The processing of the DORIS measurements relies on a precise model of the on board Ultra Stable Oscillator (USO) frequency. Unfortunately the important radiations in the South Atlantic Anomaly (SAA) perturb the USO behavior and produce localized in time (~20 minutes) peaks in the frequency when the satellite flies through it. These peaks are not modelled in standard DORIS processing, as the DORIS USO model is a third degree polynomial computed over 7 to 10 days. Therefore, DORIS measurements are not correctly processed when the satellite passes through SAA, and this can degrade significantly the orbits and station positioning (as on Jason 1).
On Sentinel3A, the GPS and DORIS receiver use the same USO. It is therefore possible to estimate independently the frequency of the USO using GNSS measurements. From the observation of this frequency anomaly, we constructed a model which can be used independently in the DORIS processing, by adjusting parameters related to the USO sensitivity to SAA.

This this model improves significantly Sentinel3A residuals.
The presentation will show how the model has been constructed, and the effects on DORIS phase residuals and station positioning.

Satellites are not impacted the same way by the SAA. It depends on the altitude and on the oscillator hardware. A model has been derived for Jason-1 in 2006 (“A corrective model for Jason-1 DORIS Doppler data in relation to the South Atlantic Anomaly”, JM Lemoine, H. Capdeville, 2006, DOI : 10.1007/s00190-006-0068-2). This study aims at trying to determine a model for Jason-3. The results and the main problems will be shown. One of the main issue is to properly place the SAA impacted area.