The SASKTRAN Thermal Infrared Radiative Transfer Model with Analytic Jacobian Calculations

Abstract

The SASKTRAN radiative transfer framework is a set of software tools for modelling the propagation of electromagnetic radiation through the atmosphere. It is an important tool in modelling the measurements made by remote sensing instruments in order to retrieve information about the structure and composition of the atmosphere from observations of spectral radiance. The primary goals of this thesis are to expand the capabilities of SASKTRAN to modelling the infrared spectral regime and test this new suite of modelling software for use with real atmospheric remote sensing data. This model will be used to retrieve information on the concentrations of greenhouse gases from measurements made by the Limb Imaging Fourier transform spectrometer Experiment (LIFE), a remote sensing instrument developed at the University of Saskatchewan.

The Thermal InfraRed (TIR) model, developed in this thesis, incorporates the absorption and emission of radiation from greenhouse gases to solve the radiative transfer problem. The fundamental output of this model is the spectral radiance that is incident on an observer positioned at a given location and looking in a given direction through the atmosphere. Radiative transfer is a non-linear problem and complex inverse methods are needed to determine atmospheric composition from instrument measurements. The inversion requires the computation of Jacobians, which quantify how the spectral radiance changes when changes are made to the atmospheric state parameters. Analytic Jacobian calculations are implemented in the TIR model and simultaneously calculated alongside the modelled radiance. With analytic calculations significant computational savings are achieved because the model does not need to be executed for each parameter that is changed. The TIR model supports the computation of analytic Jacobians for vertical profiles of greenhouse gas composition and temperatures.

Fourier Transform Spectrometers (FTS) measure interferograms of light that are converted into radiances via Fourier transform. LIFE employs the FTS technique to measure vertical images of atmospheric spectral radiance in the thermal infrared regime. This instrument is a primary motivation for the development of the TIR model, which is to be implemented to retrieve information about the vertical composition of greenhouse gases such as methane, water vapor, ozone, and nitrous oxide. In this thesis the TIR model is used to simulate LIFE measurements and demonstrate the feasibility of retrieving greenhouse gas profiles with the aid of analytic Jacobian calculations.