Evaluating Tide Models for Operational Prediction of EOPs

It has been recognized since the early days of interplanetary spaceflight that accurate navigation requires taking into account changes in the Earth's rotation. In the 1960s, tracking anomalies during the Ranger VII and VIII lunar missions were traced to errors in the Earth orientation parameters (EOPs). As a result, Earth orientation calibration methods were improved to support the Mariner IV and V planetary missions. Today, accurate EOPs are used to track and navigate every interplanetary spaceflight mission. The approach taken at JPL to provide the interplanetary spacecraft tracking and navigation teams with the EOPs that they need is based upon the use of a Kalman filter to combine past EOP measurements and predict their future evolution. This Kalman filter includes a stochastic model of the EOPs that is determined empirically by fitting autoregressive models to time series obtained from historical EOP trajectories. To model the stochastic behavior of the EOPs, the effects of physical processes that influence the rotation rate in a deterministic and, hence, predictable manner are first excluded from the observed time series. Foremost among the predictable processes are the tides of solid Earth and ocean, whose effects can be evaluated using conventional tide models. Since the stochastic model included in the Kalman filter excludes tidal effects, these effects must be removed from the input EOP time series before the Kalman filter is used to operationally combine them and generate predictions. Models are again used to do this and this presentation will summarize our evaluation of recent solid Earth and ocean tide models for the purpose of combining and predicting EOPs.