Towards a unified theoretical model of ocean backscatter for wind speed retrieval from SAR, scatterometer, and altimeter

Abstract

INTRODUCTIONEmpirical models of ocean radar backscatter have shown that spaceborne radars can provide quantitative information about wind fields. Examples include the CMOD and IFREMER-2 models developed for the ERS Wind Scatterometer. However, such models are limited because they do not include all the relevant physical effects, such as the influence of swell, rain, surface slicks and currents. Recent theoretical developments have addressed these aspects, and here we consider the construction of a 'unified' theoretical model, that is, one with a common set of parameters applicable to SAR, scatterometer and altimeter observations. MODEL DEVELOPMENTThe model requires a scattering theory to be applied to a theoretical description of the ocean waveheight spectra. We compare predictions from the composite-surface and IEM (integral equation method) scattering models, using ocean spectra described by the models of Apel (1994), Elfouhaily et al. (1997) and Lemaire et al. (1999). We also incorporate a swell component and limited fetch, and we use published models of the perturbations to the wave spectrum due to the presence of rain, surface slicks or currents. The resulting model shows the sensitivity of the radar backscatter to all these parameters. It also predicts dependences on radar parameters which are relevant for future missions such as Envisat.Finally, it predicts the circumstances when the factors neglected by the empirical models such as CMOD will be significant. For example, it is found that the greatest effect on C-band, VV-polarised backscatter from swell is at low incidence angles, while from rain it is at high incidence angles (Figure 1).Note that the atmospheric attenuation from rain is not included in the modelling.New empirical algorithms for the recovery of wind speeds from altimeter data are developed by applying nonlinear functionfitting procedures to an extensive set of Topex data collocated with buoy measurements. These eliminate a systematic bias in the Witter & Chelton (1991) algorithm. The best results are obtained when using the altimeter Ku-band backscatter coefficient together with the altimeter significant wave height measured at C band (Figure 2). MODEL TESTSThe theoretical model reproduces the empirical dependence on wind speed at nadir very well, but it under-predicts the empirical dependence on swell properties. This discrepancy is reduced if an influence of swell on the relation between wind stress and wind speed is incorporated. This model is also tested by comparison with empirical scatterometer models at C and Ku band. The results indicate that the accuracy of the theoretical model is limited by the present understanding of the description of ocean wave spectra. For example, the IFREMER-2 model for C band, VV polarisation shows best agreement with the results from the Lemaire et al. (1999) spectrum, although there is a systematic difference at high wind speeds and high incidence angles (Figure 3). Also, the directional spread of waves is not well understood; this leads to a systemat...

Detection and Processing of bistatically reflected GPS signals from low Earth orbit for the purpose of ocean remote sensing

“…It may prove more useful to concentrate on empirical methods for sensing various ocean parameters, taking as an example current scatterometry methods for sensing winds. Given the early stage of this technology, we should be open to both sides of the modeling and/or empirical debate, and eventually be able to compare, or merge, one with the other in constructive ways, as was done for the case of scatterometry in [21].…”