I recently posted on the forums area unconfirmed stellar variability discovery results I obtained by analyzing available photometry data from a CCD survey telescope. These results indicate that maybe for the first time this legacy star is shown to be variable.
Is the approach of obtaining results not personally observed but instead derived from available surveys acceptable for AAVSO posting or passing off to VSX scrutiny?
Suppose, for e.g., I grabbed some Kepler raw data and did variability analysis. Would AAVSO accept that?
In actuality I have been part of a small not-so-Citizen Sky nor Zooniverts team of amateurs analyzing CCD photometric data from a survey instrument. So far I have "discovered" around 50 stars that exhibit periodic behavior where none had been noticed before. Mostly these are pulsating variables, but a couple offer clear signs of eclipsing binary transits.
I guess I am looking for a place to park my unconfirmed results where they may be of use to variable star enthusiasts.
The answer is yes. BUT, as long as you are able to provide a complete analysis of the data.
Actually there are more new stars added to VSX from data-mining different sky surveys than original discoveries (with original I mean based on the own observer's data).
Although the submission of individual stars is welcomed, if you have a lot of stars and you can solve them properly, that sounds like a good project for publication in a journal. VSX is not an official publication venue although it ensures you that your stars will be added to VizieR and available to the community.
Or you could have your own group's webpage where the results can be presented and that can be given as a reference, as long as you are able to provide tables with all the required data (Good positions -preferrable from UCAC4-, cross-identifications, variable type, elements, range).
If you decide only to submit them to VSX individually, don't submit many stars a day because we have stars being submitted by many other observers as well (start with an individual submission so we can review it and you can get a feeling of how the moderation process works and if something needs to be changed).
Some points you should keep in mind.
If the variable star is found by data-mining a specific sky survey, use the star ID in this survey as the primary name of your star (e.g. if you used Kepler, identify the star as KIC xxxxxx, if you analyze NSVS data, then give the NSVS identifier, etc).
Follow the VSX guidelines and give the information required in every field of the VSX submission form.
Section IV is about data-mining results.
Use the New Star Wizard the first time you submit a variable, it is a step by step process with useful examples.
So, you should be able to classify your star based on its light curve.
Submit a magnitude range for it (or a standard catalogued magnitude + an amplitude).
Give a period if it is a periodic variable with a phase plot as a supporting document (a JD light curve if it is an irregular variable).
Determine an epoch of maximum for pulsating variables or stars with hot spots or an epoch of minimum for binary systems, RV and dark-spotted stars.
Use that epoch as phase 0 in the submitted phase plot.
The phase plots should show more than one variability cycle for clarity (phases 0 to 1.5 or 0 to 2 for pulsators, phases -0.25 to 1.25 work well for eclipsers so the eclipses are displayed in the middle)
If you are going to submit results based on survey data, combine the results from all the available surveys. E.g. the star may have data in NSVS, CRTS and ASAS at the same time. You can shift the magnitudes to the zero point of the survey that shows standard magnitudes.
In the above example, NSVS and CRTS give unfiltered magnitudes while ASAS has Vmags so use the ASAS zero point.
For normal stars the amplitudes will be similar, for red stars the V amplitude will be larger. Use the V amplitude.
Kepler will give you millimagnitude precission. Try to use data from several quarters to get the best result.
In all cases, the longer the time baseline you have, the more accurate the period will be.
E.g. combining CRTS data (2013) with NSVS data (1999) will give you a 14 year time baseline while if you only use NSVS data you'll have a time baseline shorter than one year.
Remember to keep in mind some database vagaries.
CRTS = MJD dates (you have to add 2400000.5 and apply the HJD correction). Useful range= 12-20 CV mag. Stars <8-10" are blended.
NSVS = Modified MJD dates (you have to add 2450000.5 nd apply the HJD correction). Useful range = 9.5-14.5 R1 mag. Stars <55" are blended (check for companions in VizieR!)
ASAS-3 = HJD dates. Useful range = 7-14 V mag. Stars <25" are blended.
That is just to give you an idea. Brighter stars will cause contamination at larger distances.
Also you need to check the error figures or data flags presented by each survey to see if the results are reliable or not.
E.g. "D" quality data in ASAS are usually bad. Flags 13824, 13826, 50, 256 and others mean NSVS data are also bad.
There are too many databases. Kepler, LINEAR, OGLE, MACHO. Each has its own pros and cons.
You need to use them and get used to them to underestand their limitations.
Give the reference to the survey used in the VSX form reference section.
Don't rush to submit new variables if the data are sparse. You'll probably be able to solve them all combining data from different sky surveys, do it (you can go directly to the survey's data for your star if you make a VSX positional search. You'll get no result if the star is not known to be variable but you'll have the external links menu available so you can click on the survey you choose and get the data). If you can't, try to get additional observations so you can at least give a possible type, and always a magnitude range.
American Association of Variable Star Observers