak in FOM about 1 second after the predicted time, but there are competing peaks which are not plausible, and the plot looks consistent with no occultation. |
Noise plot shows Gaussian form, which is good. |
Just in case the photometry aperature was too small and starlight was often spilling outside the circle, I re-did LiMovie with a larger 13/15/25 set of radii. If I can figure out how to bin the data and replot, it would be easier to tell if there was a dip.
The one minute record extracted from the mini-DV tape, plotted by LiMovie. |
Zoomed in to the 23 seconds centered on the predicted event time (frame 958). |
Derek's record, zoomed in. My interpretation? I think he's got a distinctly more convincing event than my record. It would require almost no shift in the path to produce a short event for Derek and a miss for me. Too bad it was so brief and the drop so shallow - very tough to achieve statistical confidence. |