Electromagnetic Computation in Scattering of Electromagnetic Waves by Random Rough Surface and Dense Media in Microwave Remote Sensing of Land Surfaces

Abstract

Active and passive microwave remote sensing has been used for monitoring the soil moisture and snow water equivalent. In the interactions of microwaves with bare soil, the effects are determined by scattering of electromagnetic waves by random rough surfaces. In the interactions of microwaves with terrestrial snow, the effects are determined by volume scattering of dense media characterized by densely packed particles. In this paper, we review the electromagnetic full-wave simulations that we have conducted for such problems. In volume scattering problems, one needs many densely packed scatterers in a random medium sample to simulate the physical solutions. In random rough surface scattering problems, one needs many valleys and peaks in the sample surface. In random media and rough surface problems, the geometric characterizations of the media and computer generations of statistical samples of the media are also challenges besides electromagnetic computations. In the scattering of waves by soil surfaces, we consider the soil to be a lossy dielectric medium. The random rough surface is characterized by Gaussian random processes with exponential correlation functions. Surfaces of exponential correlation functions have fine-scale structures that cause significant radar backscattering in active microwave remote sensing. Fine-scale features also cause increase in emission in passive microwave remote sensing. We apply Monte Carlo simulations of solving full 3-D Maxwell's equations for such a problem. A hybrid UV/PBTG/SMCG method is developed to accelerate method of moment solutions. The results are illustrated for coherent waves and incoherent waves. We also illustrate bistatic scattering, backscattering, and emissivity which are signatures measured in microwave remote sensing. For the case of scattering by terrestrial snow, snow is a dense medium with densely packed ice grains. We have used two models: densely packed particles and bicontinuous media. For the case of densely packed particles, we used the Metropolis shuffling method to simulate the positions of particles. The particles are also allowed to have adhesive properties. The Foldy-Lax equations of multiple scattering are used to study scattering from the densely packed spherical particles. The results are illustrated for the coherent waves and incoherent waves. For the case of bicontinuous media, the method developed by Cahn is applied to construct the interfaces from a large number of stochastic sinusoidal waves with random phases and directions. The volume scattering problem is then solved by using CGS-FFT. We illustrate the results of frequency and polarization dependence of such dense media scattering.