Data Assimilation, Analysis and Fusion across Modalities and Varying Latencies: Earth Orientation Parameters’ Estimation and Prediction

Algorithms and software have been developed to improve the assimilation of Earth Orientation Parameter data for purposes to developing Earth Orientation estimates that need to be disseminated to other military and government entities and to the general public. The work concentrated on modernizing a cubic-spline data fitting technique to express it in terms of normal equations, to enable the simultaneous fitting of function and rate data, and to allow data cross-correlations in the statistical model of noise. The resulting algorithms and software were tested on example Earth Orientation Parameter data, and the results were compared with current best estimates of the parameters based on long smoothing runs. Additional work considered novel means of using GPS data, including GPS-derived USNO Earth Orientation rate of change data, and Atmospheric Angular Momentum data.