Precession-nutation procedures consistent with IAU 2006 resolutions

¹ Space Science and Technology Department, CCLRC / Rutherford Appleton Laboratory, UK e-mail: P.T.Wallace@rl.ac.uk
² Observatoire de Paris, SYRTE/UMR8630-CNRS, 61 avenue de l'Observatoire, 75014 Paris, France e-mail: n.capitaine@obspm.fr

Received: 24 June 2006
Accepted: 28 August 2006

Abstract

Context.The 2006 IAU General Assembly has adopted the P03 model of Capitaine et al. (2003a) recommended by the WG on precession and the ecliptic (Hilton et al. 2006) to replace the IAU 2000 model, which comprised the Lieske et al. (1977) model with adjusted rates. Practical implementations of this new “IAU 2006” model are therefore required, involving choices of procedures and algorithms.

Aims.The purpose of this paper is to recommend IAU 2006 based precession-nutation computing procedures, suitable for different classes of application and achieving high standards of consistency.

Methods.We discuss IAU 2006 based procedures and algorithms for generating the rotation matrices that transform celestial to terrestrial coordinates, taking into account frame bias (B), P03 precession (P), P03-adjusted IAU 2000A nutation (N) and Earth rotation. The NPB portion can refer either to the equinox or to the celestial intermediate origin (CIO), requiring either the Greenwich sidereal time (GST) or the Earth rotation angle (ERA) as the measure of Earth rotation. Where GST is used, it is derived from ERA and the equation of the origins (EO) rather than through an explicit formula as in the past, and the EO itself is derived from the CIO locator.

Results.We provide precession-nutation procedures for two different classes of full-accuracy application, namely (i) the construction of algorithm collections such as the Standards Of Fundamental Astronomy (SOFA) library and (ii) IERS Conventions, and in addition some concise procedures for applications where the highest accuracy is not a requirement. The appendix contains a fully worked numerical example, to aid implementors and to illustrate the consistency of the two full-accuracy procedures which, for the test date, agree to better than as.

Conclusions.The paper recommends, for case (i), procedures based on angles to represent the PB and N components and, for case (ii), procedures based on series for the CIP . The two methods are of similar efficiency, and both support equinox based as well as CIO based applications.