Colloquium: Dr. Pengwang Zhai | UMBC
In-Person PHYS 401
Wednesday, September 25, 2024 · 11 AM - 12 PM
TITLE: "Radiative transfer and its applications in the Earth system”
ABSTRACT: Radiative transfer equation (RTE) is an integro-differential equation that describes the physics processes of electromagnetic waves propagating in turbid media. RTE has important applications in vast disciplines including astrophysics, Earth science, detection of cancer tissues, computer graphics, etc.. The solution of RTE exists in only a few ideal cases, i.e., isotropic or Rayleigh scattering media. In general people rely on complex numerical models to solve RTE, which is computationally expensive. Approximations are often used, aiming to both ease the development effort and increase the efficiency, which ignore the polarization of electromagnetic fields, multiple scattering of light in the medium, inelastic scattering, etc. My research is primarily focused on building a numerical model for solving RTE and applying it to a variety of remote sensing platforms for the Earth system. The model is called the Radiative Transfer model based on the Successive Order of Scattering (RTSOS) method, which uniquely combines the following features: 1.) the full polarized radiative transfer equation is solved 2.) light transportation in both the atmosphere-ocean and atmosphere-land systems are treated. 3.) scattering and absorption processes of both atmospheric and oceanic particles are included. 4.) both solar and thermal emissions are included, which enables a large spectrum coverage from ultraviolet to thermal infrared simulations. 5.) inelastic scattering processes in ocean waters are properly simulated, i.e., Raman scattering by pure ocean waters and fluorescence by plankton cells and colored dissolved organic matter. 6.) flexible sensor location, viewing geometry, and response function can be configured, which is important for remote sensing applications. The RTSOS model has been used in a number of remote sensing applications, including atmospheric correction of ocean color observation, remote sensing of aerosol and cloud microphysical properties using multiangle polarimeters. In addition, we also employ the newest machine learning technologies in expediting the numerical solutions of RTE, which is important in the operational processing of satellite observations. In this talk I will share my long-term research endeavors in these subjects as well as some personal tips to our graduate students for career development.
ABSTRACT: Radiative transfer equation (RTE) is an integro-differential equation that describes the physics processes of electromagnetic waves propagating in turbid media. RTE has important applications in vast disciplines including astrophysics, Earth science, detection of cancer tissues, computer graphics, etc.. The solution of RTE exists in only a few ideal cases, i.e., isotropic or Rayleigh scattering media. In general people rely on complex numerical models to solve RTE, which is computationally expensive. Approximations are often used, aiming to both ease the development effort and increase the efficiency, which ignore the polarization of electromagnetic fields, multiple scattering of light in the medium, inelastic scattering, etc. My research is primarily focused on building a numerical model for solving RTE and applying it to a variety of remote sensing platforms for the Earth system. The model is called the Radiative Transfer model based on the Successive Order of Scattering (RTSOS) method, which uniquely combines the following features: 1.) the full polarized radiative transfer equation is solved 2.) light transportation in both the atmosphere-ocean and atmosphere-land systems are treated. 3.) scattering and absorption processes of both atmospheric and oceanic particles are included. 4.) both solar and thermal emissions are included, which enables a large spectrum coverage from ultraviolet to thermal infrared simulations. 5.) inelastic scattering processes in ocean waters are properly simulated, i.e., Raman scattering by pure ocean waters and fluorescence by plankton cells and colored dissolved organic matter. 6.) flexible sensor location, viewing geometry, and response function can be configured, which is important for remote sensing applications. The RTSOS model has been used in a number of remote sensing applications, including atmospheric correction of ocean color observation, remote sensing of aerosol and cloud microphysical properties using multiangle polarimeters. In addition, we also employ the newest machine learning technologies in expediting the numerical solutions of RTE, which is important in the operational processing of satellite observations. In this talk I will share my long-term research endeavors in these subjects as well as some personal tips to our graduate students for career development.