Oceanography (Altimetry) 2

The ESA CryoSat mission is the first space mission to carry a radar altimeter that can operate in Synthetic Aperture Radar “SAR” (or delay-Doppler) and interferometric SAR (SARin) modes. Studies on CryoSat data have analysed and confirmed the improved ocean measuring capability offered by SAR mode altimetry, through increased resolution and precision in sea surface height and wave height measurements, and have also added significantly to our understanding of the issues around the processing and interpretation of SAR altimeter echoes.

We present work in four themes, building on work initiated in the CryoSat Plus for Oceans project (CP4O), each investigating different aspects of the opportunities offered by this new technology.

The first two studies address the coastal zone, a critical region for providing a link between open-ocean and shelf sea measurements with those from coastal in-situ measurements, in particular tide gauges. Although much has been achieved in recent years through the Coastal Altimetry community, (http://www.coastalt.eu/community) there is a limit to the capabilities of pulse-limited altimetry which often leaves an un-measured “white strip” right at the coastline.  Firstly, a thorough analysis was made of the performance of “SAR” altimeter data (delay-Doppler processed) in the coastal zone. This quantified the performance, confirming the significant improvement over “conventional” pulse-limited altimetry. In the second study a processing scheme was developed with CryoSat SARin mode data to enable the retrieval of valid oceanographic measurements in coastal areas with complex topography. Thanks to further development of the algorithms, a new approach was achieved that can also be applied to SAR and conventional altimetry data (e.g., Sentinel-3, Jason series, EnviSat).

The third part of the project developed and evaluated improvements to the SAMOSA altimeter re-tracker that is implemented in the Sentinel-3 processing chain.  The modifications to the processing scheme should support improved performance in terms of accuracy and efficiency in retrieving oceanographic geophysical parameters from altimeter data.

Finally, we describe the development of a state of the art tidal atlas for the Arctic Ocean with CryoSat altimeter data. Through its high inclination orbit, the CryoSat mission provides the most complete altimeter data set ever used in this region, and so should enable the production of a highly accurate Arctic tidal model. This in turn will improve the quality of CryoSat Sea Surface Height measurements and all derived products (e.g. mean sea surface, mean dynamic topography).

Together these studies provide an important foundation for exploiting data from the Sentinel-3 and Sentinel-6/Jason-CS missions.

The work described in this presentation was supported by an extension to the CryoSat Plus for Oceans project funded by ESA (STSE). We also acknowledge the support of CNES who provided the CNES-CCP CryoSat Products used in these studies. CNES-CPP products were developed by CNES and CLS in the frame of the  “Sentinel-3 SRAL SAR mode performance assessment” study.

State-of-the-art Arctic Ocean mean sea surface (MSS) and geoid models are used to support sea ice freeboard estimation from satellite altimeters, and for oceanographic studies. However, errors in a given model in the high frequency domain, e.g. due to unresolved gravity features, can result in errors in the estimated freeboard heights, especially in areas with a sparse lead distribution in consolidated ice conditions. Additionally these errors can impact ocean geostrophic current estimates and  remaining biases in the models may impact longer-term, multi-sensor oceanographic time-series of sea level change.

This study, part of the ESA CryoVal Sea Ice project, focuses on an inter-comparison of various state-of-the-art Arctic MSS models (UCL13/DTU13/ICEn) and commonly-used geoid models (EGM08). We show improved definition of gravity features, such as the Gakkel ridge, in the latest MSS models.  We quantify remaining errors due to unresolved gravity features and inter-satellite biases within commonly-used models and we show the implications of these potential error sources on freeboard derivation. To identify and quantify the spatial effect of the unresolved features, primarily in the high-frequency domain, we combine a climatology of lead distributions with the gradient of the slope of the sea surface anomalies. The differences between the models are analyzed and used to support improvements in future models.

The Arctic Ocean sea level remains largely unobserved by satellite altimetry missions, either due to orbit constraints (e. g. for Jason missions) or because ice coverage hinders the ability of conventional radar altimeters to retrieve sea surface height. Over the last few years, several efforts have been made to improve the observability of the Arctic Ocean and generate tailored sea level products (Prandi et al, 2012 & Andersen et al., 2015). These products remain based on the processing of 1Hz measurements, with dedicated editing and choice of geophysical corrections.

In this study, we take advantage of the waveform classification developed for the Envisat (CCI project) and SARAL/AltiKa missions (PEACHI project) that allows discriminating echo returns from leads in the ice pack, ice floes and open ocean. All echoes are retracked using the same adaptive algorithm. After editing and correction the measurements are used to build cycle-wise grids of sea level anomaly in the Arctic Ocean with unprecedented data availability in ice covered areas.

We present the methodology used to build this new dataset, its validation and the new insights on Arctic Ocean sea level variability that can be derived from it.

The Arctic Ocean is a challenging region for tidal modeling, because of its complex and not well-documented bathymetry, together combined with the intermittent presence of sea ice and the fact that the in situ tidal observations are rather scarce at such high latitudes. As a consequence, the accuracy of the global tidal models decreases by several centimeters in the Polar Regions. In particular, it has a large impact on the quality of the satellite altimeter sea surface heights in these regions (ERS1/2, Envisat, CryoSat-2, SARAL/AltiKa and the future Sentinel-3 mission).

Better knowledge of the tides improves the quality of the high latitudes altimeter sea surface heights and of all derived products, such as the altimetry-derived geostrophic currents, the mean sea surface and the mean dynamic topography. In addition, accurate tidal models are highly strategic information for ever-growing maritime and industrial activities in this region.

NOVELTIS and DTU Space have developed a regional, high-resolution tidal atlas in the Arctic Ocean, in the framework of the CryoSat Plus for Ocean (CP4O) ESA project. In particular, this atlas benefits from the assimilation of the most complete satellite altimetry dataset ever used in this region, including Envisat data up to 82°N and the CryoSat-2 reprocessed data between 82°N and 88°N. The combination of all these satellites gives the best possible coverage of altimetry-derived tidal constituents. The available tide gauge data were also used for assimilation and validation.

This paper presents the performances of this new regional tidal model in the Arctic Ocean, compared to the existing global tidal models.

We present results of investigation of wind-wave generation by tropical cyclones using satellite altimeter data. Tropical cyclones are generally relatively small rapidly moving low pressure systems that are capable of generating severe wave conditions. Translation of a tropical cyclone leads to a prolonged period of time surface waves in the right sector remain under high wind forcing conditions. This effect has been termed extended fetch, trapped fetch or group velocity quasi-resonance. A tropical cyclone wave field is thus likely more asymmetrical than the corresponding wind field: wind waves in the tropical cyclone right sector are more developed with larger heights than waves in the left one.
A dataset of satellite altimeter intersections of the Western Pacific tropical cyclones was created for 2010-2013. Data from four missions were considered, i.e., Jason-1, Jason-2, CryoSat-2, SARAL/AltiKa. Measurements in the rear-left and front-right sectors of tropical cyclones were examined for the presence of significant wave asymmetry. An analytical model is then derived to efficiently describe the wave energy distribution in a moving tropical cyclone. The model essentially builds on a generalization of the self-similar wave growth model and the assumption of a strongly dominant single spectral mode in a given quadrant of the storm. The model provides a criterion to anticipate wave enhancement with the generation of trapped abnormal waves. If forced during a sufficient timescale interval, also defined from this generalized self-similar wave growth model, waves can be trapped and large amplification of the wave energy will occur in the front-right storm quadrant. Remarkably, the group velocity and corresponding wavelength of outrunning wave systems will become wind speed independent and solely relate to the translating velocity. The resulting significant wave height also only weakly depends on wind speed, and more strongly on the translation velocity.
Satellite altimeter measurements, together with TC intensities estimates, are used to assess the proposed formulations. Compared to satellite altimeter measurements, the proposed analytical solutions for the wave energy distribution are in convincing agreement. For almost symmetrical wind field, the model quantitatively reproduces measured profiles of the wave energy with significant asymmetry between the wave-containment front-right quadrant and the rear-left quadrant where wave energy is remarkably damped. Though the differences between parametric model-wind and altimeter-wind profiles are noticeable, the energy ratios between the front-right and the rear-left quadrants are similar for both wind sources. As analytically developed, the wave enhancement criterion can provide a rapid evaluation to document the general characteristics of each storm, especially the expected wave field asymmetry.