Enter Note Done
Go to previous page in this tab
Session
  • Presentation | B34A: Emerging Machine Learning Approaches for Process Understanding and Predictions in Ecosystem Sciences II Oral
  • Oral
  • Bookmark Icon
  • B34A-02: ALIVE-KGML: A Knowledge-Guided Machine Learning Framework for Coupled Carbon and Water Flux Estimation from Geostationary Satellites
  • Schedule
    Notes
  • Location Icon265-266
    NOLA CC
    Set Timezone
  •  
    View Map

Generic 'disconnected' Message
Author(s):
Sadegh Ranjbar, University of Wisconsin Madison (First Author, Presenting Author)
Danielle Losos, University of Wisconsin
Sophie Hoffman, University of Wisconsin Madison
Qunying Huang, University of Wisconsin Madison
Ankur Desai, University of Wisconsin Madison
Paul Stoy, University of Wisconsin-Madison, Madison


We developed a new system called ALIVE-KGML that uses satellite images, machine learning, and ground measurements from research sites across the U.S. to estimate how much carbon plants take in and how much water they release into the air. What makes this system different is that it doesn’t just rely on data, it also follows basic rules of nature, like how plants grow and use water. By combining smart computer models with real-world data and natural laws, ALIVE-KGML gives reliable and easy-to-understand information about how the land and environment are changing. This can help scientists, policymakers, and others make better decisions about climate and nature.



Scientific Discipline
Neighborhood
Type
Main Session
Discussion