Develop workflows or solutions for needs identified in 2022 report for NASA Biodiversity & Eco4casting Programs

[trashbirdecology]

Idea

Develop case study solutions for some of the outstanding research needs outlined in the 2022 report, [NASA Biological Diversity and Ecological Forecasting: Current state of knowledge and considerations for the next decade]. (https://cce.nasa.gov/biodiversity/pdf/NASABiodiversityReport2022.pdf). EFI members contributed greatly to the production of this report, including contribution of white papers, and EF is a theme throughout.

For a tl;dr, see Appendix C of the report.

To do

• [ ] pull some of the most relevant topics/questions/challenges from CH5
• [x] pull some of the most relevant topics/questions/challenges from CH7
• [ ] pull from the source list of RQs from white papers (@trashbirdecology)

[cboettig]

Looks really cool!

Things that jump out to me:

• Ch 7 forecasting of essential biological variables (EBVs) -- can we design / specify a forecasting challenge specifically around 1 or more of these? To what extent do the current EFI challenges, NEON data, etc align with any already?

The end of Ch7 has this nice list to "Lead development of the community tools, models, and cyberinfrastructure needed for the next generation of forecasts."

• [ ] Automated repeatable forecast workflows
• [ ] Data products with well quantified uncertainties
• [ ] Data assimilation
• [ ] Iterative model refinement
• [ ] Explicit uncertainty accounting
• [ ] A shared clearinghouse for modeling, dissemination, and visualization tools
• [ ] Standards (e.g., NASA should require its funded projects to meet FAIR principles)
• [ ] Archiving of and shared access to forecast outputs and models/workflows
• [ ] Coordination between NASA’s Research, Applied Sciences, and ESTO programs for synergistic calls for proposals to facilitate these

Most of these (excepting the last one) elements are addressed in EFI cyberinfrastructure, though it is not well documented and in some cases could use a lot of improvement still.

[trashbirdecology]

Regardless of what topics we choose to explore during unconf, I will come to the meeting with the last bullet in mind.

Coordination between NASA’s Research, Applied Sciences, and ESTO programs for synergistic calls for proposals to facilitate these

[mdietze]

FWIW I think a number of the forecasting challenges already map onto EBVs:

• terrestrial & aquatic: Ecosystem functioning > Primary productivity
• phenology: Ecosystem functioning > phenology
• tick: Species populations > Species abundances
• beetles: Community composition > Community abundance, Taxonomic/phylogenetic diversity

On the terrestrial challenge, I think that could easily be extended to Ecosystem structure for some models, with either inventory plot or NEON AOP data serving as the target (unfortunately both only have an annual temporal resolution)

In another thread, there's also discussion of a Ecosystem functioning > Ecosystem disturbances challenge

Not currently part of the NEON challenge, but based on our own work on soil microbial forecasting I think that would hit a lot of the Community composition and Genetic composition EBVs. But there's also huge latency in the data and the pipeline for working with it is nontrivial