Search

This study examines the impact of ocean advection and surface freshwater flux on the non‐seasonal, upper‐ocean salinity variability in two climate model simulations with eddy‐resolving and eddy‐parameterized ocean components (HR and LR, respectively). We assess the realism of each simulation by comparing their sea surface salinity (SSS) variance with satellite and Argo float estimates. In the extratropics, the HR variance is about five times larger than that in LR and agrees with Argo. In turn, the extratropical satellite SSS variance is smaller than that from HR and Argo by about a factor of two, potentially caused by the insufficient resolution of radiometers to capture mesoscale features and their low sensitivity to SSS in cold waters. Using a simplified salinity conservation equation for the upper‐50‐m ocean, we find that the advection‐driven variance in HR is, on average, 10 times larger than the surface flux‐driven variance, reflecting the action of mesoscale processes. Plain Language Summary: This study explores the importance of ocean currents, evaporation, and rainfall for driving changes in the salt concentration in the upper ocean (known as salinity) in two climate model simulations with differing ocean resolutions. The high‐resolution model (HR) simulates ocean currents with dimensions of tens of km, while the low‐resolution model (LR) can only simulate currents with hundreds of km in size. When comparing their simulated sea surface salinity variations with those captured by satellites and autonomous floats from the Argo array, the salinity variability in the high‐resolution model is similar to the Argo data at mid to high latitudes and about five times stronger than that in the low‐resolution model. The satellite data show a variability two times smaller than HR and Argo in the same regions, potentially due to their insufficient spatial resolution at higher latitudes and their low sensitivity to the surface salinity in cold waters. Using a simple equation describing the conservation of salinity in the upper ocean, we have shown that small‐scale ocean currents drive most of the salinity variability in HR, while in LR, ocean currents play a much smaller role. Key Points: We investigate how advection and surface flux affect upper‐50‐m salinity variance in eddy‐resolving and eddy‐parameterized climate modelsThe extratropical variance in the eddy‐resolving run matches Argo and is much larger than in the eddy‐parameterized run and satellite dataThe larger upper‐ocean salinity variance in the eddy‐resolving run is predominantly driven by mesoscale ocean processes [ABSTRACT FROM AUTHOR]

A substantial role for atmospheric noise in simulations of decadal internal variability of SST is demonstrated by a comparison of a multicentury climate model control and a corresponding interactive ensemble (IE) simulation. The IE is designed to reduce atmospheric noise in the heat flux, wind stress, and freshwater flux at the air–sea interface. This comparison suggests that nearly all SST variability on decadal time scales is forced by internal atmospheric variability. The results are examined to determine the relative roles of atmospheric surface heat flux noise and ocean dynamics in the decadal volume-averaged heat budget of the upper ocean. The regional heat budgets in two regions, the South Pacific and the midlatitude North Atlantic, show the net atmospheric surface heat flux to be approximately in equilibrium with the ocean dynamics forcing. The IE and control results are used in the equilibrium heat budget approximation to infer the atmospheric heat flux response to SST, as well as the time series of the control atmospheric noise surface heat flux and ocean dynamics forcings for several regions. The South Pacific region SST is found to be primarily forced by the atmospheric noise surface heat flux and the North Atlantic region SST is forced by the ocean dynamics. Similar strengths for the atmospheric heat flux noise and ocean dynamics forcing, with an interdecadal atmospheric heat flux noise time scale and a centennial ocean dynamics time scale, are found for an Atlantic multidecadal variability region SST. [ABSTRACT FROM AUTHOR]

Quasi-decadal climate of the Kuroshio Extension (KE) is pivotal to understanding the North Pacific coupled ocean–atmosphere dynamics and their predictability. Recent observational studies suggest that extratropical-tropical coupling between the KE and the central tropical Pacific El Niño Southern Oscillation (CP-ENSO) leads to the observed preferred decadal time-scale of Pacific climate variability. By combining reanalysis data with numerical simulations from a high-resolution climate model and a linear inverse model (LIM), we confirm that KE and CP-ENSO dynamics are linked through extratropical-tropical teleconnections. Specifically, the atmospheric response to the KE excites Meridional Modes that energize the CP-ENSO (extratropicstropics), and in turn, CP-ENSO teleconnections energize the extratropical atmospheric forcing of the KE (tropicsextratropics). However, both observations and the model show that the KE/CP-ENSO coupling is non-stationary and has intensified in recent decades after the mid-1980. Given the short length of the observational and climate model record, it is difficult to attribute this shift to anthropogenic forcing. However, using a large-ensemble of the LIM we show that the intensification in the KE/CP-ENSO coupling after the mid-1980 is significant and linked to changes in the KE atmospheric downstream response, which exhibit a stronger imprint on the subtropical winds that excite the Pacific Meridional modes and CP-ENSO. [ABSTRACT FROM AUTHOR]