Forward model emulator for atmospheric radiative transfer using Gaussian processes and cross validation

<p>Remote sensing of atmospheric carbon dioxide (CO<span class="inline-formula"><sub>2</sub></span>) carried out by NASA's Orbiting Carbon Observatory-2 (OCO-2) satellite mission and the related uncertainty quantification effort involve repeated evaluatio...

Full description

Saved in:
Bibliographic Details
Main Authors: O. Lamminpää, J. Susiluoto, J. Hobbs, J. McDuffie, A. Braverman, H. Owhadi
Format: Article
Language:English
Published: Copernicus Publications 2025-02-01
Series:Atmospheric Measurement Techniques
Online Access:https://amt.copernicus.org/articles/18/673/2025/amt-18-673-2025.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1832086517966700544
author O. Lamminpää
J. Susiluoto
J. Hobbs
J. McDuffie
A. Braverman
H. Owhadi
author_facet O. Lamminpää
J. Susiluoto
J. Hobbs
J. McDuffie
A. Braverman
H. Owhadi
author_sort O. Lamminpää
collection DOAJ
description <p>Remote sensing of atmospheric carbon dioxide (CO<span class="inline-formula"><sub>2</sub></span>) carried out by NASA's Orbiting Carbon Observatory-2 (OCO-2) satellite mission and the related uncertainty quantification effort involve repeated evaluations of a state-of-the-art atmospheric physics model. The retrieval, or solving an inverse problem, requires substantial computational resources. In this work, we propose and implement a statistical emulator to speed up the computations in the OCO-2 physics model. Our approach is based on Gaussian process (GP) regression, leveraging recent research on kernel flows and cross validation to efficiently learn the kernel function in the GP. We demonstrate our method by replicating the behavior of OCO-2 forward model within measurement error precision and further show that in simulated cases, our method reproduces the CO<span class="inline-formula"><sub>2</sub></span> retrieval performance of OCO-2 setup with computational time that is orders of magnitude faster. The underlying emulation problem is challenging because it is high-dimensional. It is related to operator learning in the sense that the function to be approximated maps high-dimensional vectors to high-dimensional vectors. Our proposed approach is not only fast but also highly accurate (its relative error is less than 1  %). In contrast with artificial neural network (ANN)-based methods, it is interpretable, and its efficiency is based on learning a kernel in an engineered and expressive family of kernels.</p>
format Article
id doaj-art-8b5863d96c8e4ad18ce0095796b3e6d6
institution Kabale University
issn 1867-1381
1867-8548
language English
publishDate 2025-02-01
publisher Copernicus Publications
record_format Article
series Atmospheric Measurement Techniques
spelling doaj-art-8b5863d96c8e4ad18ce0095796b3e6d62025-02-06T14:48:22ZengCopernicus PublicationsAtmospheric Measurement Techniques1867-13811867-85482025-02-011867369410.5194/amt-18-673-2025Forward model emulator for atmospheric radiative transfer using Gaussian processes and cross validationO. Lamminpää0J. Susiluoto1J. Hobbs2J. McDuffie3A. Braverman4H. Owhadi5Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA, USAJet Propulsion Laboratory, California Institute of Technology, Pasadena, CA, USAJet Propulsion Laboratory, California Institute of Technology, Pasadena, CA, USAJet Propulsion Laboratory, California Institute of Technology, Pasadena, CA, USAJet Propulsion Laboratory, California Institute of Technology, Pasadena, CA, USAComputing and Mathematical Sciences Department, California Institute of Technology, Pasadena, CA, USA<p>Remote sensing of atmospheric carbon dioxide (CO<span class="inline-formula"><sub>2</sub></span>) carried out by NASA's Orbiting Carbon Observatory-2 (OCO-2) satellite mission and the related uncertainty quantification effort involve repeated evaluations of a state-of-the-art atmospheric physics model. The retrieval, or solving an inverse problem, requires substantial computational resources. In this work, we propose and implement a statistical emulator to speed up the computations in the OCO-2 physics model. Our approach is based on Gaussian process (GP) regression, leveraging recent research on kernel flows and cross validation to efficiently learn the kernel function in the GP. We demonstrate our method by replicating the behavior of OCO-2 forward model within measurement error precision and further show that in simulated cases, our method reproduces the CO<span class="inline-formula"><sub>2</sub></span> retrieval performance of OCO-2 setup with computational time that is orders of magnitude faster. The underlying emulation problem is challenging because it is high-dimensional. It is related to operator learning in the sense that the function to be approximated maps high-dimensional vectors to high-dimensional vectors. Our proposed approach is not only fast but also highly accurate (its relative error is less than 1  %). In contrast with artificial neural network (ANN)-based methods, it is interpretable, and its efficiency is based on learning a kernel in an engineered and expressive family of kernels.</p>https://amt.copernicus.org/articles/18/673/2025/amt-18-673-2025.pdf
spellingShingle O. Lamminpää
J. Susiluoto
J. Hobbs
J. McDuffie
A. Braverman
H. Owhadi
Forward model emulator for atmospheric radiative transfer using Gaussian processes and cross validation
Atmospheric Measurement Techniques
title Forward model emulator for atmospheric radiative transfer using Gaussian processes and cross validation
title_full Forward model emulator for atmospheric radiative transfer using Gaussian processes and cross validation
title_fullStr Forward model emulator for atmospheric radiative transfer using Gaussian processes and cross validation
title_full_unstemmed Forward model emulator for atmospheric radiative transfer using Gaussian processes and cross validation
title_short Forward model emulator for atmospheric radiative transfer using Gaussian processes and cross validation
title_sort forward model emulator for atmospheric radiative transfer using gaussian processes and cross validation
url https://amt.copernicus.org/articles/18/673/2025/amt-18-673-2025.pdf
work_keys_str_mv AT olamminpaa forwardmodelemulatorforatmosphericradiativetransferusinggaussianprocessesandcrossvalidation
AT jsusiluoto forwardmodelemulatorforatmosphericradiativetransferusinggaussianprocessesandcrossvalidation
AT jhobbs forwardmodelemulatorforatmosphericradiativetransferusinggaussianprocessesandcrossvalidation
AT jmcduffie forwardmodelemulatorforatmosphericradiativetransferusinggaussianprocessesandcrossvalidation
AT abraverman forwardmodelemulatorforatmosphericradiativetransferusinggaussianprocessesandcrossvalidation
AT howhadi forwardmodelemulatorforatmosphericradiativetransferusinggaussianprocessesandcrossvalidation