Day 4

Detailed paper information

Back to list

Paper title A calibrated coastal SAR winds dataset in the Australian region
  1. Salman Saeed Khan Commonwealth Scientific and Industrial Research Organisation (CSIRO)
  2. Ian Young University of Melbourne Speaker
  3. Agustinus Ribal University of Melbourne
  4. Mark Hemer Commonwealth Scientific and Industrial Research Organisation (CSIRO)
Form of presentation Poster
  • A8. Ocean
    • A8.13 Remote-sensing of Ocean Winds and Stress
Abstract text Australia has a vast marine estate and amongst the longest coastlines in the World. Offshore ocean wind measurements are necessary for monitoring for a variety of users such as offshore industries (oil and gas, fisheries etc.) and understanding wind climatology for offshore operations, ship navigation, and coastal management. Australia also has a developing offshore wind energy industry. However, there are few sustained in-situ coastal ocean surface wind measurements around Australia, and either remain largely limited to reefs, jetties and coastal infrastructure or are acquired commercially by offshore industry operators. One exception is the ocean wind record from Southern Ocean Frequency Series (SOFS) flux station (Schulz et al., 2012) several hundred kms offshore south-west of Tasmania.

Sentinel-1 A and B Synthetic Aperture Radar (SAR) satellites regularly map the wider Australian coastal region and provide an opportunity to exploit these data to compile an up-to-date database of coastal wind measurements. Such a high-resolution coastal winds database from SAR also compliments global Scatterometer wind measurements as Scatterometers provide limited data closer to the shore. Two such valuable SAR winds databases already exist in other geographical regions, including NOAA’s operational SAR derived wind products (Monaldo et al., 2016) primarily focused on North America and DTU (Technical University of Denmark) Wind Energy’s SAR winds database (Hasager et al., 2006) with a European focus. With this goal in sight, a regional calibrated coastal SAR winds database has been developed for the Australian region from Sentinel-1 missions.

SAR winds are derived using input data from Sentinel-1 level-2 ocean winds (owi) product (CLS, 2020) sourced from the Copernicus Australasia regional data hub. The owi product contains all the input variables necessary to derive SAR winds including normalised radar cross section (NRCS), local incidence angle, satellite heading, and collocated model wind speed and direction from ECMWF. The algorithm applied for wind inversion is based on a variational Bayesian inversion approach as proposed in Portabella et al. 2002, and the Sentinel-1 Ocean wind algorithm definition document (CLS, 2019) with CMOD5.N as the underlying geophysical model function – GMF (Hersbach et al. 2010). For consistency, the whole Sentinel-1 archive is processed using the same wind inversion scheme and GMF. The resulting spatial resolution of the derived winds is roughly 1 km, like the owi product. The winds are also quality flagged in a systematic manner by using the ratio of the measured to simulated NRCS as a proxy for quality of the winds retrieved and applying various thresholds of median absolute deviation to this ratio. As in-situ measurements are not available in the region, calibration of SAR wind speed is performed against the calibrated (against NDBC buoy wind speeds) Metop-A and B Scatterometer winds database (Ribal et al., 2020) matchups. Calibrated SAR wind speeds are then validated against an independent Altimeter wind speed database (Ribal et al., 2019).

Such a high-resolution coastal winds archive has numerous uses for various applications. The intention is to explore these data in the future for suitability in offshore wind resource assessment, better understanding of coastal wind climatology alongside other regional model hindcast and reanalyses data, and verification of model wind fields, whose quality is a major source of error in wave models.


Hasager, C.B., Barthelmie, R.J., Christiansen, M.B., Nielsen, M. and Pryor, S.C. (2006), Quantifying offshore wind resources from satellite wind maps: study area the North Sea. Wind Energ., 9: 63-74.

Hersbach, H. (2010). Comparison of C-Band Scatterometer CMOD5.N Equivalent Neutral Winds with ECMWF, Journal of Atmospheric and Oceanic Technology, 27(4), 721-736.

Monaldo, F. M., Jackson, C. R., Li, X.; Pichel, W. G. Sapper, J., Hatteberg, R. (2016). NOAA high resolution sea surface winds data from Synthetic Aperture Radar (SAR) on the Sentinel-1 satellites. NOAA National Centers for Environmental Information. Dataset.

Portabella, M., Stoffelen, A., and Johannessen, J. A., (2002). Toward an optimal inversion method for synthetic aperture radar wind retrieval, J. Geophys. Res., 107(C8), doi:10.1029/2001JC000925.

Ribal, A., Young, I.R. (2019). 33 years of globally calibrated wave height and wind speed data based on altimeter observations. Sci Data 6, 77.

Ribal, A., & Young, I. R. (2020). Calibration and Cross Validation of Global Ocean Wind Speed Based on Scatterometer Observations, Journal of Atmospheric and Oceanic Technology, 37(2), 279-297.

Schulz, E. W., Josey, S. A., & Verein, R. (2012). First air-sea flux mooring measurements in the Southern Ocean. Geophysical Research Letters, 39(16).

Sentinel-1 Ocean Wind Fields (OWI) Algorithm Theoretical Basis Document (ATBD). (2019). Collecte Localisation Satellites (CLS). Ref: S1-TN-CLS-52-9049 Issue 2.0. Jun 2009.

Sentinel-1 Product Specification. (2020). Collecte Localisation Satellites (CLS). Ref: S1-RS-MDA-52-7441. Issue 3.7. Feb 2020.