[LEAPSECS] next leap second

[Zefram]

John Sauter wrote:
>My second source is the estimates of delta T from Stephenson, Morrison
>and Hohenkerk.

I'd expect the IERS projections to be the more reliable over the next
few years.  The focus of Stephenson et al is the long term, and they
use general-purpose curve fitting methods.  The IERS, by contrast,
has specific models of the factors influencing Earth rotation, and its
projections are focussed on the short term.

I wonder what the intended range of usage is for the linear UT2-TAI model
described in Bulletin A.  It's explicitly said to be intended to go past
the tabulated values from the high-fidelity model, but there's nothing
saying how far after that it's valid for, or giving error estimates.
My sense is that, for the purpose for which I was using it, it was
reasonable to project two or three years into the future, but not further.
Beyond that the evolution of UT1 would be dominated by the decadal
fluctuation that is necessarily not represented in the linear model.

It would be nice to have more sophisticated projections from IERS more
than a year ahead.  It would particularly help in evaluating the proposals
that have been made involving scheduling leap seconds further ahead.

-zefram