Ocean fronts and eddies force atmospheric rivers and heavy precipitation in western North America

Atmospheric rivers (ARs), plumes of intense water vapor transport, are major contributors to extreme hydroclimate events in the extratropics, causing torrential rains and floods upon landfall, particularly along the west coast of North America. Accurate prediction of landfalling ARs is crucial for water resource management and disaster mitigation, yet current weather and climate models struggle with this task. While increasing model resolution improves overall AR prediction, the timing and location of landfall, and the associated precipitation impact, remain challenging. This study investigates a potential predictability source: the influence of mid-latitude mesoscale sea surface temperatures (SSTs) associated with ocean fronts and eddies. Previous research focused on tropical variability sources, but the impact of mesoscale SSTs along major ocean fronts like the Kuroshio Extension (KE) and Gulf Stream (GS) on the overlying atmosphere is well-documented, with evidence suggesting their influence extends beyond the atmospheric boundary layer to affect extratropical cyclones (ECs) and storm tracks. Given the close relationship between ARs and ECs, this study explores the hypothesis that mesoscale SSTs can influence ARs, particularly landfalling events and associated heavy precipitation.

Past studies have explored the predictability of ARs focusing on tropical variability sources such as the Madden-Julian Oscillation (MJO) and El Niño-Southern Oscillation (ENSO). However, the potential influence of mid-latitude mesoscale SSTs induced by fronts and ocean eddies on ARs remains largely unexplored despite evidence of their impact on the overlying atmosphere, as shown by high-resolution satellite observations and climate model simulations. Research suggests the influence of mesoscale SSTs can extend beyond the atmospheric boundary layer, affecting extratropical cyclones and mid-latitude storm tracks over large distances. Given the close association between AR occurrence and extratropical cyclones, the question arises whether mesoscale SSTs can influence ARs, particularly landfalling events and associated heavy precipitation.

The study employed two ensembles of twin simulations using the Weather Research and Forecasting (WRF) model with 27-km horizontal resolution. Each ensemble consisted of control runs (CTRL) forced with high-resolution satellite-based SST data and filtered runs (FILT) with the same SST data but spatially filtered to remove mesoscale features. The first ensemble (seasonal ensemble, SE) comprised six-month boreal winter simulations (2002-2014), analyzing the overall impact of mesoscale SST forcing. The second ensemble (cyclone ensemble, CE) consisted of two-week simulations focusing on individual winter cyclone cases to investigate the impact during cyclogenesis. The North Pacific region was chosen due to the high AR-related flood risk along the west coast of North America and the strong mesoscale SST variability generated by the KE front and eddies. The study also included analyses of ERA5 reanalysis data, GPM satellite precipitation data, and PRISM data for validation. To separate the effects of eddy-induced and front-induced mesoscale SST forcing, further experiments were performed, and sensitivity tests were conducted using different planetary boundary layer (PBL) schemes in WRF. Finally, a similar twin experiment using a high-resolution global Community Atmosphere Model (CAM) was conducted to confirm the robustness of the results. ARs were detected and tracked using a widely used approach searching for regions of integrated water vapor transport (IVT) exceeding a threshold. Heavy precipitation events were defined as area-averaged daily precipitation exceeding the 75th percentile.

The study found that including mesoscale SST forcing in the simulations resulted in a significant increase in landfalling ARs and heavy precipitation. The SE experiment showed a ~40% increase in accumulated IVT of landfalling ARs and up to a 30% increase in heavy precipitation over high terrain in CTRL compared to FILT. Analysis of the relationship between mesoscale SST forcing strength and AR/precipitation response showed that stronger mesoscale SST forcing led to stronger landfalling AR and precipitation responses. Supporting evidence from ERA-Interim reanalysis indicated an increase in landfalling AR IVT after 2002 when higher-resolution SST data became available. The CE experiment confirmed the significant increase in landfalling AR IVT and precipitation due to mesoscale SST forcing, even over a short two-week period, highlighting a delayed response with a significant increase in precipitation after 4 days. Further experiments separating eddy-induced and front-induced mesoscale SST forcing revealed that eddy-induced forcing is primarily responsible for the observed increase in landfalling ARs and heavy precipitation. A composite analysis showed a significant increase in AR IVT and precipitation in the warm sector of composite cyclones when mesoscale SSTs were present. Analysis of water vapor profiles showed that the enhanced moisture supply occurs at around 800 hPa, with higher moisture content in CTRL compared to FILT. The same phenomena were observed in the global CAM simulations, increasing the confidence in the model findings. The results suggest that common practices of using non-eddy-resolving monthly SSTs in atmospheric models underestimate AR-induced heavy precipitation.

The findings address the research question by demonstrating a significant remote influence of mesoscale SSTs on landfalling ARs and associated heavy precipitation. The results highlight the importance of including mesoscale SST forcing in prediction models, as neglecting these features leads to underestimation of AR-induced heavy precipitation events. The observed asymmetrical response to warm and cold mesoscale SSTs suggests a mechanism involving enhanced vertical moisture transport from the planetary boundary layer in the presence of warm mesoscale SSTs. This leads to increased moisture supply to developing cyclones, influencing the formation and intensity of ARs. The delayed precipitation response observed in the CE experiment suggests that the influence of mesoscale SSTs can operate on weekly time scales, which has implications for subseasonal-to-seasonal prediction of AR-related heavy precipitation events.

This study demonstrates a significant influence of mesoscale SSTs associated with ocean fronts and eddies on landfalling ARs and heavy precipitation in western North America. The results highlight the importance of incorporating eddy-resolving SST forcing in atmospheric models for improved prediction of ARs and associated extreme weather events. Future research should focus on quantifying the predictability improvement achieved by including mesoscale SST forcing and exploring the potential for extended-range forecasting of ARs.

The study's primary limitation lies in the relatively short time period of the high-resolution datasets used for analysis. The findings are based on simulations and analysis focusing on the North Pacific; the generalizability to other regions with similar oceanographic features needs further investigation. The use of high-resolution modeling also implies computational limitations that restricted the number of simulations and may lead to uncertainty in specific results. The study's focus on the influence of mesoscale SSTs on ARs does not account for other factors such as atmospheric dynamics and land-surface interactions which could also affect AR formation and intensification.