Production of Earth orientation parameter series by Kalman filter-based combination

The Kalman Earth Orientation Filter (KEOF) developed at the Jet Propulsion Laboratory (JPL) is an operational software system that generates real-time and retrospective estimates as well as predictions of the Universal Time and Polar Motion (UTPM) parameters used in Earth orientation calibrations to support all JPL interplanetary flight projects. KEOF accomplishes this task by timely assimilation of specific UTPM measurements using Kalman filter and smoother algorithms. The behaviors of KEOF are governed by the stochastic models for dynamics of the three UTPM parameters as well as the observation models that relate the dynamical variables with the measurements and quantify the measurement uncertainties. Since KEOF began operations, the structures of these stochastic and observation models and their parameter values have undergone steady updates, due to advances in scientific understanding of the forces that affect Earth's rotation and orientation as well as improvements in accuracy of the instruments that measure these phenomena. We describe various versions of the stochastic and observation models including those used currently in KEOF operations.