Abstract
Satellite orbit and clock products are the key elements for precise point positioning (PPP). Contrary to the relative technique, errors in satellite orbit and clock directly lump to the station coordinates for PPP technique. Currently final, rapid and ultra-rapid (observed-half and predicted-half) satellite products have been made freely available over the internet mainly for Global Positioning Service (GPS) satellites. Final and rapid products are used for post-processing applications. For real-time and near real-time applications, ultra-rapid products with predicted and observed parts can be used. There are several analysis centers that provide the satellite orbit and clock products. In this paper, accuracy and precision of GPS rapid and ultra-rapid satellite orbit and clock products from two services, International Global Navigation Satellite Service (IGS) and Jet Propulsion Laboratory (JPL), were investigated in the position domain of PPP technique while the final products of these services were taken as the true value. Ten IGS stations around the world were chosen for PPP processes. 24-12-8-4-2 hours of non-overlapping arc-lengths of GPS observations in 31 consecutive days (DOY 1-31 of 2018) were processed for each station. The results confirm that the shorter the arc-lengths, the larger the relative error of rapid and ultra-rapid products due to the similar Gaussian distribution pattern of the orbit errors with respect to the final products. In terms of consistency between the products, Root-Mean-Square-Errors (RMSE) of final-rapid differences of IGS and JPL are at the millimeter level. Millimeter level accuracy can be obtained using rapid and ultra-rapid products for JPL whereas only rapid products of IGS maintain millimeter accuracy with respect to the final products.