An improved long-term high-resolution surface pCO2 data product for the Indian Ocean using machine learning

Abstract Accurate estimation of surface ocean pCO2 is crucial for understanding the ocean’s role in the global carbon cycle and its response to climate change. In this study, we employ a machine learning algorithm to correct the deviations in high-resolution (1/12°) model simulations of surface pCO2...

Full description

Saved in:
Bibliographic Details
Main Authors: Prasanna Kanti Ghoshal, A.P. Joshi, Kunal Chakraborty
Format: Article
Language:English
Published: Nature Portfolio 2025-04-01
Series:Scientific Data
Online Access:https://doi.org/10.1038/s41597-025-04914-z
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849733423321579520
author Prasanna Kanti Ghoshal
A.P. Joshi
Kunal Chakraborty
author_facet Prasanna Kanti Ghoshal
A.P. Joshi
Kunal Chakraborty
author_sort Prasanna Kanti Ghoshal
collection DOAJ
description Abstract Accurate estimation of surface ocean pCO2 is crucial for understanding the ocean’s role in the global carbon cycle and its response to climate change. In this study, we employ a machine learning algorithm to correct the deviations in high-resolution (1/12°) model simulations of surface pCO2 from the INCOIS-BIO-ROMS model (pCO2 model) for the period 1980–2019, using available observations (pCO2 obs). We train the XGBoost model to generate spatio-temporal deviations (pCO2 obs − pCO2 model) of pCO2 model. The interannually and climatologically varying deviations are then added back to the original model separately, which results in an improved surface pCO2 data product. A comparison of our surface pCO2 data product with moored observations, gridded SOCAT, CMEMS-LSCE-FFNN, and OceanSODA demonstrates an improvement by approximately 40% ± 3.31% in RMSE. Further analysis reveals that adding climatological deviations to pCO2 model results in greater improvements than adding interannual deviations. This analysis underscores the ability of machine learning algorithms to enhance the accuracy of model-simulated surface pCO2 outputs.
format Article
id doaj-art-5350c95cab8a4727841c8d6376e73c1d
institution DOAJ
issn 2052-4463
language English
publishDate 2025-04-01
publisher Nature Portfolio
record_format Article
series Scientific Data
spelling doaj-art-5350c95cab8a4727841c8d6376e73c1d2025-08-20T03:08:02ZengNature PortfolioScientific Data2052-44632025-04-0112111110.1038/s41597-025-04914-zAn improved long-term high-resolution surface pCO2 data product for the Indian Ocean using machine learningPrasanna Kanti Ghoshal0A.P. Joshi1Kunal Chakraborty2Indian National Centre for Ocean Information Services, Ministry of Earth SciencesIndian National Centre for Ocean Information Services, Ministry of Earth SciencesIndian National Centre for Ocean Information Services, Ministry of Earth SciencesAbstract Accurate estimation of surface ocean pCO2 is crucial for understanding the ocean’s role in the global carbon cycle and its response to climate change. In this study, we employ a machine learning algorithm to correct the deviations in high-resolution (1/12°) model simulations of surface pCO2 from the INCOIS-BIO-ROMS model (pCO2 model) for the period 1980–2019, using available observations (pCO2 obs). We train the XGBoost model to generate spatio-temporal deviations (pCO2 obs − pCO2 model) of pCO2 model. The interannually and climatologically varying deviations are then added back to the original model separately, which results in an improved surface pCO2 data product. A comparison of our surface pCO2 data product with moored observations, gridded SOCAT, CMEMS-LSCE-FFNN, and OceanSODA demonstrates an improvement by approximately 40% ± 3.31% in RMSE. Further analysis reveals that adding climatological deviations to pCO2 model results in greater improvements than adding interannual deviations. This analysis underscores the ability of machine learning algorithms to enhance the accuracy of model-simulated surface pCO2 outputs.https://doi.org/10.1038/s41597-025-04914-z
spellingShingle Prasanna Kanti Ghoshal
A.P. Joshi
Kunal Chakraborty
An improved long-term high-resolution surface pCO2 data product for the Indian Ocean using machine learning
Scientific Data
title An improved long-term high-resolution surface pCO2 data product for the Indian Ocean using machine learning
title_full An improved long-term high-resolution surface pCO2 data product for the Indian Ocean using machine learning
title_fullStr An improved long-term high-resolution surface pCO2 data product for the Indian Ocean using machine learning
title_full_unstemmed An improved long-term high-resolution surface pCO2 data product for the Indian Ocean using machine learning
title_short An improved long-term high-resolution surface pCO2 data product for the Indian Ocean using machine learning
title_sort improved long term high resolution surface pco2 data product for the indian ocean using machine learning
url https://doi.org/10.1038/s41597-025-04914-z
work_keys_str_mv AT prasannakantighoshal animprovedlongtermhighresolutionsurfacepco2dataproductfortheindianoceanusingmachinelearning
AT apjoshi animprovedlongtermhighresolutionsurfacepco2dataproductfortheindianoceanusingmachinelearning
AT kunalchakraborty animprovedlongtermhighresolutionsurfacepco2dataproductfortheindianoceanusingmachinelearning
AT prasannakantighoshal improvedlongtermhighresolutionsurfacepco2dataproductfortheindianoceanusingmachinelearning
AT apjoshi improvedlongtermhighresolutionsurfacepco2dataproductfortheindianoceanusingmachinelearning
AT kunalchakraborty improvedlongtermhighresolutionsurfacepco2dataproductfortheindianoceanusingmachinelearning