Repository logo

What controls the variability of oxygen in the subpolar North Pacific


Dissolved oxygen is a widely observed chemical quantity in the oceans along with temperature and salinity. Changes in the dissolved oxygen have been observed over the world oceans. Observed oxygen in the Ocean Station Papa (OSP, 50°N, 145°W) in the Gulf of Alaska exhibits strong variability over interannual and decadal timescales, however, the mechanisms driving the observed variability are not yet fully understood. Furthermore, irregular sampling frequency and relatively short record length make it difficult to detect a low-frequency variability. Motivated by these observations, we investigate the mechanisms driving the low-frequency variability of oxygen in the subpolar North Pacific. The specific purposes of this study are 1) to evaluate the robustness of the observed low-frequency variability of dissolved oxygen and 2) to determine the mechanisms driving the observed variability using statistical data analysis and numerical simulations. To evaluate the robustness of the low-frequency variability, we conducted spectral analyses on the observed oxygen at OSP. To address the irregular sampling frequency we randomly sub-sampled the raw data to form 500 ensemble members with a regular time interval, and then performed spectral analyses. The resulting power spectrum of oxygen exhibits a robust low-frequency variability and a statistically significant spectral peak is identified at a timescale of 15-20 years. The wintertime oceanic barotropic streamfunction is significantly correlated with the observed oxygen anomaly at OSP with a north-south dipole structure over the North Pacific. We hypothesize that the observed low-frequency variability is primarily driven by the variability of large-scale ocean circulation in the North Pacific. To test this hypothesis, we simulate the three-dimensional distribution of oxygen anomaly between 1952 to 2001 using data-constrained circulation fields. The simulated oxygen anomaly shows an outstanding variability in the Gulf of Alaska, showing that this region is a hotspot of oxygen fluctuation. Anomalous advection acting on the climatological mean oxygen gradient is the source of oxygen variability in this simulation. Empirical Orthogonal Function (EOF) analyses of the simulated oxygen show that the two dominant modes of the oxygen anomaly explains more than 50% of oxygen variance over the North Pacific, that are closely related to the dominant modes of climate variability in the North Pacific (Pacific Decadal Oscillation and North Pacific Oscillation). Our results imply the important link between large-scale climate fluctuations, ocean circulation and biogeochemical tracers in the North Pacific.


Rights Access


decadal variability
North Pacific


Associated Publications