Atmospheric rivers (ARs), characterized by long filaments of moisture extending thousands of kilometers from the tropics, are responsible for a substantial portion of meridional water vapor transport. They play a crucial role in regional precipitation patterns, particularly in coastal regions such as California, where they contribute significantly to annual rainfall and are frequently associated with major floods and considerable economic damage. While predictors of AR activity like El Niño Southern Oscillation (ENSO) and Madden-Julian Oscillation (MJO) have been identified, the predictability of ARs remains limited, typically not exceeding 10–20 days. This limited predictability is likely influenced by air-sea interactions and feedbacks resulting from ARs' modification of local sea surface boundary conditions. ARs are characterized by extreme conditions including strong winds, high moisture content, intense precipitation, and warm air temperatures. These conditions perturb the upper ocean, affecting key variables such as sea surface temperature (SST), sea surface height, sea surface currents, and mixed layer depth (MLD). Previous research has explored the general impact of ARs on SST, indicating the importance of ocean entrainment in SST cooling downstream of the AR center and the potential for air-sea interaction to significantly impact AR predictions on synoptic timescales. However, a comprehensive quantification of the SST response to ARs, considering individual atmospheric and oceanic processes and their spatial dependency, has been lacking. This study aims to address this gap by employing an ocean state estimate and an SST budget equation to diagnose the specific physical processes that govern the SST tendency response to ARs.

Existing research on atmospheric rivers (ARs) has focused on understanding their formation, characteristics, and impacts. Studies have identified predictors of AR activity, such as ENSO and MJO, but the predictability of ARs remains limited to 10-20 days. The role of air-sea interactions in influencing AR predictability has been highlighted, with research suggesting that using interactive ocean models improves prediction skill for integrated water vapor (IWV) and integrated vapor transport (IVT). However, previous studies have not fully quantified the response of SST to ARs in terms of each atmospheric and oceanic process, nor have they identified the spatial dependency of the response, leaving a knowledge gap.

This study focuses on the October-March period when ARs are most active in the North Pacific. The researchers analyzed 25 years (1993-2017) of ocean reanalysis data from Estimating the Circulation and Climate of the Ocean Version 4 release 4 (ECCOv4) and atmospheric data from ECMWF Reanalysis Interim (ERA-Interim). The analysis utilized an SST budget equation to decompose the local SST tendency into contributions from various atmospheric and oceanic processes. These processes include shortwave radiation, longwave radiation, sensible heat flux, latent heat flux, dilution effect (rainfall impacting SST), residual velocity advection, vertical mixing (diffusion and entrainment), detrainment, and horizontal diffusion. The terms were grouped into contributions from surface forcing and ocean response. The ARs were identified using an algorithm based on integrated vapor transport (IVT), and compositing techniques were employed to analyze the SST response to ARs. The spatial distribution of the SST tendency response was examined, considering the role of various ocean processes across different regions of the North Pacific.

The analysis revealed a significant role for ocean dynamics in modulating the SST response to ARs. In the region of strong ocean modification, ocean dynamics offset over 100% of the anomalous SST warming that would otherwise occur from atmospheric forcing. Ageostrophic advection and vertical mixing were identified as the most important factors in modifying the SST tendency response. The response varied spatially: * **Coastal California:** Enhanced SST warming was driven by a reduction in ageostrophic advection due to anomalous southerly winds. * **Large North Pacific Region:** SST warming resulted from a reduction in total clouds and a consequent increase in incoming shortwave radiation. Further analysis using the SST budget equation revealed complex interactions. When surface forcing led to anomalous SST cooling, the mixed layer deepened, resulting in further cooling through vertical mixing. Conversely, when surface forcing caused anomalous warming, the mixed layer shoaled, yet vertical mixing still resulted in cooling, indicating vigorous turbulent wind work. Anomalous advection produced the strongest cooling when the surface forcing was near zero, primarily due to ageostrophic advection (attributed to Ekman advection). In a strong marine warming region ([15°N, 25°N] × [120°W, 135°W]), high SST weakened the capped inversion associated with stratocumulus clouds, reducing cloud cover and increasing incoming shortwave radiation leading to additional warming. The study also examined the seasonality of the SST response, noting a strong cancellation between surface fluxes and ocean modification in October-November and February-March. The location of strongest warming tracked the latitudinal position of the strongest meridional SST gradient. The strength of vertical mixing varied seasonally, being stronger in fall and weaker in winter and spring.

This study's findings significantly advance our understanding of AR-SST interactions. The substantial offsetting effect of ocean dynamics on surface-driven SST warming highlights the importance of including interactive ocean models in AR prediction. The identification of key processes like ageostrophic advection and vertical mixing provides crucial insights for improving model parameterizations. The spatial heterogeneity in the SST response underscores the need for geographically-specific considerations in AR impact assessments. The strong marine warming region showcases the complex interplay between atmospheric forcing, cloud cover, and SST, underscoring the need for more nuanced understanding of cloud feedback mechanisms. The temporal variability emphasizes the importance of accounting for seasonal changes in MLD and vertical temperature gradients when modeling AR impacts.

This research demonstrates the significant impact of atmospheric rivers on sea surface temperature, highlighting the crucial role of ocean dynamics in modulating the response. The findings emphasize the need for improved models incorporating interactive ocean processes for accurate AR prediction and the importance of spatially-resolved analyses to capture the complexity of AR-SST interactions. Future studies should focus on refining model parameterizations, investigating freshwater forcing effects on stratification, and incorporating higher-resolution data to better resolve eddy-driven interactions.

The study utilizes a relatively coarse resolution (1°) in the ECCOv4 ocean model, limiting the resolution of mesoscale features like eddies. The study does not include feedbacks from the ocean to the atmosphere, limiting the investigation of full two-way coupled processes. The specific mechanisms responsible for the correlation between the ENSO index and the strong marine warming region require further investigation. The study focuses on the North Pacific, and the findings may not be directly generalizable to other regions with different climatic conditions.