[Aavso-photometry] HD37605 data request

[Geir Klingenberg]

Hi Jean-Luc,

Yes, there has been quite a lot of error discussion on this list, I
guess that's because it is not trivial and there's a lot of ways to do
it.

I believe reporting standard deviation in some form gives the best
possible error estimate. So if you are satisfied with reporting a
single error value for the data set you can simply use the standard
deviation of the target star, which in your case is 0.009. Of course,
this assumes that you where not fortunate enough to catch an
ingress/egress and that trends from extinction has been removed.

If the target star varies (as it usually does) the standard deviation
will measure the stars inherit variations and not just error. In that
case one can use the standard deviation of a comp star of similar
brightness.

However, many people prefer to get error estimates for each data point
as this gives a more detailed view of the error throughout the series.
There are many ways to do this as well, one being the method you and
Radu mention. Another would be to use the standard deviation of the 10
measurements you average (if you measure each frame instead of
stacking them first).

Geir

On Sat, 01 Jan 2005 19:58:49 +0100, Jean-Luc <jeanlucla at wanadoo.fr> wrote:
> Hi Aaron,
> 
> I still have data to provide for HD 37605. Since I am not used to this
> level of precision I would like to make sure I will submit data properly.
> 
> Here is what I have done :
> 0/ I used AAVSO (e) chart,  86 (8.58) as comp star and 99 (9.89) as
> check star
> 1/ I took a set of 10 exposures (20 sec + V filter) every 10-15 minutes
> between Dec 28 - 21h00 UT and Dec 29 00h00 UT
> 2/ For each image of the set I substracted a master dark. I then
> combined the 10 of them (with MIRA) in order to get an averaged (mean)
> image.
> 3/ My SNR is good enough to have 2 millimag precision for HD 37605 and 3
> millimag for the Check Star.
> 4/ At this point I got 17  Vmag values for HD 37605
> 5/ The average value of these 17 Vmag over 3 hours is 8,626 with Stdev
> 0,009.
> 5/ I calculated Airmass (between 1.51 and 1.36) and Scintillation (0,004
> - 0,003) for each of the measures.
> 
> Now my questions are :
> 
> a) I don't know the Kv coefficient for my location, so I cannot correct
> my data for first order extinction.
>    Is it OK to report my data without applying the extinction
> correction - and provide the uncorrected data with Airmass instead ?
> 
> b) I am not sure how to calculate Scintillation error impact on my data.
> I read in a previous thread that the err should be calculated as follows :
>    err = SQRT(1/NetCountVar + 1/NetCountComp + 2*SQRT(Scintillation)).
>     If  I do so, the err which was in the 0.002 range jumps to 0.3xx
> ... just because of the Scintillation factor.
>     What does it mean ? Is Scintillation of 0,004 that bad ?
>     Since it does not make any sense to try to detect a 0,02 gap with a
> 0,3x error...what error should I report  ?
>     What is your recommendation to calculate / include scintillation
> error ?
> 
> c) A lot of things are told about about error reporting  and I am a bit
> confused. Is there an AAVSO recommendation to calculate error ?
>    The CCD Manual is giving SNR and Stdev as starting points but many
> emails I read on this list tend to say that it is not enough.
>     Is there a commonly agreed procedure/formula to calculate error
> when submitting data ?
>    (And especially for high precision measurement like for this campaign ?)
> 
> Many thanks for your advices,
> Best,
> 
> Jean-Luc
> 
> 
> Aaron Price wrote:
> 
> > I was wrong about Arto's data set, it does not include the estimate
> >ingress time if Gary's event is indeed an egress. If anyone has
> >observations of HD 37605 anywhere from JD 2453367.2 - 2453367.4 (16:48
> >-  21:36: UT on December 27, 2004) please submit them as soon as you
> >can as this is when the ingress should have occured. Thanks.
> >
> >
> >
> 
> _______________________________________________
> Aavso-photometry mailing list
> Aavso-photometry at mira.aavso.org
> http://www.aavso.org/mailman/listinfo/aavso-photometry
>