Simultaneous ocean surface current and wind vectors retrieval with squinted SAR interferometry: Geophysical inversion and performance assessment
Simultaneous ocean surface current and wind vectors retrieval with squinted SAR interferometry: Geophysical inversion and performance assessment
Simultaneous measurements of ocean surface current and wind vectors at the ocean submesoscale (O [1–10 km]) are needed to improve our understanding of upper ocean mixing, air-sea interactions, ocean biophysical processes and large-scale oceanic transports. A new satellite mission concept called SEASTAR aims to do just that. The concept is a Ku-band along-track interferometric Synthetic Aperture Radar (SAR) system with two squinted beams pointing ±45° from broadside and incidence angles around 30°. The paper presents an inversion strategy to retrieve simultaneously ocean surface current and wind vectors and reports on the performance obtained with different wind/current conditions and instrument configurations. Results are based on numerical simulations using a Bayesian approach and existing geophysical model functions (GMFs) of the microwave Normalized Radar Cross Section (NRCS) and Doppler shift.
Using the baseline two-look instrument configuration and realistic instrument noise figures (radiometric resolution: kp = 5 and 12%; Δdf = 2 and 5 Hz), the root-mean square errors (RMSE) of the retrieved current and wind vectors are typically better than [0.1 m/s, 10°] for current and [0.5 m/s, 5°] for wind. This inversion setup yields four ambiguous solutions within a current range of ∼1 m/s. The addition of dual polarization (VV, HH) capability helps to discriminate these ambiguities. The retrieval performance depends weakly on geophysical parameters such as wind speed, current velocity or current direction, but is sensitive to wind direction because of its strong effect on current retrieval through the wind-wave induced artifact surface velocity (WASV). Larger retrieval errors are obtained when the wind is aligned with one of the antenna line-of-sight (LoS) directions, although errors remain typically below [0.2 m/s, 25°] for current and [0.5 m/s, 15°] for wind. Improving the retrieval performance regardless of wind direction could be achieved either with lower noise figures on σ0, or with higher incidence angles, or by including an additional third-look direction in azimuth (e.g. to achieve a configuration similar to Metop/ASCAT scatterometers) as per the SEASTAR mission concept submitted to EE10.
Martin, Adrien C.H.
8c4f9cad-a856-4638-a13d-1b27edb73c3c
Gommenginger, Christine P.
f0db32be-34bb-44da-944b-c6b206ca4143
Quilfen, Yves
185a8597-f2ff-4a29-b0b0-2843ae1ba825
Martin, Adrien C.H.
8c4f9cad-a856-4638-a13d-1b27edb73c3c
Gommenginger, Christine P.
f0db32be-34bb-44da-944b-c6b206ca4143
Quilfen, Yves
185a8597-f2ff-4a29-b0b0-2843ae1ba825
Martin, Adrien C.H., Gommenginger, Christine P. and Quilfen, Yves
(2018)
Simultaneous ocean surface current and wind vectors retrieval with squinted SAR interferometry: Geophysical inversion and performance assessment.
Remote Sensing of Environment.
(doi:10.1016/j.rse.2018.06.013).
Abstract
Simultaneous measurements of ocean surface current and wind vectors at the ocean submesoscale (O [1–10 km]) are needed to improve our understanding of upper ocean mixing, air-sea interactions, ocean biophysical processes and large-scale oceanic transports. A new satellite mission concept called SEASTAR aims to do just that. The concept is a Ku-band along-track interferometric Synthetic Aperture Radar (SAR) system with two squinted beams pointing ±45° from broadside and incidence angles around 30°. The paper presents an inversion strategy to retrieve simultaneously ocean surface current and wind vectors and reports on the performance obtained with different wind/current conditions and instrument configurations. Results are based on numerical simulations using a Bayesian approach and existing geophysical model functions (GMFs) of the microwave Normalized Radar Cross Section (NRCS) and Doppler shift.
Using the baseline two-look instrument configuration and realistic instrument noise figures (radiometric resolution: kp = 5 and 12%; Δdf = 2 and 5 Hz), the root-mean square errors (RMSE) of the retrieved current and wind vectors are typically better than [0.1 m/s, 10°] for current and [0.5 m/s, 5°] for wind. This inversion setup yields four ambiguous solutions within a current range of ∼1 m/s. The addition of dual polarization (VV, HH) capability helps to discriminate these ambiguities. The retrieval performance depends weakly on geophysical parameters such as wind speed, current velocity or current direction, but is sensitive to wind direction because of its strong effect on current retrieval through the wind-wave induced artifact surface velocity (WASV). Larger retrieval errors are obtained when the wind is aligned with one of the antenna line-of-sight (LoS) directions, although errors remain typically below [0.2 m/s, 25°] for current and [0.5 m/s, 15°] for wind. Improving the retrieval performance regardless of wind direction could be achieved either with lower noise figures on σ0, or with higher incidence angles, or by including an additional third-look direction in azimuth (e.g. to achieve a configuration similar to Metop/ASCAT scatterometers) as per the SEASTAR mission concept submitted to EE10.
Text
admartin_performance_RSE_v3preprint
- Accepted Manuscript
More information
Accepted/In Press date: 8 June 2018
e-pub ahead of print date: 23 June 2018
Identifiers
Local EPrints ID: 422392
URI: http://eprints.soton.ac.uk/id/eprint/422392
ISSN: 0034-4257
PURE UUID: fc01a940-6a21-44f3-94d5-11710b6c106a
Catalogue record
Date deposited: 23 Jul 2018 16:30
Last modified: 16 Mar 2024 06:52
Export record
Altmetrics
Contributors
Author:
Adrien C.H. Martin
Author:
Christine P. Gommenginger
Author:
Yves Quilfen
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics