Estimating temporal and spatial variation of ocean surface pCO 2 in the North Pacific using a self-organizing map neural network technique
This study uses a neural network technique to produce maps of the partial pressure of oceanic carbon dioxide ( pCO 2sea) in the North Pacific on a 0.25° latitude × 0.25° longitude grid from 2002 to 2008. The pCO 2sea distribution was computed using a self-organizing map (SOM) originally utilized to map the pCO 2sea 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 pCO 2sea distribution and biogeochemistry of the basin. The observed pCO 2sea 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 pCO 2sea values agreed well with the pCO 2sea 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 pCO 2sea estimates could be improved by including SSS as one of the training parameters and by taking into account secular increases of pCO 2sea that have tracked increases in atmospheric CO 2. Estimated pCO 2sea values accurately reproduced pCO 2sea data at several time series locations in the North Pacific. The distributions of pCO 2sea revealed by 7 yr averaged monthly pCO 2sea maps were similar to Lamont-Doherty Earth Observatory pCO 2sea climatology, allowing, however, for a more detailed analysis of biogeochemical conditions. The distributions of pCO 2sea 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.