Evaluation of the ICRF stability from a position time series analysis

Context. The celestial reference frame is established on the basis of the absolute positions of extragalactic sources that are assumed to be fixed in space. The fixing of the axes is one of the crucial points for the concept behind the International Celestial Reference System (ICRS). However, due to various effects such as its intrinsic activity, the apparent position of the extragalactic sources may vary with time, resulting in a time-dependent deviation of the frame axes that are defined by the positions of these sources. Aims. We aim to evaluate the axis stability of the third realization of the International Celestial Reference Frame (ICRF3). Methods. We first derived the extragalactic source position time series from observations of very long baseline interferometry (VLBI) at the dual S∕X-band (2.3/8.4 GHz) between August 1979 and December 2020. We measured the stability of the ICRF3 axes in terms of the drift and scatter around the mean: (i) we estimated the global spin of the ICRF3 axes based on the apparent proper motion (slope of the position time series) of the defining sources of the ICRF3; (ii) we also constructed the yearly representations of the ICRF3 through annually averaged positions of the defining sources of the ICRF3 and estimated the dispersion in the axes orientation of these yearly frames. Results. The global spin is no higher than 0.8 μas yr-1 for each ICRF3 axis with an uncertainty of 0.3 μas yr-1, corresponding to an accumulated deformation smaller than 30 μas for the celestial frame axes during 1979.6–2021.0. The axes orientation of the yearly celestial frame becomes more stable as time elapses, with a standard deviation of 10 μas–20 μas for each axis. Conclusions. The axes of the ICRF3 are stable at approximately 10 μas–20 μas from 1979.6–2021.0 and the axes stability does not degrade after the adoption of the ICRF3