A quasi-global reduction method for astrolabe observations

Results of an evaluation of a quasi-global reduction method for astrolabe observations are presented. The method is based on two steps. The first one is to find a method to eliminate the temporal variations of some unknowns as the longitude and latitude. The second step is to find a method to suppress the connection problems between the different groups of the program. The fundamental equation is rewritten in order to introduce all the possible unknowns such as the precession, nutation, proper motion corrects, etc. An analysis of this new conditional equation shows that only five main unknowns are necessary to have a correct parameterization of the observations. It is shown that the coefficients of the unknowns may be written to contain only two variable quantities, the azimuth and the sidereal time of the observation of each star.