|
Abstract:
|
Remote sensing instruments can provide high-resolution and high-volume data to inform numerous physical and environmental processes. Typically the quantities of interest, such as the composition of the Earth's or other planetary atmosphere, must be inferred from the information in satellite radiance spectra. The inference can often involve complex, nonlinear, computationally expensive physical models, so a variety of tools for uncertainty quantification are used. Understanding the sources of uncertainty can provide additional insight for the scientific applications that benefit from the use of satellite data. The discussion will address the tradeoffs between making optimal inference and practical considerations such as data volume for global remote sensing efforts.
|
