Multi-satellite synergies for polar ocean altimetry
The research presented in this thesis uses multi-satellite synergies to make observations of sea ice, snow on sea ice, and sea surface height. While results are focussed on the Arctic, the techniques developed could equally be applied to the Southern Ocean. A novel approach for estimating snow depth...
| Main Author: | |
|---|---|
| Format: | Doctoral or Postdoctoral Thesis |
| Language: | English |
| Published: |
UCL (University College London)
2019
|
| Subjects: | |
| Online Access: | https://discovery.ucl.ac.uk/id/eprint/10087932/1/THESIS_FINAL_ONLINE_UPLOAD.pdf https://discovery.ucl.ac.uk/id/eprint/10087932/ |
| Summary: | The research presented in this thesis uses multi-satellite synergies to make observations of sea ice, snow on sea ice, and sea surface height. While results are focussed on the Arctic, the techniques developed could equally be applied to the Southern Ocean. A novel approach for estimating snow depth on sea ice from dual-frequency altimetry is applied to two pairs of satellites; AltiKa (Ka-band radar) and CryoSat-2 (Ku-band radar), and Envisat (Ku-band radar) and ICESat (laser). The AltiKa−CryoSat-2 snow product is evaluated against Operation IceBridge snow depths from the 2013, 2014, 2015 and 2016 airborne campaigns, returning root-mean-square deviations of 7.7, 5.3, 5.9 and 6.7cm respectively. Waveform data from the CryoSat-2 and Sentinel-3A missions are processed to estimate radar freeboard and sea-level anomaly for the 2017-18 winter. A 2 cm difference is found between sea-level anomaly from each satellite, which is attributed to an intermission range bias. The difference in Sentinel-3A−CryoSat-2 basin-average radar freeboard is found to be +1 cm for all six months evaluated. Since radar freeboard is a relative measurement, this discrepancy must be independent of range, and is attributed to the need for an alternative retracker bias for Sentinel-3A owing to the impact of its different operation characteristics. Following Sentinel-3B’s launch in 2018, data from Sentinel-3A, -3B and CryoSat-2 are processed for the 2018-19 winter. In line with previous findings, 1cm is removed from all Sentinel-3A and -3B floe elevation measurements, and average radar freeboards are shown to agree with CryoSat-2 to within 3mm. A merged product is developed combining data from the CryoSat-2 and Sentinel 3A/B missions, permitting basin-wide observations of Arctic sea-level anomaly and radar freeboard at synoptic time-scales. A comparison of 9-day radar freeboard variability with snowfall data from ERA5 reanalysis reveals a strong positive correlation over first-year ice, a result which appears to contradict traditional assumptions ... |
|---|