Varuna Occultation Observed After All? – Discussing the Data Recorded by the Aigua Team, Uruguay


From Uruguay:

Gonzalo Tancredi (Departamento de Astronomia, Instituto de Fisica, Facultad de Ciencias , Montevideo, Uruguay) has done a more detail reduction of the set of images taken by Alberto Ceretta and Santiago Roland in Aigua.

(See Blog: Report of the Observation Team at Aigua, Uruguay, Dec 13)


The telescope was a 30cm home made and the detector was a modified webcam (Philips TouCam 740K). More technical details about the equipment can be found in Alberto’s webpage:


The observations were done from the roof of the house of the family García-Gaudino and Bernardo Pombo, in a farm close to Aigua (Maldonado, Uruguay).

Lon = -54 48 03   Lat = -34 12 04


Gonzalo describes his data reduction:


“The sequence of images were recorded in a AVI file, and I have done the following reduction:

* Split the AVI in separate BMP files. Each image has a typical expsoure time of 5.4sec (it was the minimum exp. time for which the occulted star was barely detectable).

* Split the 3 colors (RGB) in 3 different files

* Add the 3 color images in a single Monochrome file giving a weight of 1 to each color

* Substract a Master Dark (created in a similar way and after mean combining them)

* Compute a synthetic flat with median combination of 100 files

* Flat correction

* Align the images

* Add images in sequence of 5, we have done a running mean, i.e. we add images

1+2+3+4+5, then 2+3+4+5+6, 3+4+5+6+7, 

Eveery combined images has a total exposure of ~ 27 sec.

* We have done Differential phootometry. We select a relative bright and stable star as reference (the bright star at the ~ same dec as the occulted star and to the left in Bruno’s latest chart), and 3 check stars: the 14.0 star close to the occulted star, the 11.4 star in the chart, and a star up and right from the 10.7 star.


The reduction was done with Maxim Dl Pro 5.02 and several scripts in Javascript.


I am attaching a Excel file with the results.

There are three sheets: one with the data, and two with the Plots, one for a long time interval and the other for a short one.

 Results of Reduction


Each point corresponds to the photometry of the combination of 5 images of 5.4 sec each, with a total exposure of 27 sec. The Time axis corresponds to the mid time of the total exposure.


The symbols correspond to:

* violet diamond – the occulted star

* red square – reference star

* red triangle – check star 1

* blue square – check star 2

* brown circle – check star 3


We plot for a few points the +- errors, calculated with the following eqs.:

  dM+ = -2.5 * log (1 + 1/SNR)

  dM- = -2.5 * log (1 – 1/SNR) 


There is a deep peak at 2:15:20 UT for the occulted star; where its magnitude has a clear decrease and the SNR becomes very small (~1).

This is very close to the expected time for the occultation!!


I am tempted to conclude that the occultation was detected, but they might be some critics to this conclusion:

* There are some other minor peaks of the same star, specially a few minutes later.

* The check star 1 (of similar brightness) present a noticeable peak a few seconds after the big peak.


Nevertheless, the other peaks are less deep than the supposed occultation peak.


I have done some simulations in order to estimate the duration of the occultation.

I create some synthetic occultation lightcurves and I overplot the magnitudes of the star.

From a qualitative comparison of the curves, the duration of the occultation could be estimated in the range 20-30sec.

I am attaching a plot with this comparison.




It is a pity that this is the only data that we can obtain for this occultation.

But, in view of the difficult circunstances of the occultation, we have done the best that we can.


I am doubtful that we can use this data for scientific purposes, but it would be a nice anecdote that the first occultation of a TNO was detected with a homemade 30cm telescope, with a 100 dollars webcam, from a roof of farm and at a height of 8deg !!

(Have a look to Alberto’s webpage for some nice photos of the campaign)”


From Switzerland:

Raoul Behrend (Geneva) will check the raw data for an independent reduction.


 From Australia:

Dave Herald (Canberra) has done already an interpretation of the data by the Aigua Team:


“I have to say that I am not _convinced_.


I have plotted directly the data columns Obj 1 [E], and Chk 1.


Plot by D. Herald


It is equivalent to Plot longt in the spreadsheet, limited to the purple and green points – and with the y-axis reversed. I have done this as it better shows the light curves. In my plot you can see that there are some similarities in the two light curves to the left of about 2:18:00 – with the blue curve being time-displaced compared to the red curve. Also of note is the increase in the peak noise levels from about the middle of the plot – suggesting that there was a change in atmospherics at around the time of the purported occultation.


Now let me analyse the detail of the light curve. First, let me assume that Varuna has no significant atmosphere. That is, the occultation should be relatively instantaneous, with a light drop to the magnitude of the star. The first test for the occultation is whether the light drop was sufficiently significant – with an expected drop of 5 mags. But the exposure was set at 5 secs because: „it was the minimum exp. time for which the occulted star was barely detectable“. If this is the case, excursions much beyond 1 magnitude would have been excursions into the background noise. In particular, I have trouble accepting that a 2-mag excursion in these conditions represents a real signal, rather than the effects of the background noise.


The second test is the nature of an occultation light curve in the absence of an atmosphere. In essence, there should be an abrupt drop, a flat bottom, and an abrupt rise.

The putative event could be consistent with such a light curve IF you postulate an occultation lasting about 12 secs centered a little after the mid-time of the maximum in the plot – such that the points on either side involve exposures where the occultation started/ended part-way through the exposure. But in the absence of other evidence, the light curve does not _demonstrate_ the existence of an occultation.


The alternative situation to consider is the possibility of Varuna having an atmosphere.

Assuming a duration of around 12 secs, the chord would have been very close to the edge of Varuna (on my calculation, about 98% out from the center of Varunna – assuming a spherical object). The problem then is that if there is sufficient atmosphere to have a significant effect on the occultation light curve, the effect will last for MUCH longer than the solid-body occultation. And the light curve is not consistent with this.


In conclusion:

– I cannot state with any definiteness that the light curve does NOT record an occultation

– there is nothing in the light curve that is particularly characteristic of an occultation light curve. The rational is temporal coincidence alone; and

– there are features in the light curve, and the observing conditions, which make noise a likely explanation.”


From France:

Bruno Sicardy (Paris) wrote:


Dear Gonzalo, Alberto,


 this is amazing! This could well be the first TNO occultation ever observed!!!


 The event is very close in time to the predictions. The Rio group predicted the center of event at Aigua around 02:14:55, Raoul Behrend

(Geneva) predicted an event centered around 02:14:30, and your event is at 02:15:20 or so…


 With your estimate of the event duration of 20-30 sec, and using a shadow speed of 20.8 km/sec, this would correspond to a chord of 420-620 km.

Spitzer data give a diameter of 500+/-100 km for Varuna (Stansberry et al, UoA Book 2008), while older IRAS data gave a diameter of ~1000 km (Lellouch 2002)


 Anyway, keeping the probably better 500 km diameter in mind, this would mean that the event was ~central at Aigua. In that case, the Rio prediction would the closest of the 3 we had at hand (Rio, Geneva, MIT), before the event, see my page:


 This rises a very interesting issue, as the Pico do Dias light curve has apparently a high SNR and is negative (as was Belo Horizonte), which could be used to set firm upper limits on Varuna’s apparent size on that day (if elongated, the object may vary in apparent size).


 So, I do think that this detection, if confirmed, could be of high scientific interest, and should definitively be interpreted and published!


 It also shows that small instruments, used for both predictions and observations, are probably the best tools to detect TNO’s occultations.







The story about the occultation by Varuna is not over yet.

Stay runed at this blog…




Comments are closed.