Hurricane Winds from Synthetic Aperture Radar: Recent Developments

This presentation summarizes the recent advances and progress that have been achieved in monitoring hurricane winds with Synthetic Aperture Radar (SAR), with a focus on the contributions of the Canadian HurricaneWatch program, which began with RADARSAT-1 and continues with RADARSAT-2. The capability of SAR wind retrieval under low – to – moderate wind speeds is generally accepted for co-polarization (co-pol) SAR measurements, based on the CMOD5 series of Geophysical Model Functions (GMF). However, for co-pol SAR, the radar backscattered signal becomes saturated under high wind conditions, and depends on the incidence angle of the radar measurements, causing the wind speed ambiguity problem for high winds. The speed ambiguity problem can be removed by assuming that the hurricane wind fields follow Holland vortex-type structures. However, recent progress offers an alternative. RADARSAT-2 provides fully polarimetric mode SAR images. The high radiometric calibration and very low noise floor of cross-pol images from RADARSAT-2 offers a new methodology for the retrieval of hurricane winds from SAR. Recent studies show that (1) there is a monotonic relationship between the cross-pol radar returns and wind speed, (2) such relationship doesn't saturate under high wind speed conditions. Therefore, the cross-pol GMF exhibits a very simple relationship between radar signals and wind speed, greatly enhancing the potential of this methodology to support marine forecasting in hurricane conditions. The development of GMFs for both co-pol and cross-pol SAR data has been based on the collocation of radar data and in-situ measurements, or reliable numerical model results, and covers only limited wind speeds (<30 m/s). Until now, the application of this approach to the retrieval of super high winds has been based on extrapolation. In this presentation, we present detailed evaluation of these GMFs under very high winds. It will be shown that a number of factors, such as ocean wave-breaking, and heavy rain, can influence the accuracy of the retrieved winds, discussed in the presentation.

Selected Publications: [1]Shen, H., W. Perrie, and Y. He. 2016. Evaluation of Hurricane Wind Speed Retrieval from Cross Dual-pol SAR. International Journal of Remote Sensing. Accepted. [2]Shen, H., W. Perrie, Y. He and G. Liu, 2014. Wind speed retrieval from VH dual-polarization RADARSAT-2 SAR images. Geoscience and Remote Sensing, IEEE Transactions on.vol.33, pp.1-7. [3]Li, D.and Shen H. 2014.Evaluation of wind vectors observed by HY-2 scatterometer using ocean buoy observations, ASCAT measurements, and numerical model data, Chinese Journal of Oceanology and Limnology, http://dx.doi.org/10.1007/s00343-015-4136-4. [4]Zhang, B., and W. Perrie. 2012. “Cross-polarized synthetic aperture radar: A new potential measurement technique for hurricanes,” Bull. Amer. Meteor. Soc., 93, 531–541, doi:10.1175/BAMS-D-11-00001.1. [5]Shen, H., Y. He, and W. Perrie. 2009. “Speed ambiguity in hurricane wind retrieval from SAR imagery.” International Journal of Remote Sensing, 30, 2827–2836, doi:10.1080/01431160802555879. [6]Perrie, W., W. Zhang, M.Bourassa, H. Shen, P. W. Vachon, 2008: Impact of Satellite Winds on Marine Wind Simulations. Wea. & Forecasting, 23, 290-303. doi: http://dx.doi.org/10.1175/2007WAF2006093.1 [7]Shen, H., W. Perrie, and Y. He. 2006. “A new hurricane wind retrieval algorithm for SAR images.” Geophys. Res. Lett., 33, doi:10.1029/2006GL027087.