Satellite instruments such as OMI (on NASA Aura platform 2004), TROPOMI (launched October 2017), TEMPO (upcoming NASA/SAO geostationary instrument) measure backscattered light in the UV, visible and near-infra-red parts of the electromagnetic spectrum. Target retrievals include O3 (profiles and columns), total amounts of NO2, SO2, HCHO, CH4, BrO and other trace species, and other quantities such as aerosol optical properties, cloud cover and surface properties.
To retrieve atmospheric and/or/ surface quantities from remote-sensing measurements, we need a radiative transfer (RT) model that will simulate these measurements. Retrieved quantities are determined by an inverse process based on comparisons between real and simulated light fields. RT models are based on Maxwell’s theory of electromagnetism, and in the Earth’s atmopshere we need to allow for scattering and extinction of sunlight by aerosols, molecules and gas absorbers, as well as reflected light from a variety of surfaces. For satellites measuring in the infra-red, the RT models should also deal with thermal and surface emission.
The LIDORT (scalar-only) and VLIDORT (with polarization) models are Linearized Discrete Ordinate Radiative Transfer software packages. In particular, the linearization capability – that is, the simultaneous generation of analytically-derived Jacobians or weighting functions with respect to any atmospheric and/or surface property – is extremely useful for remote-sensing retrieval applications and sensitivity analyses. The (V)LIDORT Fortran 90 codes are freely available to the public, and there are many users in the satellite and ground-based remote-sensing communities in the Americas, Europe and Asia.
In the Introduction to this talk, I will first give some general background information on satellite remote-sensing atmospheric chemistry instruments, and set the scene for retrievals of O3 and other constituents. Then I will summarize discrete-ordinate RT theory for multiple scattering in a multi-layer atmosphere, and discuss the preparation of optical property inputs to the models, plus the use of BRDF and water-leaving/SIF supplements for the treatment of surfaces. Several examples will show the flexibility and power of these RT models for a range of remote-sensing applications. Of relevance to a project here at UNM are RT simulations for the TEMPO satellite instrument. The talk concludes with a discussion of the (V)LIDORT model capabilities and some practical aspects of the models, and a brief note on other RT models available from RT Solutions.
Algunas de sus publicaciones se pueden consultar en:
Participante: Dr. Robert Spurr
Institución: Director de RT Solutions, Inc.
Fecha y hora: Este evento terminó el Miércoles, 28 de Noviembre de 2018