I wrote a simple Perl program that queries the VSX api for observations on the small set of stars I monitor and plots light curves for some recent period.
I have seen a couple of cases where an obviously discrepant observation makes its way into the chart. An example of this is the recently highlighted observation (V) of khi Cyg of mag 20.272, on JD2457910.96764. This observation does not plot in LCGv2, and is clearly marked as discrepant in LCGv1, so the observation has been clearly identified as an anomaly, but it still appears in my query results (see attachment, within the orange circle).
I searched the scattered information on the API to look for an attribute that indicates that the observation has been singled out as an anomaly, but I wasn't able to find one. How can I identify these discrepant observations? Does such an action require some kind of connection to Zapper?
Thanks in advance for any hint!
Hernán [DHEB]
Hi Hernán,
There is a data field explanation file (normally seen when downloading data using the on-line form) that discusses the various values of the validation flag. You can find it at
https://www.aavso.org/format-data-file
The discrepant value is "T". Note that all discrepant values are not flagged; only the ones that have been examined by staff and decreed discrepant. For other submitted measures, it is usually up to the researcher to decide whether they are good or bad. I'm sure the staff would like to hear of any possible AID entry, either through the use of Zapper, the LCGv2, or perhaps an email.
Arne
Thanks, Arne!
Identifying and avoiding rows where the "val" attribute is "T" solved the problem.
Best,
Hernán [DHEB]