Benefits of a second tandem flight phase between two successive satellite altimetry missions for assessing instrumental stability

Abstract. Five successive reference missions, TOPEX/Poseidon, Jason-1, Jason-2, Jason-3, and more recently Sentinel-6 Michael Freilich, have ensured the continuity and stability of the satellite altimetry data record. Tandem flight phases have played a key role in verifying and ensuring the consistency of sea level measurements between successive altimetry reference missions and thus the stability of sea level measurements. During a tandem flight phase, two successive reference missions follow each other on an identical ground track at intervals of less than 1 min. Observing the same ocean zone simultaneously, the differences in sea level measurements between the two altimetry missions mainly reflect their relative errors. Relative errors are due to instrumental differences related to altimeter characteristics (e.g., altimeter noise) and processing of altimeter measurements (e.g., retracking algorithm), precise orbit determination, and mean sea surface. Accurate determination of systematic instrumental differences is achievable by averaging these relative errors over periods that exceed 100 d. This enables for the precise calibration of the two altimeters. The global mean sea level offset between successive altimetry missions can be accurately estimated with an uncertainty of about ±0.5 mm ([16 %–84 %] confidence level). Nevertheless, it is only feasible to detect instrumental drifts in the global mean sea level exceeding 1.0 to 1.5 mm yr−1 due to the brief duration of the tandem phase (9 to 12 months). This study aims to propose a novel cross-validation method with a better ability to assess the instrumental stability (i.e., instrumental drifts in the global mean sea level trends). It is based on the implementation of a second tandem flight phase between two successive satellites a few years after the first one. Calculating sea level differences during the second tandem phase provides an accurate evaluation of relative errors between the two successive altimetry missions. With a second tandem phase that is long enough, the systematic instrumental differences in sea level will be accurately reevaluated. The idea is to calculate the trend between the systematic instrumental differences made during the two tandem phases. The uncertainty in the trend is influenced by the length of each tandem phase and the time intervals between the two tandem phases. Our findings show that assessing the instrumental stability with two tandem phases can achieve an uncertainty below ±0.1 mm yr−1 ([16 %–84 %] confidence level) at the global scale for time intervals between the two tandem phases that are higher than 4 years or more and where each tandem phase lasts at least 4 months. On regional scales, the gain is greater, with an uncertainty of ±0.5 mm yr−1 ([16 %–84 %] confidence level) for spatial scales of about 1000 km or more. With regard to the scenario foreseen for the second phase between Jason-3 and Sentinel-6 Michael Freilich planned for early 2025, 2 years and 9 months after the end of the first tandem phase, the instrumental stability could be assessed with an uncertainty of ±0.14 mm yr−1 on the global scale and ±0.65 mm yr−1 for spatial scales of about 1000 km ([16 %–84 %] confidence level). In order to achieve a larger benefit from the use of this novel cross-validation method, this involves regularly implementing double tandem phases between two successive altimetry missions in the future.