The observation-based pCO₂ fields were created using a 2-step neural network technique. In a first step, the global ocean is divided into 16 biogeochemical provinces using a self organizing map. In a second step, the non-linear relationship between variables known to drive the surface ocean carbon system and gridded observations from the SOCATv2 dataset (Bakker et al. 2014) is reconstructed using a feed-forward neural network within each province separately.

The final product is then produced by projecting these driving variables, i.e., surface temperature, chlorophyll, mixed layer depth, and atmospheric CO₂ onto oceanic pCO₂ using these non-linear relationships. This results in monthly pCO₂ fields at 1°x1° resolution covering the entire globe with the exception of the Arctic Ocean and few marginal seas. The air-sea CO₂ flux is then computed using a standard bulk formula. More details can be found in Landschützer et al. 2013 and Landschützer et al. 2014.

All the related data and detailed description has been published at the CDIAC website:

http://cdiac.ornl.gov/oceans/SPCO2_1982_2011_ETH_SOM_FFN.html