[LEAPSECS] JD & MJD, UT1 & UTC

[Tom Van Baak]

Now is a good a time as any to bring up an issue I've been meaning to ask.

My understanding is that JD has always been a day count, based on rotations of the earth. So the timescale from which JD is calculated is UT1 (not TAI or UTC). JD is widely used in astronomy. Presumably when JD is calculated from UTC one applies a DUT1 correction.

The textbook definition of MJD is simply JD - 2400000.5, which implies it is also based on UT1. But MJD is widely used in time metrology and in this context it is based on UTC, not UT1. Almost every report, graph, database and table you see from a timing laboratory uses MJD. There's a massive amount of code that takes UTC, does some arithmetic, divides by 86400.0 and out comes MJD.

A side-effect of this is that there is no valid MJD representation for the leap second we had a few days ago. I think most people sweep this under the rug and move on. For example, when making plots with MJD the error or ambiguity caused by a leap second is far less than one pixel and nobody notices.

But since some of you are real picky about formal definitions, my question is what to do about MJD -- is it UT1-based or it is UTC-based? And if UTC-based, what's the right thing to do when the day has a +/- leap second?

And if leap seconds were to vanish, would MJD become UTC-based instead of UT1-based? Would metrology and astronomy fight over who gets to define the timescale of MJD?

/tvb