New study utilizes machine learning to improve ocean pCO2 products for model evaluation

  • 18 February 2022
New study utilizes machine learning to improve ocean pCO2 products for model evaluation

The ocean plays a significant role in absorbing carbon dioxide from the atmosphere; however, determining how much carbon the ocean absorbs—and when, and where—is challenging since ocean observations to measure this are limited. To address this challenge researchers have used cutting edge machine learning algorithms to fill in observational pCO2 gaps, or used models to represent ocean processes. 

A new study, funded in part by the Climate Observations and Monitoring program, and led by a team from Lamont Doherty Earth Observatory (LDEO) Columbia University (PI: McKinley), recently published a study that utilizes machine learning to merge these observation-based and model-based methods to improve the quantification of ocean carbon uptake. Lead author L. Gloege highlights this new LDEO-HPD pCO2 product and demonstrates that it agrees better with independent data than currently available products. The new product can be used as a valuable diagnostic and visualization tool to evaluate spatio-temporal model fields.

Access the study in the Journal of Advances in Modeling Earth Systems »

For more information, contact Virginia Selz.

Print

x

CPO HEADQUARTERS

1315 East-West Highway Suite 100
Silver Spring, MD 20910

ABOUT US

Americans’ health, security and economic wellbeing are tied to climate and weather. Every day, we see communities grappling with environmental challenges due to unusual or extreme events related to climate and weather.