Feedstock Production
LiveBreeding and agronomy for crops grown for energy — what was planted where, how it was managed, and how much biomass it returned.
- Field sites 8
- Plots & subplots 42
- Crops 5
- Sample observations 1,392
CABBI Data Hub
Bioenergy research data, built to be discovered
Field trials, eddy-covariance flux, weather and geospatial coverage from CABBI's research sites — published in one model where every value carries its variable definition, its units, its protocol and the dataset it came from. Structured for a researcher with a question, and for the machines that will help answer it.
8
Field sites
42
Plots, stations & units
22
Datasets
44
Catalogued variables
808,586
Observations
18,700
Flux half-hours
CABBI's research, in data
Where the themes meet
CABBI grows bioenergy crops, converts biomass into valuable chemicals, and measures what that costs the land and the climate. Two of those meet here in one record — the towers and weather stations stand in the same fields as the biomass plots, sharing sites, units, seasons and a single variable catalog, so a yield and a carbon flux can be read against each other rather than reconciled afterwards. The third is kept in the specialist repositories built for it.
Sustainability
LiveWhat these systems do to carbon, water and soil — measured continuously at the tower and the weather station, and mapped across the landscape they sit in.
- Flux half-hours 18,700
- Weather & soil values 807,194
- Raster layers 11
Conversion
Held elsewhereTurning biomass into fuels and bioproducts. These data are kept in domain-specific repositories built for them, rather than in the Data Hub.
- CABBI Pathway Tools / BioCyc Metabolic pathways and genome databases
- LabCollector Laboratory samples, strains and records
Integrating these into the Data Hub is not planned at this time.
6 of 8 sites carry both — continuous measurement and campaign sampling at the same place, on the same units and seasons. That is what makes them readable together, as on Sustainable Advanced Bioeconomy Research Farm →
What you can do here
Find it, see it, take it with you
Find the data
Search and filter 22 datasets by site, crop and variable — or go straight to a plot and see everything ever measured on it.
Browse datasets →See where it comes from
Every site, plot, station and chamber on one map, with real boundaries and click-through to the record behind each one.
Open the map →Read a place end to end
A site profile gathers its continuous measurements, crop seasons, campaign sampling and management history into one page. Start with Sustainable Advanced Bioeconomy Research Farm.
Open a site profile →Take it away
Model-ready CSV from any site, unit or dataset — the same values you see on the chart, with units and provenance columns intact.
Preview and export →Built for machine-readable discovery
Data an agent can read without being told how
Most research data needs a person to explain it before anything can use it — which column is which, what the units were, whether two spreadsheets mean the same thing by "yield". The Hub's job is to carry those answers with the data itself. Some of that is working today; the rest is being built. Here is honestly where each part stands.
A governed variable catalog
Now44 variables defined once, with label, units and domain. Adding a measurement is a catalog row, not a schema migration — so nothing has to guess what a column meant.
Provenance on every value
NowDataset, protocol, instrument and units travel with the number, not in a readme beside it. 10 of 22 datasets already carry a DOI.
One shape for every store
NowDense flux, narrow sensor series and campaign samples resolve to a single long form, so one query pattern reads all of them.
Typed validation & controlled values
In progressRanges, allowed categorical values, conditional requirements and derived traits — so bad data is caught at the door rather than found later.
Crosswalks to standards
In progressTerms can declare their ICASA, BrAPI and AGROVOC equivalents, making the Hub legible to tools that never heard of CABBI.
A JSON API over the same data
PlannedThe second door onto the contract below — stable identifiers over HTTP, for apps and agents that would rather not speak SQL.
For CABBI collaborators
PlannedQuery the data directly, with your own credential
Instead of asking someone to send a spreadsheet, collaborators will connect straight to a read-only replica and query published data in SQL — from a notebook, from a database client, or from an agent working on their behalf.
What that credential sees is a curated data_public schema of
published views, documented inside the database, so column meanings
and units are readable by a person and discoverable by a tool without anyone
having to explain the schema first.
The limits are grants, not guidance: the role cannot write, cannot reach accounts or unpublished data, and long queries time out. A wrong or hostile query hits a wall rather than a policy.
# your own read-only credential, on the replica psql "host=<replica-host> dbname=cabbi sslmode=require" SET search_path = data_public; SELECT site, ts, nee FROM flux WHERE site = 'Energy Farm, Urbana IL' AND ts >= '2022-06-01' LIMIT 100;
Start with a place
8 sites, 42 units, and 808,586 observations published so far.