A012
Advancing AI and Machine Learning for Improved Subseasonal-to-Seasonal (S2S) Forecast Skill
Session ID#: 252232
Session Description:
Timely and skillful subseasonal-to-seasonal (S2S) predictions are a critical tool for reducing economic risk and improving decision making across sectors such as energy, agriculture, and transportation. Additionally, industries rely on dependable S2S guidance to inform decisions on supply chains, logistics, and resource allocation. Artificial intelligence (AI) and machine learning (ML) are increasingly being explored to improve forecast skill by correcting systematic biases, enhancing model calibration, and sharpening representations of key drivers like the Madden-Julian Oscillation (MJO) and El Niño-Southern Oscillation (ENSO). This session invites contributions that apply AI/ML to bias correction, hybrid modeling, and process diagnostics—especially where methods demonstrate operational potential or measurable gains in skill. We also welcome work integrating AI tools with traditional modeling frameworks to improve efficiency and decision support. We particularly encourage submissions that show how AI-driven innovations can strengthen model reliability, reduce costs, and support the delivery of actionable forecasts across public and private sectors.
Index Terms:
1622 Earth system modeling [GLOBAL CHANGE]
3238 Prediction [MATHEMATICAL GEOPHYSICS]
Primary Convener: Christine Bassett, FedWriters, Fairfax, United States; NOAA, Weather Program Office, Silver Spring, United States
Conveners: Mark A Olsen, NASA GSFC, Greenbelt, MD, United States, Nachiketa Acharya, Pennsylvania State University, Center for Earth System Modeling, Analysis, and Data (ESMAD), University Park, United States and Johnna Infanti, Climate Prediction Center College Park, College Park, MD, United States
Student/Early Career Convener: Marybeth Arcodia, Colorado State University, Fort Collins, United States
Summer Greenland Blocking Index Prediction Using Nonlinear System Identification and Interpretable Machine Learning Methods (1862810)
Hua-Liang Wei, University of Sheffield, Dept Automatic Control & Systems Engineering, School of Electrical and Electronic Engineering, Sheffield, United Kingdom, Yiming Sun, University of Sheffield, Dept Automatic Control and Systems Engineering, School of Electrical and Electronic Engineering, Sheffield, United Kingdom, Edward Hanna, University of Lincoln, Department of Geography, School of Natural Sciences, Lincoln, United Kingdom and Linh Luu, University of Lincoln, Department of Geography, School of Nature Sciences, Lincoln, United Kingdom
A Hybrid Dynamical-Machine Learning Forecast Tool for Subseasonal Precipitation Prediction Based on ENSO and the MJO (1874297)
Cheng Zheng1, Hyemi Kim1,2, Emerson LaJoie3, Shan He4 and Edmund K M Chang4, (1)Stony Brook University, School of Marine and Atmospheric Sciences, Stony Brook, United States, (2)Ewha Womans University, Department of Science Education, Seoul, South Korea, (3)Climate Prediction Center College Park, College Park, MD, United States, (4)Stony Brook University, School of Marine and Atmospheric Sciences, Stony Brook, NY, United States
Analysis of the feasibility of deep learning models for subseasonal solar and wind forecasting based on time-series patterns (1877113)
Myeongchan Oh1, Donggun Oh2, Minkyu Lee2 and Chang Ki Kim2, (1)Korea Institute of Energy Research, Renewable Energy Big Data Laboratory, Deajeon, South Korea, (2)Korea Institute of Energy Research, Renewable Energy Big Data Laboratory, Daejeon, South Korea
Benchmarking Machine Learning Models for Subseasonal Temperature Forecasting Across the Contiguous U.S. (1895587)
Maike Holthuijzen1, Thomas Ehrmann2, Meredith G. L. Brown3, Katherine Goode1 and Jacob Johnson1, (1)Sandia National Laboratories, Albuquerque, United States, (2)Sandia National Laboratories, Albuquerque, NM, United States, (3)University of Maryland College Park, College Park, United States
Advancement in NOAA’s DynAmic Forecasting of Air Composition using Optimized Machine Learning (DAFCOM) approach (1900644)
Dr. Beiming Tang1, Barry Baker1, Youhua Tang1, Patrick C Campbell2, Tianfeng Chai1 and Daniel Tong3, (1)NOAA Air Resources Laboratory, College Park, United States, (2)North Carolina State University at Raleigh, Raleigh, United States, (3)George Mason University, Department of Atmospheric, Oceanic & Earth Sciences, Fairfax, United States
Using data science tools to investigate S2S predictability of near-coastal sea surface height (1901407)
Laura Thapa, Colorado State University, Atmospheric Sciences, Fort Collins, United States, Marybeth Arcodia, University of Miami, Miami, United States, Elizabeth Barnes, Boston University, Computing and Data Sciences, Earth and Environment, Boston, United States; Colorado State University, Atmospheric Science, Fort Collins, United States and Dillon J Amaya, NOAA Physical Sciences Laboratory, Boulder, United States
Deep Learning-Based Postprocessing of East Asian Summer Precipitation Forecasts for S2S Prediction Using Multiple Architectures, Loss Functions, and Ensemble Means (1906752)
Jieun Wie, Jeonbuk National University, Jeonju, South Korea, Dongjin Kim, Jeonbuk NAtional University, Jeonju, South Korea, Yejin Lee, Jeonbuk National University, Jeonju, Jeonbuk State, SOUTH KOREA and Jong-Min Yeom, Jeonbuk National University, Jeonsu-si, South Korea
HiResCastNet: A 3D Deep Learning Surrogate Model for Probabilistic Regional High-Resolution Subseasonal Forecast Through Downscaling. (1908956)
Sofien Resifi1, Thang Manh Luong2, Elissar AL Aawar2 and Hoteit Ibrahim3, (1)King Abdullah University of Science and Technology, Physical Sciences and Engineering, Thuwal, Saudi Arabia, (2)King Abdullah University of Science and Technology, Physical Sciences and Engineering, Jeddah, Saudi Arabia, (3)King Abdullah University of Science and Technology, Thuwal, Saudi Arabia
Deep learning based short to sub-seasonal streamflow forecast for the flood-prone regions in the United States. (1932109)
Anukesh Krishnankutty Krishnankutty Ambika, Oak Ridge National Laboratory, Earth Science, Oak Ridge, TN, United States, Kshitij Tayal, University of Minnesota Twin Cities, Department of Computer Science/Engineering, Minneapolis, United States, Vimal Mishra, Indian Institute of Technology Gandhinagar, Department of Civil Engineering, Gandhinagar, India and Dan Lu, Oak Ridge National Laboratory, Oak Ridge, United States
Can AI Data-Driven Models Capture the Temporal Clustering of Unique Atmospheric Rivers and the Influence of Climate Modes? (1938831)
Zhiqi Yang1, Michael J Deflorio2, Agniv Sengupta3, Bin Guan4, Zhenhai Zhang5, Kristen Guirguis1, Alexander Gershunov6 and Luca Delle Monache3, (1)University of California San Diego, Scripps Institution of Oceanography, Center for Western Weather and Water Extremes (CW3E), La Jolla, United States, (2)Scripps Institution of Oceanography, UC San Diego, Center for Western Weather and Water Extremes (CW3E), La Jolla, United States, (3)Center for Western Weather and Water Extremes (CW3E), Scripps Institution of Oceanography, UC San Diego, La Jolla, United States, (4)University of California, Los Angeles, Joint Institute for Regional Earth System Science and Engineering, Los Angeles, United States, (5)Scripps Institution of Oceanography, UCSD, Center for Western Weather and Water Extremes (CW3E), La Jolla, United States, (6)Scripps Institution of Oceanography, University of California San Diego, Center for Western Weather and Water Extremes (CW3E), La Jolla, United States
NIVA: An Earth System Foundation Model Used for Subseasonal to Interannual Predictions (1955757)
Aodhan John Sweeney1, Kalai Ramea2, Anisha Pal2, Kyle Heyblom2, Hansi Singh2 and Anisha Pal, (1)Planette AI, Washington, United States, (2)Planette AI, San Francisco, United States
Human-centered operational benchmarking of AI weather models for subseasonal monsoon onset forecasts in India (1962102)
Rajat Masiwal1, Mayank Gupta2, Adam Marchakitus1, Colin Aitken2, Y Qiang Sun1, Tyler Yang3, Hamid Pahlavan4, Katie Kowal5, Amir Jina6, William R Boos7 and Pedram Hassanzadeh8, (1)University of Chicago, Department of the Geophysical Sciences, Chicago, United States, (2)University of Chicago, Development Innovation Lab, Chicago, United States, (3)University of California Berkeley, Berkeley, United States, (4)Rice University, Department of Earth, Environmental, and Planetary Sciences, Houston, TX, United States, (5)University of Chicago, Data Science Institute, Chicago, United States, (6)University of Chicago, Harris School of Public Policy, Chicago, United States, (7)University of California Berkeley, Department of Earth and Planetary Science, Berkeley, CA, United States, (8)The University of Chicago, Department of the Geophysical Sciences, Chicago, United States
Supermodelling towards improved seasonal prediction (1964083)
Noel S Keenlyside1, Francois Counillon2, Tarkeshwar Singh3, Ping-Gin Chiu4 and Lennard Montag3, (1)University of Bergen, Geophysical Institute, Bergen, Norway, (2)Nansen Environmental and Remote Sensing Centre; Bjerknes Centre for Climate Research, Bergen, Norway, (3)Nansen Environmental and Remote Sensing Center, Bergen, Norway, (4)University of Bergen, Bergen, Norway
Does DL Downscaling Affect S2S Forecast Performance? (1980444)
Alejandro Valencia1, Tyler Mitchell Fenske2, Benjamin P Kirtman3 and Samantha J Kramer1, (1)Sonoma Technology, Inc., Petaluma, CA, United States, (2)Sonoma Technology, Inc., Petaluma, United States, (3)University of Miami - RSMAS, Atmospheric Sciences, Miami, FL, United States
Sub-seasonal to seasonal (S2S) forecast of North American weather regimes and associated precipitation extremes in the U.S. using artificial intelligence. (2003453)
Swatah Snigdha Borkotoky1, Kathleen A. Schiro2, Kevin M Grise1 and Antonios Mamalakis3, (1)University of Virginia, Charlottesville, VA, United States, (2)University of Virginia, Department of Environmental Sciences, Charlottesville, United States, (3)University of Virginia, Department of Environmental Sciences, Charlottesville, VA, United States
