Piotr Sobieski

Rain generated ring-waves: Measurements and modelling for remote sensing

We present an analysis of ring-wave and scatterometer data from a water surface that was agitated by simulated rain. Water droplets of 2.8 mm diameter impacted the water surface at almost terminal velocity, and the rain rates cover a wide range of conditions (5 to 200 mm hr(-1)). Both the ring-wave energy and backscattered power from the GHz scatterometer increase as R increases, but the growth rates slacken at higher rain intensities. Ring-wave frequency spectra and wavenumber spectra are well represented by log-Gaussian spectral models. The results can be used to guide development of microwave scattering models.

An Approximate Model for the Microwave Brightness Temperature of the Sea

Abstract A modified two-scale model is proposed for scattering and emissivity calculations for certain classes of random rough surfaces. It is based on an approach by Burrows and by Brown, but it has been extended to bistatic scattering by lossy dielectric surfaces, and it incorporates modified Fresnel reflection coefficients and a simple correction for multiple-scattering effects. The method is shown to be applicable to the ocean surface for light and moderate winds. A contracted form of the radiative-transfer equation is proposed and the included Wentz correction for surface scattering is discussed. This could lead to a method that could be both simple and accurate enough for real-time inversion algorithms in microwave remote sensing.

An analysis of scatterometer returns from a water surface agitated by artificial rain: evidence that ring-waves are the main feature

Both wind and rain roughen the sea surface, but whereas wind generates waves, rain generates craters, stalks and ring-waves. Average backscattered power for scatterometer returns from water surfaces is closely related to small scale features on the water surface, so we use backscattered power from short wind-waves as a basis to evaluate the importance of ring-waves. Experiments were conducted with a 13.5 GHz scatterometer (30-degrees incidence angle, vertical polarization) in a wind-wave tank that is enhanced by a rain simulator. Rain intensities ranged from 3-30 mm h-1 and wind friction velocities were between 10 and 50 cm s-1. The variance of sur-face elevation for small scale features xi(sm)2, i.e., ring-waves and short wind-waves, was computed for each case using data from a capacitance probe. Comparison of the data sets shows that the range of xi(sm)2 for the rain cases is comparable to that from light to moderate wind cases-so ring-wave amplitudes are not negligible. Analysis of the radar data provides evidence that ring-waves are the dominant feature contributing to the backscattered power. Thus ring-waves need to be included in scatterometer numerical models that contain rain effects.

Scatterometric signatures of multivariate drop impacts on fresh and salt water surfaces

Microwave signatures of the ocean surface are affected by wind and rain. To support the development of theoretical models for remote sensing applications, radar scatterometry experiments at 13.5 and 36 GHz with VV polarization were conducted at the Rain-Sea Interaction Facility at NASA Wallops. Backscatterings from rain drop impacts on fresh and salt water surfaces were measured with an incidence angle of 30 degrees for eight different drop sizes. Results are also presented for terminal and non-terminal fall velocities. Surface features were imaged by an ultrahigh-speed digital camera synchronized with radar data acquisition and their geometrical characteristics were determined. Backscattered powers from crowns, craters, stalks and ring-waves were measured and compared. These measurements confirm that for slant radar configurations, ring-waves are the dominant scattering contribution, even though stalk scattering is not negligible.

Full-range sea surface spectrum in nonfully developed state for scattering calculations

A new form for the spectrum of the ocean surface vertical displacements is derived for the case of nonfully developed states. The gravity range is expressed as a function of the fetch x and the significant slope /spl conint/ as well. The capillary-gravity range is assumed dependent on the wind friction velocity only. Recent wavenumber spectrum measurements in this spectral domain and ocean conditions are analyzed. Tobas spectral shape is shown to represent correctly these experimental data when updated with an equivalent amplitude factor. An expression for this factor is proposed. It is weakly wind friction velocity dependent, as observed by Mitsuyasu in the late 1970s. The proposed spectrum is then combined with a boundary perturbation model for electromagnetic scattering computations. Empirical scattering models and radar data collocated with assumed ground-truth data are used for comparison. This is shown to give consistent results for both C- and Ku-bands as well as large ranges of wind speeds and incidence angles. Comparisons of backscattering coefficients computed using other sea spectra from the literature are presented. The significant slope is found to be an important factor for scattering at low incidence angles. The proposed spectrum thus constitutes a useful basis for physically based inversion algorithms.

Scattering by artificial wind and rain roughened water surfaces at oblique incidences

Rain affects wind retrievals from scatterometric measurements of the sea surface. To depict the additional roughness caused by rain on a wind driven surface, we use a ring-wave spectral model. This enables us to analyse the rain effect on K-u band scatterometric observations from two laboratory experiments. Calculations based on the small perturbation method provide good simulation of scattering measurements for the rain-only case, whereas for combined wind and rain cases, the boundary perturbation method is appropriate.

The Two-Scale BPM Scattering Model for Sea Biogenic Slicks Contrast

Sea oil slick observation by means of synthetic aperture radar is still a scientific and operational challenge. In this paper, the sea surface scattering with and without biogenic slicks is analyzed by using the two-scale Boundary Perturbation Method scattering model. The surface slick is supposed to modify both the full-range sea surface spectrum and the slope probability density function by means of the Marangoni damping and by a reduced friction velocity. In this paper, the full-range Universite Catholique de Louvain sea surface spectrum is considered. A new contrast model, which overcomes the drawbacks of the contrast model based on the untilted Small Perturbation Method scattering model, is presented and illustrated in some L- and C-band biogenic slick cases.

Drop size effects on rain-generated ring-waves with a view to remote sensing applications

This paper presents an analysis of drop size effects on ring-wave spectra and radar scatterometer returns from a water surface agitated by artificial rain. For this purpose, monodisperse and polydisperse rain events were generated in the laboratory for a wide range of rain rates and various drop sizes. The water droplets reached the surface at terminal velocity. In all cases, the radar average power is well modelled by a linear function of the power spectral density at the Bragg resonant wavelength. The drop size is found to have a strong impact on the spectral shape of the ring-waves and on their total energy. A log-Gaussian model characterizes well the ring-wave spectra and empirical expressions of the spectrum parameters are given. Ring-wave energy increases with rain rate and drop size, and is found to be proportional to the kinetic energy of a single drop, indicating that one may use a model in which all drops contribute to the ring-wave energy in proportion to their squared momentum. The results from the monodisperse rain experiments are used to construct a model for natural rain. Data from the polydisperse rain experiments show that a nonlinear model which relates dissipation to the total rainfall rate provides excellent agreement with the measurements. This analysis also shows the important impact of a few large drops on the ring-wave spectrum. The model proposed can be extended to natural rains either by using measurements of the drop size distribution or by assuming a drop size distribution model that is appropriate to the study region. It is concluded that it is important to characterize ring-wave spectra as a function of rain rate and drop size distribution to develop robust radar scattering models for rain-roughened seas.

Synergic inversion technique for active and passive microwave remote sensing of the ocean

A unified approach for combining active and passive microwave measurements for remote sensing applications is described. A synergic inversion technique has been developed and applied to the retrieval of geophysical parameters of the ocean surface and of the atmosphere. It is based on the combination of radiometric and radar measurements at the electromagnetic and cell level and not only on the correction of radar measurements by radiometric measurements, or conversely. Such a combination is performed through a common quantity: the bistatic scattering coefficient of the observed surface. This is used in a direct model to simulate combined measurements from active and passive sensors. It requires a rather complete and accurate calculation of the scattering of microwaves by the rough sea surface. >

Ring-waves generated by water drops impacting on water surfaces at rest

Radar observations of the ocean surface can be affected by impacting raindrops. Ring-wave measurements are presented for drops of 2.2 and 2.8 mm in diameter impacting on fresh and salt water surfaces initially at rest. They are based on the observation of the mirror image of a sharp edge on the perturbed surface. The retrieved wave profiles show a rather stable characteristic wavenumber (0.2 mm/sup -1/) and very small wave amplitudes: the fraction of the incident kinetic energy converted into ring-waves is of the order of 1%.