Expanded record of air-sea CO2 fluxes created using machine learning techniques improves our ability to model annual changes in the ocean carbon sink.

New study highlights the application of Neural Network Ensembles in detecting relationships within Earth system models.

A new interior ocean carbon relevant data product: GLobal Ocean Data Analysis Project version 2 (GLODAPv2) was just released. CPO's Climate Observation Division supported this effort. 

Research funded by CPO’s Atmospheric Chemistry, Carbon Cycle, and Climate (AC4) program was published in Nature Geosciences.  The article assesses the relationship between the organic carbon content of sea water and freshly emitted sea spray aerosol in the North Atlantic as well as the coastal waters of California.