Comments on the ERA-2005 numerical theory of Earth rotation

Two papers recently published in Celestial Mechanics (Krasinsky 2006, and Krasinsky and Vasilyev 2006) have presented a model for Earth-rotation variations, called ERA-2005, based on numerical integration of a new set of equations for the rotation of a deformable Earth followed by a fit of the results of the integration to VLBI data. These papers claimed that this model was superior to any other existing model. The purpose of this Note is to bring to light fundamental errors in the derivation of the basic equations of the new theory, compounded by serious deficiencies in the process of fitting to the data; they make ERA-2005 unsuitable for consideration as a geophysics-based model of nutation and precession.