Linking Antarctic Atmospheric River Characteristics with Their Landfalling Impacts

Linking Antarctic Atmospheric River Characteristics with Their Landfalling Impacts James Butler1, Michelle Maclennan2, Fernando Pérez1, Jon McAuliffe1,3 1UC Berkeley, Department of Statistics; 2British Antarctic Survey; 3The Voleon Group AGU25 Session A12D: Atmospheric Rivers: Processes, Impacts, Observations, and Uncertainties II Oral What are Antarctic Atmospheric Rivers (ARs)? Long, narrow filaments of intense poleward moisture transport AR landfalls relatively rare (compared to elsewhere on the globe) Credit: NASA But they have tremendous impacts! But they have tremendous impacts! Precipitation 10-20% of annual snowfall budget (Wille et al. 2021) But they have tremendous impacts! Precipitation 10-20% of annual snowfall budget (Wille et al. 2021) But they have tremendous impacts! Precipitation Temperature Warm midlatitude air associated with… 10-20% of annual snowfall budget (Wille et al. 2021) But they have tremendous impacts! Precipitation Temperature Credit: NASA Warm midlatitude air associated with… Collapse of Larsen B (Wille et al. 2022) 10-20% of annual snowfall budget (Wille et al. 2021) But they have tremendous impacts! Precipitation Temperature Credit: NASA Warm midlatitude air associated with… Collapse of Larsen B (Wille et al. 2022) 40°C temp. anomalies (Wille et al. 2024) Blanchard-Wrigglesworth et al. (2023) Commonly studied via Eulerian AR Detection Tool 1980 vIVT Threshold Catalog (Wille et al. 2021) lat. lon. AR pixel 2022 Impacts and Characteristics of Interest Impact Variables Cumulative snowfall (Gt) 2m Temp. Anomaly (°C) For each of 3101 landfalling ARs from 1980-2022 Impacts and Characteristics of Interest LH integrated water vapor (kg/m2) poleward 850 hPa wind (m/s) SLP gradient (Pa/km) max southern extent (° lat) cumulative landfalling area (km2 day) omega (Pa/s) Impact Variables Characteristic Variables Cumulative snowfall (Gt) 2m Temp. Anomaly (°C) For each of 3101 landfalling ARs from 1980-2022 To get started… To get started… Wetter storms produce more snowfall Windier storms produce more snowfall To get started… Wetter storms produce more snowfall More variables and their interactions? Windier storms produce more snowfall To get started… Wetter storms produce more snowfall More variables and their interactions? Typical vs. rare events? Windier storms produce more snowfall A Statistical Modelling Approach AR impact Vector of 6 AR characteristics A Statistical Modelling Approach AR impact Vector of 6 AR characteristics Model average and extreme quantile of these distributions A Statistical Modelling Approach AR impact Vector of 6 AR characteristics Model average and extreme quantile of these distributions Learn via xgboost (Chen 2016) Learn via gbex (Velthoen 2023) A Statistical Modelling Approach Interrogate fitted models to see what patterns they learned.. In the meantime… QR code time! Thanks! Tomorrow’s cloud-based workflows session info GitHub repo for dataset construction workflow Session IN23A: Open-Source Geospatial Workflows in the Cloud Academic website References and Acknowledgements Wille, J. D., Favier, V., Gorodetskaya, I. V., Agosta, C., Kittel, C., Beeman, J. C., et al. (2021). Antarctic atmospheric river climatology and precipitation impacts. Journal of Geophysical Research: Atmospheres, 126, e2020JD033788. Wille, J.D., Favier, V., Jourdain, N.C. et al. (2022). Intense atmospheric rivers can weaken ice shelf stability at the Antarctic Peninsula. Commun Earth Environ 3, 90 (2022). Wille, J. D., et al. (2024). The Extraordinary March 2022 East Antarctica “Heat” Wave. Part I: Observations and Meteorological Drivers. J. Climate, 37, 757–778. Blanchard-Wrigglesworth, E., Cox, T., Espinosa, Z. I., & Donohoe, A. (2023). The largest ever recorded heatwave—Characteristics and attribution of the Antarctic heatwave of March 2022. Geophysical Research Letters, 50, e2023GL104910. https://doi.org/10.1029/2023GL104910 Tianqi Chen and Carlos Guestrin. 2016. XGBoost: A Scalable Tree Boosting System. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD '16). Association for Computing Machinery, New York, NY, USA, 785–794. https://doi.org/10.1145/2939672.2939785 Velthoen, J., Dombry, C., Cai, JJ. et al. Gradient boosting for extreme quantile regression. Extremes 26, 639–667 (2023). https://doi.org/10.1007/s10687-023-00473-x Special thanks to all those who helped me prepare this presentation, including Ryan Giordano, Alex Strang, Chris Paciorek, Anthony Ozerov, and Sequoia Andrade!