Reference time scales International Atomic Time (TAI) and
Universal Coordinated Time (UTC) have been computed regularly and have been published
in the monthly BIPM Circular T. Definitive results for 1999, 2000, 2001 and
2002 have been available, in the form of computer-readable files from the BIPM
website (http://www.bipm.org), and on
printed volumes of the respective Annual Reports of the BIPM Time Section. Work has been done to automate the
calculation of the time scales TAI and UTC published monthly in the BIPM
Circular T. Several modifications have been introduced in Circular T, starting
in May 2003: results are given to 0.1 ns; the list of time links used in the
calculation of the current circular is provided; and a new format has been
adopted for the layout of the circular.
Research concerning time scale algorithms includes studies
to improve the long-term stability of the free atomic time scale EAL and the
accuracy of TAI. The weighting procedure of clocks participating in TAI has
been revised and modified. Until 31 December 2000, the maximum relative weight
of clocks participating into TAI was
fixed to 0.700%. With the improvement of the commercial atomic clocks, this
method became inefficient to discriminate between the best clocks. Since
January 2001 the maximum relative clock weight has been established each month,
depending on the number of participating clocks. This modification improves the
stability of the international time scales.The medium-term stability of EAL,
expressed in terms of the Allan deviation, is estimated to be 0.6 ´ 10-15 for averaging times of 20 to
40 days over the period.
Primary frequency standards developed and operated by the
National Institute of Standards and Technology (NIST, USA), the Communications
Research Laboratory (CRL, Japan), the Paris Observatory (OP, France) and the
Physikalish-Technische Bundesanstalt (PTB, Germany) reported their measures to
the BIPM. The global treatment of these individual measurements led to a
relative departure of the duration of the TAI scale unit from the SI second on
the geoid ranging, in 2002, from +6 ´ 10-15 to +11´ 10-15, with an uncertainty smaller
than 3 ´ 10-15.
The frequency offset between TAI and EAL is changed when necessary; this
operation is referred to as the “steering of TAI”. Following the
recommendations of the Consultative Committee on Time and Frequency, changes
were implemented to render the data used in TAI, as well as the results, more
accessible to the users and to make the procedures of calculation even more
transparent and traceable. Since April 2000 two modifications had been implemented:
a new model to characterise the instability of the free atomic scale EAL, and a
more complete representation of the uncertainty of the deviation of the TAI
scale interval relative to that of the Terrestrial Time TT.
Two techniques of time transfer are used at present to
compare clocks in TAI: GPS common-views based on C/A measurements and two-way
satellite time and frequency transfer (TWSTFT). In the last decade the time
links computed at the BIPM used the classical GPS common-view technique based
on C/A-code measurements obtained from single-channel receivers. The commercial
availability of newly developed receivers has stimulated interest in extending
the classical common-view technique for use of multi-channel dual-code
dual-system (GPS and GLONASS) observations, with the aim of improving the
accuracy of time transfer. Since July
1999 GPS multi-channel links and TWSFTF links have been progressively
introduced in TAI. Ionospheric maps and precise operational satellite
ephemerides provided by the International GPS Service (IGS) are routinely used
to correct all links in regular TAI calculations since May 2000. In addition,
the BIPM Time section carries on research on new techniques of time transfer,
such as the utilisation of geodetic type receivers. These activities had been
developed in the period 1999-2002 in the frame of the IGS/BIPM pilot project to
study accurate time and frequency comparison using GPS phase and code
measurements, and have been incorporated to the IGS as a current activity since
2003. A pilot experiment (TAIP3) was proposed in April 2002 to laboratories
participating to TAI. The goal was to study time links computed with GPS P3
data obtained from geodetic-type dual-frequency receivers. Comparisons of such
P3 time links with other techniques presently used for TAI have been conducted
in the aim of evaluating the long term stability of P3 time links and, as a
matter of fact, to assess the long-term stability of the other techniques. It has been concluded that the reliability
and long-term stability of P3 links are adequate for their use in TAI
computation.
The BIPM/IAU Joint Committee on general relativity for
space-time reference systems and metrology (JCR), concluded his work in January
2001. Its activities have been undertaken by the IAU working group on
Relativity, Celestial Mechanics, Astrometry and Metrology (RCMAM). Studies have
been conducted at the BIPM in collaboration with other members of the
JCR/RCMAM, they concern the extension of the relativistic framework to allow a
correct treatment for time transformations and the realisation of barycentric
coordinate time at the full post Newtonian level and the realisation of
geocentric coordinate times. Since January 2001 the BIPM, jointly with the
USNO, is the Conventions Product Centre of the International Earth Rotation
Service (IERS).