This prospective study provided very promising results and suggests undertaking the following actions:
- Confirmation and refinement of the conclusions reached in this study with an extended evaluation of the AltiDop/SMRT simulator. For that purpose, it would be necessary to define even more realistic synthetic snowpacks, built from in-situ snow measurements (over land ice AND sea ice). The waveforms, simulated using these snowpacks, must then be compared to real altimetry acquisitions overflying in-situ sites measurement.
=> IGE is currently writing a scientific publication following this workplan, with the support of CLS
- Simulations performed with SMRT can be improved in Ka band (and lower frequencies) by taking into account 2nd order scattering. Progress has also to be made in the understanding of snow surface roughness and its impact into radar altimetry measurements.
- Sensitivity studies must be conducted at global scales to account for the snow parameters variation over the cryosphere. Over ice sheets, seasonal variations of the snowpack must also be studied.
- Most of the results have been obtained for ice sheet surfaces. Sea-ice surfaces must be investigated as well. Snow depth and freeboard estimations are among the main objectives of the CRISTAL mission. However, the lack of snow in-situ data over sea-ice prevents currently from precisely setting the synthetic snowpack.
- Specific level-2 algorithms employed for acquisitions made over snow surfaces still have to be improved, to reach user requirements. In particular the waveform retracking, and the algorithms exploiting Ku/Ka measurements (to estimate snow depth and/or volume scattering). This is true for the future CRISTAL mission but also for improving the current altimetry dataset. The simulation analysis made on this study (WP3) demonstrates that the surface elevation derived from the waveforms is sensitive to several snowpack parameters (namely: snow density, snow grain size & mean square slope), for both LRM and SAR modes and both the Ku and Ka bands. Therefore, alternatives to empirical retrackers have to be carefully and urgently considered. We strongly recommend to define, develop and validate retrackers based on physical modeling of the backscattered signal. The number of parameters impacting the shape of the waveforms is great but we are confident that CRISTAL configuration, based on simultaneous Ku/Ka and LRM/SAR/SARin modes, will provide enough measurements to discriminate/retrieve these parameters. This is the big challenge we have to face for the CRISTAL mission. Simulations must definitely help to do that job and the simulation tools (coupling radiometric and altimeter models) that have been developed must play an important role for better processing the future measurements over the cryosphere.
- An important point that has not been exploited in this study is that that radiometric model used is valid for simulating altimeter measurements but also for generating radiometer brightness temperatures and thus simulating radiometer measurements as well. We already know the great benefit of exploiting simultaneous observations from altimeter and radiometer looking at the same surface. Recent studies have shown the interest of such an approach for determining properties of the snow pack (age of the ice, type of swow etc, …). CRISTAL will embark a radiometer. Working on the coupling of altimeter and radiometer simulation could be largely beneficial for improving the future performance of the CRISTAL mission.