Methods Altimetry 2

Quality assessment of altimeter data on the long-term is critical when one is looking for climate signals. Tide gauges measurements provide external and independent Sea Surface Height measurements which are used to check the quality and stability of altimeter sea levels. Comparisons to tide gauges are either performed at dedicated and carefully monitored calibration sites or through global analyses where a much larger set of in-situ stations is used. The main interest of both techniques is the detection of any altimeter drift or sudden bias.   

The latter class of methods is based on existing tide gauges networks (GLOSS/CLIVAR, PSMSL, REFMAR), quality controlled and corrected to provide a physical measurement comparable with altimetry measurements. Results are highly dependent on processing choices like station selection, altimetry collocation, geophysical corrections for both in-situ and altimetry data. In the present study, we look for the source of observed discrepancies between tide gauges and satellite altimetry: do they reflect errors in satellite altimetry or tide gauges networks. 

Results from comparisons between the main altimeter missions (TOPEX/Poseidon, Jason-1&2, ENVISAT, AltiKa) and in-situ stations are considered here. We try to separate true altimetry problems from artifacts linked to in-situ events. This represents a new step in establishing the accuracy of our methodology for performing global comparisons between altimetry and in-situ.

The Sentinel-3 satellite mission with its SRAL instrumentation contains new features compared to the conventional radar altimeter mission that form the basis for new innovative scientific analyses of both ocean and inland water levels. To utilize the full potential of the new data source, new methods and processing chains need to be developed. Subsequently, new potential Copernicus products should be developed that utilize the improved along-track resolution over both the oceans and over land. The main objective of the LOTUS project is to prepare the scientific and operational use of data from Sentinels 3. Then new operational processing, validation and delivery mechanisms need to be developed and implemented for generating the new dynamic products. Finally, the take-up of the new Copernicus products by the value-adding sectors needs to be stimulated and demonstrated to ensure that they will be used for commercial activities. LOTUS will develop processing scheme for extracting high-resolution sea surface heights, wave heights and wind speeds from SAR mode data. Over land, the LOTUS will develop processing scheme for extracting high-resolution river and lake heights, soil moisture, and snow water equivalents. This presentation shows results based on analyses using CRYOSAT data and available S-3 data. Furthermore, new DEMO data sets are presented and examples of scientific impact demonstrated.

Cryosat-2 is ESA’s ice mission which primary objective is to serve Cryosphere science. Nevertheless, Cryosat-2 has, in theory, the potential to be a mission of opportunity for oceanography. Indeed, the satellite embarks an innovative radar altimeter, and high-precision orbit determination (POD), which are expected to be at least as accurate as ENVISAT’s.

Even if Cryosat-2 is not an optimised payload for mesoscale observation (no radiometer and single frequency altimeter, no repetitive ground track), Labroue et al. (2011) and Scharoo et al. (2012) showed that the Cryosat-2 system gives very good performance in its traditional mode (Low Resolution Mode).  

The Cryosat-2 altimeter is operated almost continuously over ocean, either in Low Resolution Mode (like conventional pulse-limited altimetry sensors) or in the so-called  Delay Doppler/SAR mode (higher-resolution and lower noise level). Boy et al (2012, 2013) developed dedicated SAR retracking processing and showed that the results obtained over a full year of data were very promising. On the one hand, this technique appears to be mature enough to replace LRM mode in order to retrieve mesoscale signal. On the other hand, very compelling results have been obtained in the shortest scales below 80 km where the SAR provides more accurate data.

The geodetic orbit with a cycle of 369 days coupled with the noise reduction on SAR data makes Cryosat-2 data set very attractive for the improvement of the mean sea surface. The data quality of the mission has been assessed on the extended CPP SAR time series through different metrics that are presented here. We analyse more deeply the main features to check the reliability and the improvements of the SAR processing (noise reduction, increased along track spatial resolution, check of the dependencies that may induce geographically correlated errors...). A focus is done on the continuity between LRM and SAR processing to provide a seamless transition between both kinds of observations.

The Poseidon-3 altimeter onboard Jason-2 includes a significant new capability with respect to its prede- cessors, an open-loop (DIODE/DEM) tracker mode. This innovative mode is capable to successfully track and lock the backscatter signal over rapidly variant terrains, and thus overcome one of the limitations of the closed-loop tracker on Poseidon-2 on board Jason-1. DIODE/DEM achieves this improvement thanks to a pre-determined topography on board (a Digital Elevation Map or DEM) which combined with the Doris Im- mediate On-board orbit DEtermination (DIODE) allows for improving acquisition timing, and for reducing data loss at the coastal zone. In a further enhancement, Jason-3 and Sentinel-3A will be capable of switching to this innovative mode for selected regions and Sentinel-6 is also envisioned to embark a similar tracker. In order to recommend how these missions should operate DIODE/DEM, we have studied the impact of the tracker modes on the accuracy and precision of wave heights and wind speed, on the continuity of the sea level climate data record and the coverage of coastal regions. The results show close agreement between the two tracker modes over open ocean with the exception of high-tide areas, coastal zone and sea-ice regions. The DIODE/DEM tracker shows better performance than closed-loop at the coast and at the presence of sea ice. Jason-2 when operating in open loop allows for a ∼ 5% increase of successful acquisitions at the ocean to land transition, and it also better locks the backscatter signal position within the observation window when there is presence of sea ice. However, it exhibits less sensitivity to high tides than closed-loop. Although, topography differences observed at the presence of these extreme events are negligible and wave height differences shows to be within mission requirements (mm level). If data is available, the authors will repeat the Jason-2 investigations on Sentinel-3 and provide a first assessment of the tracker performance on this forthcoming mission.  

Sea Surface Height estimation derived from Nadir Altimetric missions results from the sum of a tenth of corrections, geophysical, instrumental, atmospheric…  They are affected by an error which can be characterized by different methods, mostly based on the differences between independent solutions of a same correction (envelopes of statistic monitoring, geographical analysis at crossovers, multimission comparisons…). Furthermore, each correction and associated errors also have their own temporal and geographical behavior and sign at different spatial scales.

To address different temporal and geographical scales skills on the past and current missions, Sentinel3, Cryosat-2, Envisat… the integrated error budget can declined as a function of the considered wavelength highlighting that the error are not homogeneous, depending on the temporal/spatial scales.

This study addresses the spectral content of the errors derived from nadir altimetry, to give a “wavelength dependant” coloration to the usual error budget of altimetric signals. It focuses on the hypothesis it implies and notably underlines the sensitivity of such analysis to the geographical zones and temporal period and missions