AstroNote 2019-37

Primary tabs

2019-06-23 12:16:05
Type: Announcement-Tool/Utility
The revised treatment of object coordinates (astrometric accuracies) is now active on the TNS
Authors: Ofer Yaron, Avishay Gal-Yam, Eran Ofek, Avner Sass (Weizmann)
Revision of the treatment of object coordinates (astrometric accuracies) was deployed today - 2019-06-23 - on the TNS. Currently 9 groups have a defined astrometric accuracy (that is below the default value of 5 arcsec). Coordinates were updated backwards for a total of 3157 objects, not affecting the actual objects and existing AT reports, and without any merging or splitting of objects. We encourage further feedback from the reporting groups.

Following the details described in AstroNote 2019-15, we have deployed today the revision in the treatment of object coordinates on the TNS, according to the defined astrometric accuracies of the reporting groups/surveys.

From now on, if an additional public AT report is received for an existing object, and this new report is from a group that has a better astrometric accuracy, the object's "principal" coordinates will be updated according to the new report.

Moreover, this revised treatment affects the decision whether to create a new object or associate a new report with an existing one. A new object will be created whenever the separation of a new report from all known objects is larger than the square root of the sum of the defined astrometric accuracies (uncertainties) squared of the new report and existing objects, or if there's a temporal difference of >= 5 years between the discovery dates of the new report and the existing objects.

As part of the deployment of this new mecahnism, we've updated backwards the principal object coordinates for a total of 3157 objects, as listed in the attached log file (sorted by the new group and object names). Comments were also added to all relevant object pages on the TNS. We DID NOT merge or split any objects in a bulk automated way, so there is no change to the objects and their IDs, and no changes to the AT reports - all are kept untouched.

As shown in the table below (and also on the TNS Groups page), we currently have in place values for the astrometric accuracy of  9 specific groups, based on the 95 percentiles (of the separation distributions w.r.t. the Gaia reported coordinates), and we have decided to begin by introducing a more conservative 30% safety buffer (instead of 20%; see AstroNote 2019-15). For the GOTO group, the value is based on private communication. For all other groups, the default accuracy of 5 arcsec is applied, which doesn't mean that this is their true accuracy but just that we have no other reliable value in hand for applying. Therefore, we encourage the surveys to contact us and provide us with comments and feedback. 

In this respect, another rule that is currently implemented is that an object's coordinates will not be updated if the existing coordinates are from a group for which the default value of 5" is set, and the new AT report originates from a group with an accuracy worse than 2".

Last, if an object's coordinates are set by some group and the same group submits an additional report for the same object but with slightly different coordinates (within the defined accuracy), the new most recent coordinates will replace the previous ones as the object's coordinates.


Here are the accuracies as they have currengly been applied for the 9 groups:

Survey 0.95 quantile Applied Accuracy (with 30% margin rounded)
GaiaAlerts ----- 0.2
Pan-Starrs1 0.38 0.5
OGLE 0.43 0.6
ZTF 0.44 0.6
ATLAS 0.83 1.1
POSS 1.34 1.7
GOTO ----- 3.0
PTSS 2.82 3.7
ASAS-SN 3.40 4.4