Drivers of Atmospheric and Oceanic Surface Temperature Variance: A Frequency Domain Approach

Abstract

Ocean-atmosphere coupling modifies the variability of Earth’s climate over a wide range of timescales. However, attribution of the processes that generate this variability remains an outstanding problem. In this manuscript, air-sea coupling is investigated in an eddy-resolving, medium-complexity, idealized, ocean-atmosphere model. The model is run in three configurations: fully coupled, partially coupled (where the effect of the ocean geostrophic velocity on the sea surface temperature field is minimal), and atmosphere-only. A surface boundary layer temperature variance budget analysis computed in the frequency domain is shown to be a powerful tool for studying air-sea interactions, as it differentiates the relative contributions to the variability in the temperature field from each process across a range of timescales (from daily to multidecadal). This method compares terms in the ocean and atmosphere across the different model configurations to infer the underlying mechanisms driving temperature variability. Horizontal advection plays a dominant role in driving temperature variance in both the ocean and atmosphere, particularly at timescales shorter than annual. At longer timescales, the temperature variance is dominated by strong coupling between atmosphere and ocean. Furthermore, the Ekman transport contribution to the ocean’s horizontal advection is found to underlie the low-frequency behavior in the atmosphere. The ocean geostrophic eddy field is an important driver of ocean variability across all frequencies and is reflected in the atmospheric variability in the western boundary current separation region at longer timescales.

Role of Ocean and Atmosphere Variability in Scale‐Dependent Thermodynamic Air‐Sea Interactions

“…If all spatial scales are retained in the spectral analyses, the obtained results indicate that, in HR, ocean forcing prominently enhances the extratropical SST variability over periods shorter than about 500 days while atmospheric forcing dominates the variability at longer timescales, results compatible with the findings of Patrizio and Thompson (2022) and Martin et al. (2021). More specifically, Patrizio and Thompson (2022) employed a stochastic climate model formulation similar to the one used here, further accounting for a feedback term associated with ocean dynamics computed as the 1‐month lag‐regression coefficient between OHFC and SST, and found that ocean processes significantly enhanced the temperature variability over periods shorter than about 2 years.…”

“…Both Patrizio and Thompson (2022) and Martin et al. (2021) found that, although ocean processes also contribute to the upper‐ocean temperature variability at lower frequencies, their relative importance becomes secondary to that of atmospheric processes.…”

“…Within the framework of FCL98, atmospheric processes solely modulates THF through stochastic variability in near‐surface atmospheric temperature. This approach neglects various mechanisms through which internal atmospheric processes can affect SST and THF, including: The role of the atmosphere in generating Ekman currents, which are significant contributors to OHFC and consequently to the SST variability at large spatial scales (Larson et al., 2018; Small et al., 2020) and at long time‐scales (Martin et al., 2021). This implies that using OHFC to represent the ocean noise forcing overestimates the role of ocean processes for driving the SST and THF variability at these scales. The influence of stochastic fluctuations in surface wind speed, which drives significant fractions of the THF variability (e.g., Alexander & Penland, 1996; Frankignoul et al., 1998; Lee et al., 2008; Proistosescu et al., 2018; Sura et al., 2006).…”

“…Fully‐coupled simulations show similar results, and further indicate that the THF variability is also enhanced (Chang et al., 2020; Small et al., 2020). The importance of ocean phenomena on driving changes in SST and THF over intraseasonal to interannual timescales was also revealed in an eddy‐resolving coupled simulation of an idealized western boundary current system (Martin et al., 2021) and by the Estimating the Circulation and Climate of the Ocean (ECCO) ocean state estimate in the Gulf Stream and subpolar North Atlantic (Buckley et al., 2014, 2015) and in the extratropics including oceanic regions away from strong currents (Patrizio & Thompson, 2021, 2022).…”

“…The study by Martin et al ( 2021 ), based on a frequency-space temperature variance budget analysis of an idealized, high-resolution air-sea coupled simulation of a western boundary current system analogous to the Gulf Stream, found that the ocean dynamics prominently contributes to the upper-ocean temperature variability over annual timescales and shorter. Both Patrizio and Thompson ( 2022 ) and Martin et al ( 2021 ) found that, although ocean processes also contribute to the upper-ocean temperature variability at lower frequencies, their relative importance becomes secondary to that of atmospheric processes.…”