Gravity Field from Cryosat-2 SAR Altimetry: The Merits of the Empirical Leading Edge Retracker
Jain, Maulik; Andersen, Ole Baltazar; Stenseng, Lars; Dall, Joergen
National Space Institute / DTU Space, DENMARK

Precise gravity field determination can be done using Cryosat-2 SAR Level 1B data. Traditional gravity field determination using ship experiments is expensive and is difficult to be evaluated at places where ship navigation is difficult. Using the satellite based altimetry accurate sea surface heights can be computed. The gravity field can be determined by considering the height variations of the sea surface along track. Therefore accurate sea surface height determination through processing of satellite data is critical for reliable gravity field determination. This presentation deals with the improvement in traditional empirical waveform retrackers for sea surface height determination. Classical retrackers work well in open ocean, but in coastal areas and areas with presence of significant sea ice, the traditional retrackers either fail or provide inaccurate and consequently less accurate sea surface heights. The Arctic has this typical problem where the SAR echo waveforms are contaminated and hence a need of customization/improvement in empirical retrackers is necessary. With this improvement in mind, there is a focus on extracting the leading edge of the SAR waveform and isolating it from the rest of the waveform. Further the traditional OCOG and Threshold methods are applied on this extracted Leading Edge rather than the complete waveform. The extraction of the leading edge is critical as it is the part where interaction between the EM Altimeter Pulse and the surface occurs. This extraction is based on start/stop thresholds based on the standard deviation of power difference in consecutive/alternate bins. After the extracted leading edge is processed using the ' Leading Edge ' retrackers the sea surface heights are computed and converted to gravity fields. There are two methods of checking the quality of the improved retrackers. Firstly the sea surface heights are corrected, the mean sea surface is removed and all corrections are applied, the residual being the sea surface height anomaly. A lower value of this standard deviation indicates a better retracking algorithm. Secondly, the sea surface height anomaly's slope (along the track) and derived sea surface heights are evaluated as possible methods to compute the gravity field A lower value of the standard deviation with marine data indicates a more accurate gravity field and less noisy sea surface height/slope determination and hence a more efficient retracker for marine gravity field determination. The above two quality check techniques were employed for the traditional OCOG and Threshold retrackers, the Leading Edge retrackers and the ESA retracker used for Cyrosat-2 Level 2 heights. The results in both the quality checks confirmed that the Leading Edge retrackers performed better than the traditional/ESA retrackers. Hence a conclusion is reached that focusing on just the leading edge rather than the complete waveform yields better sea surface height/gravity field