Effects of sea maturity on satellite altimeter measurements

Abstract

For equilibrium arid near-equilibrium sea states, the Wave.slope variance is a function of wind speed U and of the sea maturity. The influence of both factors on the altimeter measurements of wind speed, wave height, and radar cross section cr ø is studied experimentally. The necessary theoretical background, both on wave statistics and on the near-nadir radar backscatter, is outlined. It is based on a recently proposed view that the effective fractal dimension for a surface patch corresponding to the equilibrium range of wave spectra increases as the wave field matures. The experimental part is based on ! year's worth of Geosat altimeter observations colocated with in situ wind and wave measurements by 20 NOAA buoys. Errors and biases in altimeter wind speed and wave height measurements are investigated. A geophysically significant error trend correlated with the sea maturity (i.e., with Co/U, where Co is the phase velocity of waves corresponding to the spectral peak and U is the mean wind speed well above the surface) is found in wind speed measurements. This trend is explained by examining the effect of the generalized wind fetch, X oc gH•/3/U 2, on the curves of the observed dependence cr ø =f(U, X). We conclude, in particular, that unambiguous measurements of wind speed by altimeter, in a wide range of sea states, are impossible without accounting for the actual degree of wave development. An approach to a more accurate retrieval of wind speeds is suggested as based on accounting for the actual degree of wave development, which can be achieved through processing the total radar cross section simultaneously with the significant wave height information contained in altimeter waveforms. The potential capability of a satellite altimeter to estimate wave age and similar measures of wind-wave coupling is discussed. This capability expands the use of satellite altimeters for air-sea interaction studies. lated with errors in microwave radiometer-and scatterometer-reported winds [Glazman, 1987; Glazman et al., 1988; Glazman and Pilorz, 1989]. An explanation of the error trend in the scatterometer data was proposed based on the notion of a non-Bragg ("spike") component ors ø of the total radar return, cr ø = o'• ø + o'•, arising at oblique incidence. Here o-• ø is due to the Bragg-type scattering from small tipples, while cr• o• o-iv accounts for an impact of wedgelike cusps and breaking wavelets [e.g., Kwoh and Lake, 1984; Glazman et al., 1988]; o 'l represents the mean contribution due to an individual, spike-generating wavelet, and v is the rate of steep wavelet occurrence. This rate is a strong function of the surface slope variance 72 = (IV•r12): v oc 7 -2 exp (-F2/272) where F is the threshold wave steepness causing a radar spike [Glazman, 1985, 1986; Glazrnan et al., 1988; Glazrnan and Welchman, !989]. Since the scales of the spike-producing wavelets pertain to the equilibrium range of the wave spectrum, the rms s16pe •/of these gravity waves is influenced by the degree of wave development. In the case of the a...

Influence of the type of sea waves on the backscattered radar cross section at medium incidence angles

“…The first classification of wave types was proposed, and it was shown that the backscattered radar cross section during nadir sounding (radio altimeter) and consequently the retrieved wind speed depend on the type of waves. Accounting for this effect in the algorithm made it possible to increase the accuracy of wind speed retrieval from the Geosat radio-altimeter data (Glazman and Pilorz, 1990).…”

SAR raw signal simulation of oil slicks in ocean environments

“…Such an equilibrium depends on the wind strength. When the wind stops blowing, the short waves decrease quickly, while the long waves propagate far from their source (and, in this case, are usually referred to as swell [18]). Accordingly, long waves are generated by strong far winds, at variance of short waves, which are generated by local winds.…”

“…They depend on the far wind intensity and direction and on the fetch area [18]. The algorithm for the long wave generation is detailed in [25].…”

Technical note: Impacts of collocation window on the accuracy of altimeter/buoy wind-speed comparison-a simulation study

“…In uence of the wind-speed variability Figure 4 (b) and (c) can be related reasonably to the cases of Geosat and TOPEX/ Poseidon, respectively. Fortunately, it was possible to nd some previous studies by Glazman and Pilorz (1990), Ebuchi et al (1992 ) and Gower (1996), who performed qualitative or quantitative analyses of the impacts of window sizes on the comparison results using real Geosat and TOPEX/Poseidon altimeter data in conjunction with the US, Japanese and Canadian buoy data. These give an opportunity to check the validity and applicability of the present simulation results.…”

“…Their conclusions, however, are somewhat divided. One argument suggests that the variations of space and time windows do not have a crucial impact on the accuracy of satellite/buoy comparisons (Dobson et al 1987, Glazman andPilorz 1990). In contrast, another argument suggests that reducing the spatial lag persistently improves the agreement of windspeed measurements from altimeters and ocean buoys, while time lags of up to 1 h do not produce signi cant changes in the key statistics, such as the root mean square (RMS) diOE erence and the correlation coe cients (Gower 1996, Hwang et al 1998 ).…”