Optimal interpolation of global dissolved oxygen: 1965–2015

Abstract Oxygen inventory of the global ocean has declined in recent decades potentially due to warming‐induced reduction solubility as well circulation and biogeochemical changes associated with warming increasing stratification. Earth System Models predict continued oxygen decline for this century profound impacts on marine ecosystem fisheries. Observational constraint rate loss is crucial assessing ability models accurately simulate these changes. There are only a few observational assessments oceanic reporting range loss. This study develops gridded data set dissolved oceans using optimal interpolation method. The resulting product includes full‐depth map 5‐year moving average from 1965 2015 uncertainty estimates. can come unresolved small‐scale high‐frequency variability mapping errors. multi‐decadal trend −281 −373 Tmol/decade. estimate more conservative than previous works. In study, grid points far observations essentially equal zero anomaly climatology. Calculating approach produces relatively estimate; thus, results likely provide useful lower bound