The question of how to handle sets of single-epoch analyses is an
interesting
one, but this is NOT the data that will be present in the final
publically-queriable
deep catalogue.
Each night we'll be analysing the data for variables, and these
variables
will be stuffed into a database (which will be public, as will all LSST
data). There will be cases of pairs of variable objects that need to be
handled as the seeing comes and goes.
However, the deep detection stuff will be run simultaneously on all the
data that we have in hand, and this will assign unique IDs to OBJECTS
not to detections. The SDSS analysis is a [trivial] example, where we
process all 5 bands simultaneously, and [almost always] correctly handle
cases such as asteroids moving in front of stars [main belters move
a couple of arcsec between g and r], and the fact that the u band seeing
is significantly worse than the i band, and that there are objects that
only show up in z [T-stars; z ~ 6 QSOs].
R
_______________________________________________
LSST-data mailing list
[email protected]
http://www.lsstmail.org/mailman/listinfo/lsst-data
