The deadline to submit a session, town hall, or pre-conference workshop proposal for AGU25 has passed. Please search for and review proposed sessions below. Please note this site is frozen as of 4/29/25 and will not reflect any changes to proposals made after this date.


Conveners will be contacted by the AGU25 Program Committee with any requests for mergers or edits to proposals. Conveners will be notified of the acceptance of their proposal in mid-June.


Abstract submissions will also open in mid-June and the deadline for abstract submissions is 30 July, 2025.


Please visit the AGU25 website for more information or please contact AGU with any questions.

Session Proposal: Advancing biogeochemical cycle modeling with artificial intelligence (AI): Bridging data-driven methods and process-based approaches (AGU25)

B011
Advancing biogeochemical cycle modeling with artificial intelligence (AI): Bridging data-driven methods and process-based approaches

Session ID#: 251955

Session Description:
Formulating closed-form physical parameterizations for complex, heterogeneous biogeochemical processes remains challenging. AI offers promising integration with process-based models. However, biogeochemical data properties, such as multiple scales, small-sized observation samples, and spatial autocorrelation, introduce uncertainty in model evaluation, requiring systematic approaches to evaluate model structure, interpretability, robustness, and data uncertainty.  This session seeks to bring together the latest knowledge and innovations at the intersection of AI and biogeochemistry, with a focus on advancing ecosystem and Earth system models. We invite contributions across a wide range of related topics, including but not limited to: 1. Hybrid modeling approaches combine AI with process-based biogeochemical modeling. 2. AI-assisted parameter optimization for modeling global or regional carbon, nutrient, or other biogeochemical cycles. 3. AI-derived data products for parameterizing, validating, or benchmarking biogeochemical processes.  Our goal is to build a collaborative, interdisciplinary community to enhance predictive understanding of biogeochemical cycles across ecosystems and the Earth system.
Co-Sponsor(s):
  • GC - Global Environmental Change
  • IN - Informatics
Index Terms:

0428 Carbon cycling [BIOGEOSCIENCES]
0430 Computational methods and data processing [BIOGEOSCIENCES]
0469 Nitrogen cycling [BIOGEOSCIENCES]
1942 Machine learning [INFORMATICS]
Primary Convener:  Yang Song, University of Arizona, Department of Hydrology and Atmospheric Sciences, Tucson, United States and forrest M hoffman
Conveners:  Umakant Mishra, Sandia National Laboratories, Livermore, United States, Forrest M. Hoffman, Oak Ridge National Laboratory, Computational Sciences & Engineering Division, Oak Ridge, TN, United States and forrest M hoffman
 
Digital Twins to Enhance Model-Experiment (ModEx) Integration (Invited) (1872562)
Yiqi Luo1, Jian Zhou2 and Ning Wei2, (1)Cornell University, School of Integrative Plant Science, Ithaca, NY, United States, (2)Cornell University, School of Integrative Plant Science, Ithaca, United States
 
A Microbially-Enabled E3SM Land Model Reveals Contrasting Seasonal Responses of ecosystem respiration Across NEON Grassland Sites (1898023)
Chang Gyo Jung, Sagar Gautam and Umakant Mishra, Sandia National Laboratories, Livermore, United States
 
DeepCarbon: A Global Data Product of Upscaled Carbon Fluxes from Knowledge-Guided Deep Learning Models (1912682)
Vipin Kumar, Aleksei Rozanov and Arvind Renganathan, University of Minnesota Twin Cities, Department of Computer Science & Engineering, Minneapolis, United States
 
An Artificial Intelligence-based emulator for western US crop productivity and evapotranspiration in land surface and water resource models: The upper Colorado River basin as a testing case (1938902)
Hongyuan Zhang, University of Arizona, Department of Hydrology and Atmospheric Sciences, Tucson, United States, Tianyi Hu, University of Arizona, Department of Hydrology and Atmospheric sciencs, Tucson, AZ, United States, Patrick Neri, University of Arizona, Department of Hydrology and Atmospheric Sciences, Tucson, AZ, United States and Yang Song, Oak Ridge National Laboratory, Climate Change Science Institute & Environmental Sciences Division, Oak Ridge, United States
 
Assessing the Implication of Mineral Adsorption for Soil Carbon Stock and Emissions Across the CONUS: an integrated A.I. and process-based parameterization. (1941133)
Sabrina Wilson, Organization Not Listed, Washington, DC, United States and Yang Song, University of Arizona, Department of Hydrology and Atmospheric Sciences, Tucson, United States
 
A Process-Guided Machine Learning Approach Improves Stomatal Conductance Prediction and Reveals Its Dynamic Association With Photosynthesis (1941492)
Hongmei Li1, Gang Zhao2, Wenpeng Xie3, Roman Olson4 and Kei Yoshimura1, (1)The University of Tokyo, Kashiwa, Japan, (2)Tokyo Institute of Technology, Tokyo, Japan, (3)The University of Tokyo, Institute of Industrial Science, Tokyo, Japan, (4)The University of Tokyo, Chiba, Japan
 
Weaker CMIP6 Carbon Sink Projections and Enhanced Confidence Level Constrained by Artificial Intelligence (1952207)
Zherong WU1,2, Qing Zhu3, John D Albertson4, Wu Sun5, César Terrer6, Trevor Cambron6, Richard Norby7, William Kolby Smith8, Flavio Lehner9, Jiaming Wen10, Yiqi Luo11, Feng Tao1, Ning Wei1, Peifeng Ma12, Anthony Walker13, Xiangzhong Luo14, Joshua Fan15, Carla Gomes16 and Ying Sun1, (1)Cornell University, School of Integrative Plant Science, Ithaca, United States, (2)The Chinese University of Hong Kong, Institute of Space and Earth Information Science; Department of Geography and Resource Management, Hong Kong, Hong Kong, (3)Lawrence Berkeley National Laboratory, Climate & Ecosystem Sciences Division, Berkeley, CA, United States, (4)Cornell University, Ithaca, NY, United States, (5)Carnegie Institution for Science Stanford, Department of Global Ecology, Stanford, United States, (6)Massachusetts Institute of Technology, Cambridge, United States, (7)Oak Ridge National Laboratory, Environmental Sciences Division and Climate Change Science Institute, Oak Ridge, TN, United States, (8)University of Arizona, School of Natural Resources and the Environment, Tucson, United States, (9)Cornell University, Department of Earth and Atmospheric Sciences, Ithaca, United States, (10)Cornell University, School of Integrative Plant Science, Soil and Crop Sciences Section, Ithaca, United States, (11)University of Oklahoma Norman Campus, Department of Microbiology and Plant Biology, Norman, United States, (12)The Chinese University of Hong Kong, Department of Geography and Resource Management, Shatin, Hong Kong, (13)Oak Ridge National Laboratory, Oak Ridge, TN, United States, (14)Lawrence Berkeley National Lab, Department of Geography, Singapore, Singapore, (15)Cornell University, Department of Computer Science, Ithaca, United States, (16)Cornell University, Ithaca, United States
 
Stage-Specific Biogeochemical Controls on Soil Respiration in Cotton: Insights from Machine Learning and Explainable AI (1976498)
Jing Yang1, Jing Hu2, Cary Daniel McCraine1 and Joby Czarnecki3, (1)Mississippi State University, Geosystems Research Institute, Starkville, United States, (2)University of Central Florida, Department of Civil, Environmental, and Construction Engineering, Orlando, United States, (3)Mississippi State University, Geosystems Research Institute, Mississippi State, United States
 
AI-based spinup for the land component of an Earth system model (1979183)
Peter E Thornton1, Dali Wang2, Daniel M Ricciuto1 and Xiaojuan Yang3, (1)Oak Ridge National Laboratory, Environmental Sciences Division, Oak Ridge, United States, (2)Oak Ridge National Laboratory, Climate Change Science Institute and Environmental Sciences Division, Oak Ridge, TN, United States, (3)Oak Ridge National Lab, Oak Ridge, TN, United States
 
Experimental-data-informed, AI/ML-enabled framework for evaluation and development of land carbon cycle in Earth system models (1980396)
Zheng Shi1, Yuxiang Zhang2, Forrest M Hoffman3 and Jizhong Zhou2, (1)University of Oklahoma, Institute for Environmental Genomics, Department of Microbiology and Plant Biology, Norman, United States, (2)University of Oklahoma, Norman, United States, (3)Oak Ridge National Lab, Oak Ridge, United States
 
Machine Learning-Enhanced Vegetation Demography and High-Resolution Earth System Modeling (Invited) (1986382)
Dr. Lingcheng Li, PhD1, Yilin Fang2, Gautam Bisht3, Peishi Jiang3, Pengfei Shi1, Hoang Tran3, Mingjie Shi1, Zhangshuan Hou1, Zhonghua Zheng4, Marcos Longo5, Jennifer Holm5, Rosie Fisher6, Jeffrey Q Chambers7, Charles Koven8, Nathan McDowell3 and Ruby Leung3, (1)Pacific Northwest National Laboratory, Richland, WA, United States, (2)Battelle, Pacific Northwest National Laboratory, Richland, United States, (3)Pacific Northwest National Laboratory, Richland, United States, (4)The University of Manchester, Department of Earth and Environmental Sciences, Manchester, United Kingdom, (5)Lawrence Berkeley National Laboratory, Berkeley, United States, (6)CICERO Center for International Climate Research, Oslo, Norway, (7)University of California, Berkeley, United States, (8)University of California Berkeley, Energy and Resources Group, Berkeley, United States
See more of: Biogeosciences
<< Previous Session Proposal | Next Session Proposal >>