Atmospheric effect on microwave polarimetric passive remote sensing of ocean surfaces

Abstract

A theoretical emission model of combined ocean surface and atmosphere is presented to predict the microwave emissivity of the ocean. The modeled ocean surface is one‐dimensional with a random rough profile. The electromagnetic scattering from the surface is calculated based on the extended boundary condition method. Realizations of rough surfaces are created using Monte Carlo simulations. The bistatic scattering coefficients are computed from the ensemble average. The millimeter‐wave propagation model is used to evaluate the absorption of microwave radiation at all height levels in the atmosphere. An expression for the observed brightness temperatures is derived by solving the radiative transfer equations. The radiative transfer model results show a good agreement with the measured data from the 1995 NASA Jet Propulsion Laboratory WIND radiometer (WINDRAD) campaign. An approximate model is provided to estimate the atmospheric effect on the ocean brightness temperatures based on the overall atmospheric attenuation. The approximate model also compares well with the WINDRAD data. Further comparisons are made between the approximate formula and the radiative transfer results on the ratio of the third Stokes parameter in the atmosphere to the one in free space by varying the atmospheric conditions, surface roughness, and radiation frequencies. The approximate formula shows its usefulness for the prediction of the ocean brightness temperatures.

An updated analysis of the ocean surface wind direction signal in passive microwave brightness temperatures

“…This process can be characterized by the atmospheric liquid cloud water density profile: . From the theory of radiative transfer [18], [44]- [46] one obtains for (3) The and are the up-and downwelling atmospheric brightness temperatures and are given by the atmospheric integrals (4) Hereby is the total atmospheric absorption coefficient and is the atmospheric temperature at location . The atmospheric transmittance between and is given by…”

“…It is therefore necessary to remove the isotropic part from (8) before doing the harmonic fit. In [18], [44]- [46] a model function has been derived for the isotropic part that is based on the theory of radiative transfer and empirical observations. Because the full atmospheric profiles for temperature , water vapor density and liquid cloud water density are not readily available, a simplified version has been established that depends only on the effective temperature of the ocean-atmosphere system (similar than in (10)) and the vertical columnar integrals for water vapor (15) and for liquid cloud water (16) Here, denotes the vertical distance from sea level and is the satellite altitude.…”

“…For fixed EIA, the model function for polarization depends therefore on , , , and . The signal (9) is then determined as (17) For our new analysis, we have used the most updated version of the isotropic model function, which can be found in [46]. This model function is able to reproduce the SSM/I brightness temperatures below 37 GHz to a RMS accuracy of 0.6 K [45].…”

On The Effect of Atmospheric Emission upon the Passive Microwave Polarimetric Response of an Azimuthally Anisotropic Sea Surface

“…This influence can be expressed only in terms of atmospheric attenuation as done in [31] or in the approximated model in [32]. However, we have verified that it is also function of the wind speed and this has to be properly taken into account in the model function to be considered for polarimetric radiometers.…”

“…In order to properly investigate the sensitivity of passive polarimetric response to the ocean wind field, the impact of atmospheric emission on the measured Stokes vector has to be quantified. The simulation of the apparent brightness temperature vector at the platform height has to include both the upwelling and downwelling atmospheric radiation together with the bistatic scattering of downwelling brightness temperatures due to the rough sea surface as done in [9] and in the more recent paper by Yeong et al [32]. To some extent, even the atmospheric stability in the boundary layer can have an impact on the polarimetric passive measurements [19].…”