Effect of ambiguity resolution on the draconitic errors in sub-daily GPS position estimates

Spurious draconitic errors with periods of (351.4/N) days are detected in GPS position time series. This study assesses the performance of ambiguity resolution in sub-daily position estimates on reducing the draconitic errors. Precise point positioning solutions covering 132 global stations from 2002 to 2017 are generated with and without ambiguity resolution using 24-, 8- and 4-h processing sessions. Different satellite products from the European Space Agency (ESA) and the Massachusetts Institute of Technology (MIT) are employed. Although the ambiguity fixing rate decreases with the processing session shortening, a higher percentage of the draconitic errors are reduced by ambiguity resolution for nearly all the solutions and components (but the East component of ESA 4-h solution). At its best, fixing ambiguities reduces the draconitic errors by more than 60% in the East component and by about 20% in the North and Up components for 4-h solutions. Moreover, the draconitic errors and their reduction by ambiguity resolution show a latitudinal dependence. The draconitic errors are also reduced by sub-daily solutions with the largest reduction of 25.4%, 12.6% and 25.3% in the North, East and Up components, respectively, for ESA 4-h solution with ambiguity resolution, but less reduction for MIT solutions. It indicates that part of the draconitic errors result from the alias of certain unmodeled diurnal and semidiurnal signals beating against standard 24-h processing, and these signals mainly come from orbit-related errors, rather than station-specific errors.