Working Groups Annual Meeting 2025

Environmental System Science (ESS) Working Groups Annual Meeting

Hybrid Meeting - April 14, 2025

Participation

  • Virtual participation via Zoom.
  • In person participants will be in the main room for overview talks and the group discussions. Breakouts will be in physically separated rooms, with each breakout room connected to a corresponding breakout room in zoom.
  • A variety of online tools, including Google Docs and Jam boards will be used to ensure both in person and virtual participants can contribute effectively to the discussions and outcomes. In person participants are expected to bring a laptop to use online tools.

Agenda

8:00-9:00 Breakfast
9:00-9:10 Welcome, meeting objectives and structure, orientation to goals
9:10-9:20 Cyberinfrastructure Working Groups (CIWG Leads)
  • Current working groups and activities over the past year
    • Data Management
    • Data-Model Integration
    • Software Engineering and Interoperability
    • Computing Infrastructure
9:20-9:30 ESGF Update (Forrest Hoffman)
  • Back-end architecture changes for improved scalability and reliability
  • Front-end changes for better data discovery and access with Python and Jupyter
9:30-10:30 ESS-DIVE Updates (Charuleka Varadharajan and Shreyas Cholia)
  • ESS-DIVE features, services and plans
  • ESS-DIVE partner project updates (5-min flash talks)
    • Kim Ely - ESS-DIVE data curation support
    • Stephanie Pennington - sensor data reporting format
    • Amy Goldman - harmonization of hydrological and water quality reporting formats
    • Zhi Li - Improving Advanced Terrestrial Simulator (ATS) model data management and archiving standards
  • Discussion on proposed large data guidance
  • Discussion on repository scope
10:30-10:45 Break
10:45-12:00 Flash Talks (Terri Velliquette)

Description: Communicating new, exciting CI capabilities and engaging the ESS community in shared CI needs (5 min). Add your questions and comments to the Google Doc. There should be time for a question between talks and discussion at the end.
  1. Benjamin Sulman (ORNL): Coupling a reactive transport simulator with the E3SM Land Model to simulate wetland biogeochemistry
  2. Katrina Bennett (LANL): NGEE Arctic ELM Updates: Snow-Vegetation-Permafrost Interactions
  3. Sergi Molins (LBNL): Modeling solute transport and geochemical processes in watershed hydrology
  4. Tasneem Ahmadullah (PNNL): Understanding Water Column Respiration Using Substrate Explicit Thermodynamic Model
  5. Xingyuan Chen (PNNL): Designing Watershed Testbeds in the Digital Age
  6. Dipankar Dwivedi (LBNL): Toward Transferable Watershed-Scale Models Using Process-Based and Machine Learning Approaches
  7. Yunxiang Chen (PNNL): Advancing river observations with AI-powered cloud computing infrastructure: a demo
  8. Joan Damerow (LBNL): Toward an AI-assisted platform for integrating BER data and simulations
  9. Meredith P. Goins (Univ of TN, Knoxville): World Data System: Repository support and how it aids CI teams
12:00-12:45 Lunch
  • Continue informal collaborative discussions with participants.
    • ESS-DIVE team members will be available
12:45-1:45 BER data integration across BSSD and EESSD (Lee Ann McCue)

Description: We are developing the capability to integrate diverse biological and ecological datasets through a new BER initiative – The Bio-Eco Data Integration. Two BER User facilities (EMSL and JGI) and three BER data resources (ESS-DIVE, KBase and NMDC) are working together to to enable facile data discovery and integration across BER resources, to support interoperability of BER data needed for the span of research involving genomics to (plot-scale) ecosystems, to enable scientific understanding across spatial and temporal scales, and support new science including the application of AI/ML.
  • Speakers:
    • JKjiersten Fagnan: How a BER Data Ecosystem Can Support Your Science and Our Mission
    • John Bargar: MONet Data Integration Use Cases
  • Group Discussion:
    • Beyond the initial API queries that are currently being implemented
    • What do you wish was possible
    • Additional use case ideas
1:45-2:30 What’s possible already … What’s next (David Moulton)

Description: A quick overview of a few examples of what people are already doing in the realm of data processing pipelines, data collections, ensemble simulations using process-based models etc. Followed by a talk and demo, and then group discussion.
  • Speakers:
    • Stijn Wielandt: Distributed environmental sensing using advanced wireless connectivity
    • Roelof Versteeg: Highlights of a data processing workflow from a recent SBIR project
    • Forrest Hoffman: Highlights of the Rapid Evaluation Framework (REF), an automated system that is under development that would evaluate and intercompare model output as soon as it is submitted to ESGF.
  • Group Discussion:
    • Do you have examples to share
    • What do you wish was possible
    • What are some common challenges in the community
2:30-2:45 Break
2:45-4:15 Model / Data - Accelerating ModEx / completing the cycle (Xingyuan Chen, Tim Scheibe)

Description: In recent work we’ve seen there’s significant potential to accelerate ModEx iterations by taking a co-design approach to support hypothesis driven research. We’ll discuss this hand-in-hand approach to Modeling and Experimentation, have three short talks highlighting ways it can work, and then breakout to brainstorm ideas on how we can expand this acceleration across your projects.
  • Speakers (~5 min each)
    • Tim Scheibe: Data driven approaches with AI-based models guiding data collection.
    • Scott Painter: Using ATS’s multiscale transport model to evaluate tracer test designs
    • Sundar Niroula: Site specific simulations studying disturbance impacts with ATS to guide data collection
  • Breakout Session (~1 hour including report back):

    Brainstorm ways to use data-driven models, synthetic and simplified process-based models and preliminary data in process-based models in a co-design iterative approach to guide sampling, build and focus models, and accelerate ModEx.
4:15-5:15 A future vision of Cyberinfrastructure and the CI Working Groups (David Moulton)

Description: Given all that we’ve been discussing, it is clear that the data collection/curation and storage, the integration and workflows, and the software needed to support it is changing and growing. This is particularly true as the additional challenges of more fully embracing the potential of AI/ML to further enhance this acceleration of ModEx and science is true. Here we have a breakout session to brainstorm what the future of the CI Working groups should look like to meet these needs for the community. Key topics for brainstorming the needs and challenges of the community include:
  • Sensor streaming data
  • Computing infrastructure, Edge computing, Networking, Automation (data transfer and processing), containerization
  • AI/ML role/potential
  • Model coupling (IHTMs and LMs), surrogate models, …

We will have an open discussion on these topics based on what we learned throughout the day and particularly in the ModEx breakouts.
5:15 Adjourn
Final Note: Thanks for everyone’s participation, it was a really fun and productive day!