Welcome, Guest. You are not logged in.
Log in
Submitting Variables to VSX

A full manual in PDF format, containing many examples, may be downloaded here.


We encourage individual submission of new variables to the Variable Star Index (VSX). These can be single stars that you have discovered through your own observations or through data-mining; batches of stars, such as from a publication that is not already included in VSX; or changes/modifications/comments on existing VSX stars.

As opposed to comet or asteroid discovery, you will get little or no credit for discovering a new variable star. It is pure science and the willingness to contribute to the community. A new variable will typically have some assigned name (from an existing catalog like GSC, 2MASS, etc.), and sometime later, it may also have an assigned GCVS name (this can take years). You may provide a discover's abbreviation and sequence number, but there is no guarantee this name will stick or will be used by the community. There are rare occasions when an unusual object garners attention, and an informal discoverer's name gets attached: McNeil's Object, Sakurai's Object, Barnard's Star, etc. Don't count on it, though!

Adding stars to VSX helps the community. When studying a particular star, observers typically use one or more neighboring stars as comparison stars. By letting the community know about every variable, observers can search VSX when selecting their comparison stars and know that they are not selecting a known variable. Adding variables to VSX also broadens the database of variables, giving more selection to researchers who are interested in a particular class, period, location for their studies. You may think that adding only one star is not important, but it is like voting - thousands of people adding one star apiece can be extremely important, and your particular star might be that really unusual one that everyone wants to study.

So, let's get into the details. While the discussion below is geared towards someone submitting original observations leading to the discovery of a variable, you can also do data-mining of existing catalogs like ASAS and NSVS. Just beware that there is a double standard: we are more critical of data-mined submissions because we expect the submitter to do some work, such as period analysis, rather than just regurgitating information already available from a given survey's site. Also, don't be too critical of objects already contained within VSX: we have accepted complete lists of previously published objects rather than applying criteria to each and every object on those lists. New submissions are subject to more strict quality control.


Keep the following in mind before submitting a new object or revision:
  • Read the whole guidelines before attempting your first VSX submission.
  • At least a basic knowledge on variable star light curves and analysis is required. Moderators will help you through the submission process but cannot teach you how to analyze a variable star.
  • Observers undertaking systematic searches of variable stars resulting in a large number of discoveries are encouraged to submit their results to journals like the JAAVSO. For bulk uploads to VSX, contact the administrators.
  • Check the literature (VizieR/SIMBAD) to confirm your variable star is a new object. We try to keep VSX up to date but there are lots of recent lists/papers that are still not included in VSX and you cannot be credited as the discoverer if the object has already been published (you are welcomed to submit it with the proper reference anyway).
  • Make sure the object is identified correctly. Give the J2000 position based on Gaia DR2 as given by VizieR or from another astrometric catalog such as UCAC4 if the object is not present in Gaia DR2.
  • If you find a new variable, you should be able to combine your new data with archival survey data like those from ASAS-SN (all-sky ASAS-SN observations have become available between Vmag. 10 and 16.5). CRTS, ASAS-3, NSVS, SuperWASP or other available and relevant datasets depending on the star's position and magnitude should be checked and used if they are useful to get much better elements and secure the object's classification.
  • VSX is primarily a catalog of Galactic variable stars. Variable stars in globular clusters and extra-galactic objects like supernovae and quasars are normally not included, unless for historical reasons or for AAVSO observing campaigns.
  • Only 5 submissions per week will be reviewed per user/group.
  • If you are performing a systematic search of variables resulting in lots of new objects, we encourage you to publish those results in a journal (e.g. JAAVSO or PZP) instead of submitting them individually to VSX.
  • Submissions that have still not been accepted after two months (60 days) when modifications were required, will be deleted by the system.


To make it into VSX, the object must be proven variable. What constitutes proof, though? Here are some basic guidelines. We will bend the rules in unusual circumstances, but in general, if you follow these guidelines you have a very good chance of having your favorite object added to the system.

a. Variability must be determined beyond a shadow of a doubt. If you submit few datapoints, you need to justify your conclusion with supporting evidence: large variability (include measurement errors), red color/mira typing using 2MASS, etc. We strongly encourage that you observe your object often and submit a true light curve rather than a few discrepant measures. Submission of discoveries from visual observations is discouraged, except for transient objects like novae. If you are suspicious of visual variability, we recommend posting to the AAVSO Visual Observing Forum and asking questions or requesting confirmation on this or another AAVSO Forum.

b. Unfiltered observations are discouraged. There are many systematic problems with unfiltered observations. If you must submit unfiltered photometry, you need to check a plot of airmass for the observational dataset so that you can see if your variation mimics or mirrors the airmass change (a common problem). You should also check the color of the object and of the comparison stars, if possible (Tycho B-V, 2MASS colors, something that helps you understand the color differences). We may request more supporting evidence for unfiltered submissions. Observing with at least two filters is highly useful for discovery work—you can often roughly classify a star if you have its color and light curve shape.

c. For transient objects, such as novae, we strongly recommend that you first submit your discovery to the IAU Central Bureau for Astronomical Telegrams. That disseminates the information the fastest, and also is the avenue for obtaining confirming observations of your object (CBAT usually contacts the AAVSO for followup observations). You can also announce your discovery on many of the alert maillists, such as vsnet-alert and baavss-alert.


a. First, you need highly accurate coordinates of your new object. There are many software packages that can do this for you, but you should get a position with no worse than an arcsec or two of error.

b. Minor planets are the vermin of the sky. They get into many images, and when near stationary points, can be very slow moving. Check to see if one of these might be your new star at the SN Candidate Minor Planet Checker. We also highly recommend that your observations be made over several days so that you can look for even small motion.

c. Check to see if this is already a known variable. The best place to start is VSX itself. You should do an area search, as an exact coordinate match is unlikely, and some reported variables had large error in their discovery report. Other sources to search include SIMBAD and VizieR. (VizieR will also give you cross-identifications and photometry from existing catalogs, useful information to submit with your report). When you search by position in VSX you will get links to search by position in a number of other catalogs without having to enter the position again.

d. If you are observing unfiltered, don't compare your results against any of the photographic surveys (DSS, POSS, etc.) to demonstrate "variability". Many red objects will appear brighter on your images than on the archival plates; this can often be many magnitudes of difference.

e. Look carefully at your images. There are many defects that can lead to false variability. Airmass effects were already mentioned, but there are ghost images of bright stars, scattered light, hot/cold pixels that migrate through images during time series, etc. Think of any way that a problem could effect your data and guard against it before assuming variability. If you are uncertain, ask the experts on one of the maillists that have been mentioned, or on the AAVSO Forum, etc.


a. You need accurate coordinates. Anything with more than an arcsec of error is discouraged. It is easy to get subarcsec precision with the modern astrometric reference catalogs now available. If the object appears in some common catalog, you will be asked for that cross-id. Coordinates should come from an astrometric catalog. Use UCAC4, PPMXL or 2MASS (when Gaia DR2 is released in 2018, use that). Some common examples of cross-identifications:
  • GSC 01234-06789 (5 digits before and after the dash, use leading zeroes if necessary)
  • 2MASS J11431012-5804040 (use a J before the coordinates)
  • USNO-A2.0 0300-13671194
  • USNO-B1.0 0319-0360318 (use a dash between USNO and the catalog version, it is part of the acronym)
  • GSC2.3 S111210165373
  • UCAC4 810-003941
b. You need to provide the maximum and minimum magnitude. If you have calibrated reference stars in your image, use those to determine the zeropoint and the target magnitude. Use AAVSO sequences, ASAS-3 (V-filtered) or VizieR catalogs like APASS, CMC15 or UCAC4 (through transformations) to get calibrated maximum and minimum magnitudes in a standard passband (preferably in V).
You can also use USNO-A2.0, USNO-B1.0, GSC2.3, etc. if the star is too faint and not present in the other surveys, but you should indicate the type of magnitude and its source. If it is impossible to provide an accurate calibrated minimum and maximum magnitude in the same passband you may provide a magnitude (e.g. in V) derived from one of the catalogs mentioned above (which may be considered to be the mean magnitude for stars with a relatively small amplitude) and the observed amplitude in another passband (e.g. clear, without filter). For outbursting stars like novae or dwarf novae however, the maximum magnitude should always be given.
Note: throughout when maximum magnitude is used, it is meant to be the magnitude at maximum, when the star is at its brightest and thus with the smallest number for the magnitude (e.g. a dwarf nova in outburst) as it is customary in variable star studies. The minimum magnitude is then the magnitude at minimum, when the star is at its faintest and the number for the magnitude is largest (e.g. an eclipsing binary at mid eclipse).

c. Submit supporting evidence. We request at least a plot of the photometry with time on the horizontal axis and magnitude on the vertical axis (brightest at the top, faintest at the bottom). If you have sufficient information to analyze the light curve, determining period and classifying the star, then you should also submit a phase plot. You can also include the actual data in a tabular form as it may clarify suspicious light curves or help in a better determination of type and/or period later on. Submitting the data in tabular form only will slow down the acceptance process as someone then has to plot that data. Do not submit the actual data under the assumption that this Web site will archive that data, as VSX is not the AAVSO International Database. We do not guarantee that any submitted data will be retained.
The New Star Form gives you the choice to upload two supporting documents. If you have more than two plots to upload (e.g. finding chart, phase plot of one period and phase plot of another period), you can save a draft of your submission and then you will be able to edit your draft and upload as many files as you want (saving a draft version each time you upload a file).

d. Most 14th magnitude stars are easily identifiable on the DSS with good coordinates. However, for crowded fields, or very faint stars, or stars that do not appear in one of the major catalogs like Gaia or 2MASS, you should submit a finding chart. The rule of thumb is that if there is any chance for confusion, identify. You want people to be able to find this new variable! These finding charts should be in jpeg, png or gif and reasonably small (don't submit bitmaps or fits files!); north and east marked, and some indication of scale (verbal is ok).


In general, the data submitted to VSX for new variables found in (public) survey data should be as complete as possible: the position, variability type, and maximum and minimum magnitude should always be given. For periodic variables also the period and epoch (time of minimum for eclipsers, and of maximum for pulsating and eruptive variables) are required.

a. Always give the id of the object from the survey, e.g. NSVS 123456 or ASAS J123456-7840.1.

b. The positions given by surveys like NSVS and ASAS-3 are not very precise and may have errors of 10" and more. If it is possible to unambiguously link the object to one from an astrometric catalog, give the id from that catalog and use its position. Depending on their brightness, stars closer than 60" from each other may be blended and measure as one object in some of the surveys, try to choose the most probable identification based on colors or magnitudes.

c. Do not attach the light curves provided by the NSVS or the ASAS-3 web site, nor the NSVS or ASAS-3 data themselves. These are easily available from within VSX. Instead provide phase plots (for periodic variables) or combine light curves from several surveys into a single plot, making sure data from different surveys are distinguishable. When two or more synonyms are present in the NSVS database, combine these on a single plot as well, preferably also with different symbols and/or colors. Data from other surveys for which there is no link-out from VSX should be attached or an url should be provided.

d. Some general remarks on the use of data from particular surveys:
  • The date given by NSVS is MJD-50000 = JD-2450000.5. VSX expects a JD for the epoch, so add 2450000.5 to the NSVS date (and make the heliocentric correction)
  • Use mask 6420 for NSVS data.
  • In general, do not use ASAS-3 data of category D.
  • Beware of low quality nights and different zero points in SuperWASP data.
  • Submissions based only on Tycho photometric data are not acceptable.
e. Some notes about the resolution of stars by particular surveys. Stars closer than the following distance will be blended:
  • NSVS: ~55"
  • SuperWASP: ~60"
  • ASAS-3: ~23" (up to 40" in the case of bright stars, e.g. mag. 8-9)
  • APASS: ~12"
  • CSS: ~11"
  • SSS: ~9"
  • MLS: ~7"

a. A number of variable star catalogs has been imported into VSX. However a large number of variables have been described in the literature, which have not been assigned a GCVS name and have not been incorporated into VSX. For other variables up-to-date information exists in the literature (e.g. corrections of GCVS or survey data) but it has not yet been entered into VSX. Your help to make VSX more complete and up-to-date is also welcome.

b. Provide all data found in the paper. Use the position as given the paper, or provide a better position (from an astrometric catalog) based on the id given in the paper. Information which cannot readily be placed into the standard VSX fields, should be added as a comment. E.g.: magnitude at secondary minimum for an eclipsing binary, epoch of secondary minimum for an eccentric eclipser, etc.

c. Provide the bibcode of the publication when it exists (in which case a link should not be given). For preprints published on astro-ph, use the temporary bibcode (e.g. 2014arXiv1401.0635T). When a bibcode does not exist, give the url of the online publication whenever possible. When a publication cannot be accessed online, provide a scanned copy, so that the moderators can verify your submission. Notes in email discussion groups such as vsnet-alert, cvnet-discussion, etc. are acceptable as references when they provide enough information to decide that the object is a variable (in principle the same rules as in sections I and II are valid here).
If you have more than one reference to give (e.g. data from multiples sources), you can save a draft of your submission in the New Star Form and then you will be able to edit your draft and add them (saving a draft version after each reference you added).

d. If there are more than 10 or so new objects or revisions available electronically in tabular form from a single publication, send a message to vsx at aavso.org. Often these can be more easily imported into VSX, rather than be entered manually. Preparing a list (a text file or an Excel spreadsheet) with the relevant data separated in columns will save you (and VSX moderators) a lot of time.


Make sure to double-check everything before pressing the submit button. There is a "Draft only" checkbox at the bottom of the New Star form that will let you work with your submission until you are sure you are ready to submit it. Use it.
Be sure to use the Spot check link in the Position field and the"Check name" link in the Primary name field to avoid duplicates.
Once you have submitted a new star for inclusion in VSX, the request is sent to a group of volunteer "moderators." These folk are experts in variable stars, usually having looked at thousands of light curves and often long-time observers themselves. One or more of the moderators will look at your submission. If you have been diligent and have submitted an obvious candidate, it will often be approved within a very short time (up to a few days), depending on the workload.

There will be submissions where the moderator thinks insufficient or incorrect information has been submitted. He or she will then usually contact the submitter, asking for clarification. Do not consider this as confrontation or lack of respect; these people are truly trying to help. They may offer suggestions as to how to submit later discoveries to make their process easier in the future. Submissions that have still not been accepted after two months (60 days) when modifications were required, will be deleted by the system.

Also, do not be angry or discouraged if your submission is rejected. The moderators will give you their reasoning for the rejection. Read it carefully! You will get rejection often in life, and astronomy is no different. Most professionals submit dozens of proposals during their career with the majority being rejected by their peers. Learn from the comments and try again!

At any time, the moderators may invoke their right to revise or otherwise update any submission made to VSX for clarity, consistency, or convention, whenever they believe the modification can be made without adversely affecting the integrity of the submission. Typically, you will be informed of any changes made by the moderators.


Note that these comments are geared towards submission of individual objects to VSX. VSX submission does not preclude sending your data analysis to some other additional source, perhaps as a formal paper for publication. We do not retain any kind of right to your data. There are many journals that accept variable star research, and we recommend that you look at them all carefully. The ones we have found most useful are:

OEJV (http://var.astro.cz/oejv/)
A good basic journal. Electronic only. Now refereed.

Peremennye Zvezdy (http://www.astronet.ru/db/varstars/)
Run by the GCVS folks. Electronic only. Refereed. The supplement contains primarily observational results and matches most discovery projects.

JAAVSO (http://www.aavso.org/publications/jaavso/)
Run by the AAVSO; fully refereed and both electronic and hardcopy. There is a page charge for non-members.

JBAA (http://www.britastro.org/jbaa/)
Primarily for members of the BAA. Nice journal published 6 times per year. No electronic version.

PASP (http://www.journals.uchicago.edu/PASP/)
This fully refereed, professional journal will accept quality articles from amateurs. Page charges.

Many other professional journals accept variable-star articles, but the ones listed above are the most common. Good luck!


Check list for new submissions from your observations (Novice discovery):

  • Get accurate coordinates - preferably Gaia DR2 J2000.0 positions as computed by VizieR (available through the VSX external links after a VSX coordinate search).
  • Ensure that the object is truly variable - visual inspection of a handful of images at the location of the star, should suffice.
  • Check VSX (or other databases) to see if the object is a known variable.
  • Determine the variability range and provide the passband of the observations or alternatively give a mean catalogue magnitude plus the recorded amplitude.
  • Create a plot of your data to submit with your report (JD vs magnitude for irregular variables and phase vs magnitude for objects with sufficient information to derive a period).

Check list for new submissions from existing archival (survey) data:

In addition to the above,
  • Provide the name of the star from the survey.
  • Provide phase plots (for periodic variables) or combined light curves from several surveys (if available) into a single plot, making sure data from different surveys are distinguishable.
  • Provide the proper survey references.