10 / 2023-07-22 19:59:47

Causality of Air–sea Interaction Inferred from Satellite Observations

air-sea interaction,causality analysis

A recently developed causal inference formalism, which is quantitative in nature, and rigorously derived from first principles, is implemented to measure the causality of air–sea interaction, and map out the underlying causal structures based on satellite observations. The formalism is first validated with a classical stochastically forced, coupled energy balance model; it robustly reveals the cause-effect relation of the surface temperature (SST) and air–sea surface heat fluxes (SHFs) in the coupled system. By applying the formalism to the satellite-observed time series of SST and SHFs at a daily temporal resolution and a 0.25° spatial resolution, we identify an overall asymmetry of causality between the two variables in the global ocean, that is, the strength of causality from the SHFs to the SST is significantly larger than that from the SST to the SHFs almost everywhere, even in the western boundary current regions where it has been recently argued that a paradigm of ocean driving atmosphere dominates. We further show that the identified causality is scale dependent. Generally, the causal influence from the SST (SHFs) to the SHFs (SST) decreases (increases) with increasing spatial scale. The amplitudes of the interactions are characterized by strong seasonal and interannual variations. Finally, it is suggested that applications of the novel causal inference formalism to the air–sea interaction would contribute to the improvement of the potential predictability of the coupled ocean-atmosphere system.