[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
A New List of 10
I am working on refining my algorithm for finding new variables in the
data. I warn everyone that I am doing this for recreation. Mostly I am
working to take data and get it to the data base. From time to time I look
at the data. This helps to uncover problems as one of the plots attached
illustrates. I am not keeping records of this work. So If you don't, then
any possibly interesting information will be lost until it is found in a
later rigorous analysis.
For the following list, I just started at an arbitrary point in the TOM1
"probably variable" output list and looked at 10 stars in a row by plotting
them and looking them up in VizieR. The first columns are the position,
then the WS statistic, then any comments or the result of the VizieR lookup:
07 52 05.9 -03 45 16 39.55 LPV?
07 52 52.7 +13 22 56 250.71 LPV?
07 54 07.0 +03 39 20 303.86 SPV?
07 57 43.3 -00 41 05 1500.81 GCCS 1960
07 59 50.1 -03 28 36 1611.71 SPV CMC 1007680
08 04 25.5 +04 27 51 188.45 IRAS LPV?
08 04 30.6 -03 07 49 290.33 LPV?
08 06 21.6 +03 23 02 196.33 IRAS LPV?
08 06 38.5 +01 55 46 489.50 YY Cmi EB
08 08 49.5 -04 10 22 264.72 IRAS LPV???
I have attached one plot from this list, the fifth item, that might be of
interest to one of you.
As you can see, only one of the stars was found in a VizieR search. Of the
rest, all were obviously variable except the tenth item which is an IRAS
source so I guess it is not surprising that it appears to be variable.
Define "obvious" as > 3 sigma. I think the current scheme is thus pretty
good for finding new variables. This will be one of my recreations, keep
trying to improve the search scheme.
I have done enough of this now so that I can observe that some areas of the
sky have a much higher percentage of "found" variables than others. BP
said this was true. I think he is right.
The JO71829.3 star is put up for someone to explain. I think the possible
errors are such that we I cannot explain the data by the error. One
problem with our VI pair requirement is that red stars tend to disappear
from our data when the V component drops into the noise, as in this case.
So we do not get a clue as to whether the I component is also changing.
Tom Droege
J075950.1-032836.png
J071828.3-004041.png
