Estimating temporal and spatial variation of ocean surface <i>p</i>CO₂ in the North Pacific using a self-organizing map neural network technique

Abstract

This study uses a neural network technique to produce maps of the partial pressure of oceanic carbon dioxide (pCO2sea) in the North Pacific on a 0.25° latitude × 0.25° longitude grid from 2002 to 2008. The pCO2sea distribution was computed using a self-organizing map (SOM) originally utilized to map the pCO2sea in the North Atlantic. Four proxy parameters – sea surface temperature (SST), mixed layer depth, chlorophyll a concentration, and sea surface salinity (SSS) – are used during the training phase to enable the network to resolve the nonlinear relationships between the pCO2sea distribution and biogeochemistry of the basin. The observed pCO2sea data were obtained from an extensive dataset generated by the volunteer observation ship program operated by the National Institute for Environmental Studies (NIES). The reconstructed pCO2sea values agreed well with the pCO2sea measurements, with the root-mean-square error ranging from 17.6 μatm (for the NIES dataset used in the SOM) to 20.2 μatm (for independent dataset). We confirmed that the pCO2sea estimates could be improved by including SSS as one of the training parameters and by taking into account secular increases of pCO2sea that have tracked increases in atmospheric CO2. Estimated pCO2sea values accurately reproduced pCO2sea data at several time series locations in the North Pacific. The distributions of pCO2sea revealed by 7 yr averaged monthly pCO2sea maps were similar to Lamont-Doherty Earth Observatory pCO2sea climatology, allowing, however, for a more detailed analysis of biogeochemical conditions. The distributions of pCO2sea anomalies over the North Pacific during the winter clearly showed regional contrasts between El Niño and La Niña years related to changes of SST and vertical mixing.