We are a home for Earth science data and computing professionals. Our sessions bring together the community for hands-on, interdisciplinary deep dives as we explore "Innovation to Impact" this year. Learn more about ESIP: esipfed.org
Log in to bookmark your favorites and sync them to your phone or calendar.
ESIP Data Readiness Cluster is exploring to expand the current AI-readiness checklist (https://github.com/esipfed/data-readiness) by better supporting diverse geoscience use cases. The current AI-readiness checklist provides general guidelines for data producers and data managers to evaluate the quality and usability of open environmental data for AI developments. However, different AI use cases in geosciences may have further requirements for datasets that are suitable for efficient and reproducible AI research and development. This session will focus on the development of extensions of AI-readiness checklist for different types of geoscience datasets.
Value to Session Participants: Session participants can contribute to the development and provide feedback to the AI-readiness checklist.
Recommended Ways to Prepare: Getting familiar with the AI-readiness checklist from ESIP Data Readiness Cluster.
I am currently a Research Scientist at North Carolina Institute for Climate Studies, affiliated with NOAA National Centers for Environmental Information. My current research at NCICS focuses on generating a blended near-surface air temperature dataset by integrating in situ measurements... Read More →
Tuesday January 21, 2025 1:30pm - 3:00pm EST
Room 1
We will talk on a range of ideas centering on data, AI models, and AI product life cycles. We will determine what we should do next to help community realize practical, trustworthy, and ethical AI.
Value to Session Participants: Clear the mind on what AI projects should be carried out, participate to draft the community paper, be part of the big effort on navigating AI efforts in Earth Science Data community.
Recommended Ways to Prepare: Read the AI readiness checklist and the meeting notes of machine learning cluster.
Share your trained models as cloud-executable functions with DOIs with the NSF Funded Garden Service. You can bring your models to the session and together we will package them up and publish them as hosted Python functions. The Garden platform makes it easy for other researchers to find, understand, and try out your models.
If you don’t have a model to share, feel free to stop by to learn how to use Garden to help make your research more FAIR. Value to Session Participants: Researchers will be able to turn their trained models into citable research objects and extend the reach of their research by allowing others in the field to try the model out and possibly integrate them into their own work. Researchers will learn methods to most simply share their models and make them reusable by the community.
Recommended Ways to Prepare: Gather up your trained models Obtain a Hugging Face account
This session explores AI applications in geoscience, focusing on automated data analysis workflows and deposit type classification techniques using NSAI guided by KGs. We will demonstrate how AI agents leverage open LLMs to perform mineral data analysis and showcase a neuro-symbolic approach that embeds knowledge graphs of deposit rules within neural networks to enhance mineral deposit classification accuracy.
Value to Session Participants: Participants will gain insights into scalable AI-driven workflows and the integration of NSAI guided by KGs to enhance geochemical-based deposit type classification, exploring practical applications for automating data analysis and enhancing classification in geoscience and beyond.
Recommended Ways to Prepare: We will provide a Jupyter Notebook demo for our audience to interact with.
Introduce Taylor Geospatial Institute (TGI) to the ESIP community and the initiatives TGI is running that are open to all researchers, government, industry, and academic institutions. These include the TGI Geospatial Innovation for Food Security Challenge that is open to all to participate in; the TGI monthly GeoAI working group that discussing running the Clay foundation model for others to develop applications on top and eventually the NASA / IBM foundational model within a cloud environment. As well as the TGI Spatial Humanities working group that meets regularly and always has a guest speaker and the AWS $1M credit challenge that will already be underway but we expect to build on this challenge to have more in the future. With a desire to support the ESIP federation and its members, TGI is inviting organizations and individuals to collaborate on initiatives and projects to develop, advance, and implement geospatial capabilities across domains, disciplines, and borders. Value to Session Participants: Understanding the organizations and members objectives and engaging with them to discover where we can support one another.
Recommended Ways to Prepare: No preparation required
