Cosmic fireworks...

...or not. 
We are entering into a very special 2 month interval centered on predicted mid-eclipse, 4 August 2010 (JD 2,455,413 +/- 2 weeks).  Some expect a "mid-eclipse brightening" - as much as several tenths of a magnitude - which would demonstrate a substantial central clearing in the dark disk. 

Based on what we know from the initial round of interferometric imaging, the disk is so close to edge on, that seeing the "hole in the doughnut" is improbable.  However, there are some clues that surprises may await the persistent observer:

- the B star at the center of the disk is a significant source of ultraviolet photons, capable of vaporizing dust, possibly enlarging the central opening;

- the ingress light curve being reported here shows signs of "speed bumps" - short term increases in brightness that correspond with "ring" structure announced by Robin Leadbeater last summer, while monitoring neutral potassium (K I) in the spectrum of epsilon Aurigae (see detailed discussion of this by Leadbeater and Stencel, 2010 http://arxiv.org/pdf/1003.3617 ).  Potassium, a mineral easily ionized, is a sensitive tracer of dust vaporization.  It potentially reveals density variations across the disk of epsilon Aur.  These speed bumps can be more easily seen in our visual light curve data, if a 10 day smoothing is used, or detected with digital photometry.  Of course, the relationship of these to the "out of eclipse" light variations needs clarification.

However, if we interpret these light curve blips as a decrease in the areal coverage of the F star by the dark disk, then a narrowing of the disk thickness can be estimated by the amplitude and duration of each.  The minor speed bump near RJD 55160 lasts only a week or so, and deviates less than 0.1 mag from the ingress trendline.  The K I data suggests a gap between the "C" and "D" rings at that time, with less than a 10% reduction in the disk thickness, based on the magnitude change.  A second, subtle bump appears near 55125 with a ~2 week duration, but only 0.05 mag amplitude at best.  This corresponds to the narrow, K I deduced gap between rings D and E.  The most obvious bump is the most recent, centered near 55275, lasting ~45 days and with an amplitude of < 0.1 mag.  It is possible that this is/was the central clearing, showing up a couple of months prior to predicted mid-eclipse, or another ring-like gap situation.

With Denver students Brian Kloppenborg and RJ Wall, I've been observing epsilon Aurigae and its neighbors with twin 28 inch RC telescopes at our Mt.Evans observatory situated at 14,148 ft in the Colorado Rockies.  The clear air of morning allows for direct visual inspection when the stars are near the zenith, eliminating airmass ambiguities.  We've seen epsilon appearing no brighter than zeta (V=3.75) during all of June.  Using an SSP-4, we've also been attempting to measure epsilon in the near-infrared J and H bands - it remains faint compared to nearby standards like zeta, iota and Capella, although daytime sky variations make this challenging.  In any event, we can state that no strong mid-eclipse brightening was underway during most of June 2010 - epsilon Aurigae appeared to be deep in eclipse. 

Your persistent observing during the next 2 months will tell the tale:
-- will we see another speed bump surrounding predicted mid-eclipse (early August)?
-- are we already in the post-mid-eclipse interval?  In that case, significant brigthening should begin to signal egress soon, if that's the case.  In this system, surprises abound.

Thanks for your efforts, and I hope to meet you at the next Citizen Sky Workshop in San Francisco in early September!

References:
Ring structure in the eps Aur disk based on K I spectra: 
http://arxiv.org/pdf/1003.3617
Mt.Evans photometry with the SSP-4 instrument:
http://adsabs.harvard.edu/abs/2008JAVSO..36..127B

The challenge is that epsilon Aurigae is now a morning star, so extra effort is needed to catch its light.  The good news for those with day jobs is that things change slowly enough so that a once a week observation will help contribute to a complete light curve when many reports are combined.