These comparisons were made with selected frames from the same Digital 8 video recording I used in the MPEG movie clip on my homepage. I oriented them north-up with the aid of the excellent Pic du Midi Observatory online archive, which includes a H alpha coronograph taken at 13:10:34 UT on 2001 June 21 (click here for image). For reference points I used the prominences in the coronagraph that were also visible in my images. With the compositing transformation function in Picture Window Pro 3, I registered the images (2-point rotate and scale) and then placed tick marks at the north and south points of the lunar limb. I registered the resulting image again, this time with Figure 8 (The Lunar Limb Profile) from Espenak's and Anderson's eclipse bulletin (see accreditation above). Finally I performed an absolute difference composite which gives the effect of the figure overlaying the image.
This exercise pointed out the difficulty of judging just when and where second and third contacts occur. From our observation site on the southern edge it was even more so as the roughest part of the moon's profile presented itself. The three sets of predictions come from the Bureau des Longitudes' very elegant eclipse site, Chris O'Byrne's slick Javascript Eclipse Calculator (which employs Espenak's math), and Tom Van Flander's calculations. Tom is the astronomer for Eclipse Edge Expeditions. He did not provide any position angles because, "the contact points near the path edges are dominated by local topography. Moreover, small changes in location make large changes in the angles. Even the assumption made about the Moon's mean radius, average or typical minimum, would make a big difference in angles."
For determining contact times I used the standard Espenak gives in his Fifty Year Canon of Solar Eclipses: 1986-2035 (page 16); i.e., there is no totality "as long as any photospheric rays reach the observer through valleys along the lunar limb". Thus my second contact time is appreciably later than in the predictions. Third contact, however, is very close. This may be because third contact occurs in relatively smoother terrain. |
