TLE time reference

Frank Reed (f-reed@metabien.com)
Thu, 16 Jul 1998 13:54:04 -0700

Dr. T. S. Kelso's FAQ page at: 


http://celestrak.com/columns/v04n03/index.html#FAQ03

indicates that the time frame for satellite predictions fropm the the
commonly used TLE files is UT (Universal Time), but doesn't specify which
flavor of UT (UTC or UT1 or ??) is to be used.  

As I understand them, UT1 is the actual mean solar time as determined from
observations, while UTC is the time kept by and coordinated among stable
atomic clocks at several locations worldwide.  UTC is the time broadcast by
WWV and other time services.  UT1 is subject to variations in the rotation
of the earth that are not entirely predictable, therefore it can differ by
substantial fractions of a second from UTC.  By international agreement,
UTC is adjusted to approximate UT1 by insertion/deletion of "leap seconds"
as needed.  Therefore UTC is discontinuous, while UT1 is not.  Further
information can be found at:


http://sadira.gb.nrao.edu/~rfisher/Ephemerides/times.html

According to the IERS Bulletin-A for 16 July 1998 found at:


file://maia.usno.navy.mil/ser7/ser7.dat

the difference between UT1 and UTC today (presumably at midnight UT) is:

UT1-UTC = -0.10618 second on 7/16/1998

and the prediction for a year from now is:

UT1-UTC = -0.67250 second on 7/15/1999.

Obviously these time differences are of little importance for visually
sighting satellites.  However, for some types of analysis of obervation
data they could be significant.

Does anyone know whether the people who maintain the TLEs at OIG use UT1 or
UTC (or something else)?

- Frank

-- Frank Reed
Scottsdale, Arizona, USA  111.898W  33.484 N  1227 feet