Assessing improvements in global ocean pCO2 machine learning reconstructions with Southern Ocean autonomous sampling
The Southern Ocean plays an important role in the exchange of carbon between the atmosphere and oceans and is a critical region for the ocean uptake of anthropogenic CO 2 . However, estimates of the Southern Ocean air–sea CO 2 flux are highly uncertain due to limited data coverage. Increased samplin...
Published in: | Biogeosciences |
---|---|
Main Authors: | , , , , |
Format: | Text |
Language: | English |
Published: |
2024
|
Subjects: | |
Online Access: | https://doi.org/10.5194/bg-21-2159-2024 https://bg.copernicus.org/articles/21/2159/2024/ |
id |
ftcopernicus:oai:publications.copernicus.org:bg114808 |
---|---|
record_format |
openpolar |
spelling |
ftcopernicus:oai:publications.copernicus.org:bg114808 2024-09-15T18:36:58+00:00 Assessing improvements in global ocean pCO2 machine learning reconstructions with Southern Ocean autonomous sampling Heimdal, Thea H. McKinley, Galen A. Sutton, Adrienne J. Fay, Amanda R. Gloege, Lucas 2024-04-30 application/pdf https://doi.org/10.5194/bg-21-2159-2024 https://bg.copernicus.org/articles/21/2159/2024/ eng eng doi:10.5194/bg-21-2159-2024 https://bg.copernicus.org/articles/21/2159/2024/ eISSN: 1726-4189 Text 2024 ftcopernicus https://doi.org/10.5194/bg-21-2159-2024 2024-08-28T05:24:15Z The Southern Ocean plays an important role in the exchange of carbon between the atmosphere and oceans and is a critical region for the ocean uptake of anthropogenic CO 2 . However, estimates of the Southern Ocean air–sea CO 2 flux are highly uncertain due to limited data coverage. Increased sampling in winter and across meridional gradients in the Southern Ocean may improve machine learning (ML) reconstructions of global surface ocean p CO 2 . Here, we use a large ensemble test bed (LET) of Earth system models and the “ p CO 2 -Residual” reconstruction method to assess improvements in p CO 2 reconstruction fidelity that could be achieved with additional autonomous sampling in the Southern Ocean added to existing Surface Ocean CO 2 Atlas (SOCAT) observations. The LET allows for a robust evaluation of the skill of p CO 2 reconstructions in space and time through comparison to “model truth”. With only SOCAT sampling, Southern Ocean and global p CO 2 are overestimated, and thus the ocean carbon sink is underestimated. Incorporating uncrewed surface vehicle (USV) sampling increases the spatial and seasonal coverage of observations within the Southern Ocean, leading to a decrease in the overestimation of p CO 2 . A modest number of additional observations in Southern Hemisphere winter and across meridional gradients in the Southern Ocean leads to an improvement in reconstruction bias and root-mean-squared error (RMSE) of as much as 86 % and 16 %, respectively, as compared to SOCAT sampling alone. Lastly, the large decadal variability of air–sea CO 2 fluxes shown by SOCAT-only sampling may be partially attributable to undersampling of the Southern Ocean. Text Southern Ocean Copernicus Publications: E-Journals Biogeosciences 21 8 2159 2176 |
institution |
Open Polar |
collection |
Copernicus Publications: E-Journals |
op_collection_id |
ftcopernicus |
language |
English |
description |
The Southern Ocean plays an important role in the exchange of carbon between the atmosphere and oceans and is a critical region for the ocean uptake of anthropogenic CO 2 . However, estimates of the Southern Ocean air–sea CO 2 flux are highly uncertain due to limited data coverage. Increased sampling in winter and across meridional gradients in the Southern Ocean may improve machine learning (ML) reconstructions of global surface ocean p CO 2 . Here, we use a large ensemble test bed (LET) of Earth system models and the “ p CO 2 -Residual” reconstruction method to assess improvements in p CO 2 reconstruction fidelity that could be achieved with additional autonomous sampling in the Southern Ocean added to existing Surface Ocean CO 2 Atlas (SOCAT) observations. The LET allows for a robust evaluation of the skill of p CO 2 reconstructions in space and time through comparison to “model truth”. With only SOCAT sampling, Southern Ocean and global p CO 2 are overestimated, and thus the ocean carbon sink is underestimated. Incorporating uncrewed surface vehicle (USV) sampling increases the spatial and seasonal coverage of observations within the Southern Ocean, leading to a decrease in the overestimation of p CO 2 . A modest number of additional observations in Southern Hemisphere winter and across meridional gradients in the Southern Ocean leads to an improvement in reconstruction bias and root-mean-squared error (RMSE) of as much as 86 % and 16 %, respectively, as compared to SOCAT sampling alone. Lastly, the large decadal variability of air–sea CO 2 fluxes shown by SOCAT-only sampling may be partially attributable to undersampling of the Southern Ocean. |
format |
Text |
author |
Heimdal, Thea H. McKinley, Galen A. Sutton, Adrienne J. Fay, Amanda R. Gloege, Lucas |
spellingShingle |
Heimdal, Thea H. McKinley, Galen A. Sutton, Adrienne J. Fay, Amanda R. Gloege, Lucas Assessing improvements in global ocean pCO2 machine learning reconstructions with Southern Ocean autonomous sampling |
author_facet |
Heimdal, Thea H. McKinley, Galen A. Sutton, Adrienne J. Fay, Amanda R. Gloege, Lucas |
author_sort |
Heimdal, Thea H. |
title |
Assessing improvements in global ocean pCO2 machine learning reconstructions with Southern Ocean autonomous sampling |
title_short |
Assessing improvements in global ocean pCO2 machine learning reconstructions with Southern Ocean autonomous sampling |
title_full |
Assessing improvements in global ocean pCO2 machine learning reconstructions with Southern Ocean autonomous sampling |
title_fullStr |
Assessing improvements in global ocean pCO2 machine learning reconstructions with Southern Ocean autonomous sampling |
title_full_unstemmed |
Assessing improvements in global ocean pCO2 machine learning reconstructions with Southern Ocean autonomous sampling |
title_sort |
assessing improvements in global ocean pco2 machine learning reconstructions with southern ocean autonomous sampling |
publishDate |
2024 |
url |
https://doi.org/10.5194/bg-21-2159-2024 https://bg.copernicus.org/articles/21/2159/2024/ |
genre |
Southern Ocean |
genre_facet |
Southern Ocean |
op_source |
eISSN: 1726-4189 |
op_relation |
doi:10.5194/bg-21-2159-2024 https://bg.copernicus.org/articles/21/2159/2024/ |
op_doi |
https://doi.org/10.5194/bg-21-2159-2024 |
container_title |
Biogeosciences |
container_volume |
21 |
container_issue |
8 |
container_start_page |
2159 |
op_container_end_page |
2176 |
_version_ |
1810480716358615040 |