[vsnet-chat 1079] Re: helio(bary)centeric correction (Thorstensen)

[Taichi Kato <tkato>]

Dear Dr. Thorstensen,

   The two points are very important.  I have traditionally computed
heliocentric corrections in publishing results, mainly for consistency with
much of other publications.  Still only few optical, ground-based reports use
BJD(TT).  Probably it may be time to turn fully to BJD.

   One of my first concerns is the correctness of many versions (including
mine) of heliocentric correction programs.  I have observed in several
independent programs differences up to (sometimes even greater than) 1 min.
My approach of the program uses analytic approximations, using a free
code for the Newcomb expansion (a bit old standard, I must confess).  The
provider of the code assures an accuracy of 10^-6 AU, but independent
confirmation is always important.  A few years ago I had a chance to compare
the result with another using the JPL ephemeris (probably DE200); the report
was satisfactory (in terms of HJD, not yet for BJD).

   The correction term to HJD to BJD being small (an order of a few seconds),
may I first suggest to compare heliocentric corrections between us?
I presume and hope the observed discrepancy between us results only from
HJD-BJD.

   The UTC-TAI problem is subtle, but important.  Following the most
observer's convention, I exclusively use UTC when observing.  Fortunately
at my longitude, (night-time) observations will not jump across leap seconds;
we have only to adjust the clock before each observation.  (I wonder how
about at other longitudes).  The correction to TAI is an even simpler
problem using only a small table, but surely needs attention when fine
analyzing period changes esp. across leap seconds.  However, providing
predicted eclipse times in TAI will be only misleading.

Regards,
Taichi Kato

VSNET Home Page

Return to Daisaku Nogami

---

vsnet-adm@kusastro.kyoto-u.ac.jp