Abstract
This article sets forth a practical methodology for uncertainty quantification of physical state estimates derived from remote sensing observing systems. Remote sensing instruments observe parts of the electromagnetic spectrum and use computational algorithms to infer the underlying true physical states. In current practice, many sources of uncertainty are not accounted for in this process, leading to underestimates of uncertainties on quantities of interest. We propose a procedure that combines Monte Carlo simulation experiments with statistical modeling to approximate distributions of unknown true states given point estimates of those states. Our method is carried out post hoc, that is, after the operational processing step. We demonstrate the procedure using four months of data from NASA's Orbiting Carbon Observatory-2 mission and compare it to validation measurements from the Total Column Carbon Observing Network.
Highlights
The ability of space-borne remote sensing observations to address important Earth and climate science problems rests crucially on how well geophysical quantities of interest (QOIs) can be inferred from these data
While much of this sounds familiar in the context of the uncertainty quantification discipline [32], existing techniques do not address the problem in a practical way that can be applied comprehensively to very large data sets produced in routine operations
Our perspective is that the uncertainty to be quantified is that of the entire, end-to-end observing system shown in Figure 1, and so our goal is to provide the conditional distribution P (X| X\^ )
Summary
Observing systems that collect and process this information must address uncertainties arising from measurement errors, and from imperfect physical models and their parameters, computational artifacts, and potentially other unknowns that affect the conversion of observations to QOI estimates. While much of this sounds familiar in the context of the uncertainty quantification discipline [32], existing techniques do not address the problem in a practical way that can be applied comprehensively to very large data sets produced in routine operations. These spectra carry information about the properties of Earth's atmosphere and surface, as encountered in each individual observational unit corresponding to a specific ground footprint ( sometimes called a`sounding""), because photons at different wavelengths are scattered and absorbed in characteristic ways, depending on the makeup, function, and prop-
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
