Epsilon Totality-OOE / Eclipsing object shape

Hi Roger, Nice plot! It is very similar to what SMEI has observed so I think your analysis is going in the right direction. To answer your question, the Feb. observation had extremely minimal UV coverage (amounting to about two observations). Nevertheless, we were able to salvage enough data to show that if the southern spot was there, we should have seen something. I take this to mean the disk is thicker. Over the last few weeks I've been putting together a model for the system that uses all of the interferometric and astrometric (i.e. orbital information) at the same time. I have to solve for everything all at once which makes it a difficult problem. In the mean time, the poster shows some diameter estimates using single-epoch model fitting. In this frame, the disk appears to show little variation in thickness (perhaps growing slightly) once you consider an ~5% uncertainty. See ya, Brian

Updating on March 7th, epsilon now back to a steep brightening ! Egress engaged ? Green curve: observations details showing the OOE Red Curve: OOE subtracted Yours truly, Roger

The 0.12 mag step occured unresolvable close to the seasonal gap / solar conjunction and mid-eclipse. This fact should make everbody cautious:

• - Either its real. Then its probably related to the mid-eclipse and an asymetric disk.
• - Or its a systematic error. Then its probably related to the seasonal gap / solar conjunction (e.g. due to an hour-angle effect in the vis obs or an not 100% extinction and/or color correction that are often noticable for far northern/southern vars in visual data and also measurements before vs. after the seasonal gap / solar conjunction).
• - Or its accidential due to some kind of unexpected/unknown variability.
• - Or its a combination of two to all of these points.

As the SMEI LC doesn't show such a step, I would favour that its a systematic error in the ground based data. CS Wolfgang

HI Wolfgang, Brian, It seems to me there is something strange in the SMEI data. I have made the exercise to superpose the two curves at the same scale. The SMEIcurve shapefit reasonnably well at the beginning and at the end of the totality period covered. But between JD 5260 and 5440 the SMEI seems translated by 0.12 mag and the (few) existing data of thetranslated section seems well following the variation of the ground based observation (To seeSMEI_Eps-OOE3.png) Surprising no ? I would admit the conjunction period was difficult( ! ) and I have used a very specific color and extinction technique to process it (determining the extinction gradient in the image using a group of star, the gradient being characterizedboth in amplitude and direction, the technique also includes a blue extinction correction. Anyhow the translation of the SMEI curve occured (both ways)at dates atwhich therewasNOISSUE with the extinction and the background: about at 50~60 deg above horizon at the stepping-up and 30~40 deg atthe stepping-down. I have also made the exercise to superpose my curve (the green one)on theresults collected by Jeff Hopkins and his observergroup. The green curve shape fits very well the average of their results. That means a large number of ground observers have seen the same thing. Brian, have you any idea of the reasons for such difference ? My second point is the last update of my observation. The brighteningkeeps in progress,even steeper ! The slope of the curve is anyhow much steeper than such of the OOE at any time of the records. And at present the OOE is supposed to be in a dimming phase ! (To see Eps-OOE3.png) Last result: 11/3/11 3.655 SE:0.001 (My data are systematically 0.03 brighter than the Jeff's group average) Yours truly, Roger

Hi Roger, I haven't noticed anything that seems odd in the SMEI data. It does appear to show a brightening in/around mid-eclipse time which we think is real (we don't see through the disk though) and is probably caused by less absorption because there is less material in the way (when going through the disk... think doughnut). I've seen a similar offset in the light curve on Jeff's site so I've considered it to be real. Keep in mind that SMEI has raw, unfiltered CCD cameras (three of them to be exact) and that the SMEI magnitude isn't like any other magnitude system. We should see similar trends, but comparing it with V, or R, or I isn't a good idea. SMEI's bandpass is very odd and spans a little bit of V, all of R, and some of I-band. I haven't a copy of the data with reference stars on hand so I can't comment further... sorry. Brian

Hi Brian, I have not attached the superposed graphs, mine and Jeff's one, as Jeff requires permission for reproducing his data. But I can say our data are very similar and I do not see at all that "rectangular step up" from JD5260 to 5445 in our common data. I did check the R and I curves fromJeff, they are much less detailed but they look similar to the Vj one showing also a brighteningat thesame date (JD 5380) The 5260-5445"rectangular step up"subtracted, the SMEI curve looks very similar to the ground base observations, including the 0.12 mag brightening at JD 5380. The "rectangularstep up" isinterestingly perfectly centered on the conjunction date(~5354)and symetrical ? It would be very interesting to see similarlight curves from eta AUR and zeta AUR from SMEI. Yours truly, Roger

Hi Wolfgang, It seems you dohighly rely on SMEI. Did you check its photometry technique ? On my side I wonder how a camera with suchonedegree wide PSF, very asymetrical, variable depending position, rotating depending date... would permit an accurate stellar photometry. The catalog for the background elimination is limited tomag 6, the comparison stars are faraway from eps AURlocation...And itworks on white light. The mission of SMEI is no way the stellar photometry but solar meteorology. Usage for stellar photometry seems only secondary. Yours truly, Roger

Hi Roger, You are correct in your accessment of SMEI in that it wasn't designed for photometry. It's name says everything: Solar Mass Ejection Imager. It does, however, do photometry very well. The satellite was designed for 0.1 % photometry and often performs at or close to that limit for stars. The calibration process for SMEI is indeed cumbersome. John (Clover) has done much to improve the accuracy and characterize the long-term stability of the system. The PSF, albeit complicated, is actually very well determined and their fitting routines do well. When I have inspected the traditional calibration stars (eta Aur., lam Aur) they have been stable to the limit of the instrument, so the changes we see in the F-star are likely to be real. After the eclipse has finished John and I will write up a paper discussing the data and release all of the SMEI data (2003-present) on eps Aur so that you all may inspect the data as well. I just re-read your post as I think I missed something. I think the jump between 5260 - 5440 peaks on April 1 is something that hasn't been fixed in the calibration process. All of the calibration stars have a U-shaped bowl that hits minimum in the middle of Camera 2 and hits maximum near the edges of the CCDs (i.e. near the gaps). It is likely that this still contaiminates the data. John and I will address this issue using more standard calibration routines. As with any space-based instrument, you can get great data, but it often is a lengthy calibration process. Brian

Hi Brian, Thanks for your explanation, it's clear that huge flow of data will need long time processing to deliver all its secrets ! We could remember Hipparcos needed years to end up with Tycho 2. Indeed SMEI is aclever space mission ! Yours truly, Roger