EventsChallenge 01 · AgTech data
One Farm, One Picture
Merge weather, satellite and farm-management data into one clear, trustworthy view a real farmer would use.
- LEAF Open Farm Sunday
- Jun 2027
- data sources to join
- 3+
- all data free to use
- Open
- Countdown
- Sun 13 Jun 2027
01 — The problem
Data-rich, insight-poor on the farm
A single mixed farm generates data from dozens of unconnected sources across one season — tractor telematics, soil and weather sensors, drone and satellite imagery, farm-management software, market prices. Each speaks its own format and rarely talks to the others.
The result: growers are data-rich but insight-poor. The answers to their biggest questions sit hidden across the silos.
01Poor interoperability
No common standard — systems that should share data cannot.
Why this is hard
Each system was designed to be complete on its own terms rather than to hand its data to the next one. Two can both be working correctly and still disagree about where a field ends, when a day starts, or how a yield is counted. Nothing is broken; there is simply no agreed way to line the records up.
02Inconsistent quality
Miscalibrated, duplicated or missing sensor readings erode trust.
Why this is hard
Sensors drift, fail quietly, or record the same event twice from two devices. The number on its own does not say how much it can be trusted, so one doubtful reading can cost confidence in the whole set.
03Patchy connectivity
Rural coverage makes data hard to move and reconcile in the first place.
Why this is hard
Much of this data is generated exactly where coverage is weakest. Readings queue on a device and arrive late, out of order, or not at all — so even well-formed data can reach a system after the moment it was needed.
02 — The question
How might a farmer get a trustworthy answer about their own land, when the data that holds it sits in systems that were never built to talk to each other?
The question deliberately does not say what an answer looks like. A build, an analysis, a design or a shared standard can all answer it — see the four tracks.
03 — The mission
Show, in miniature, what “one farm, one picture” could look like
Build a working prototype that pulls together at least three data sources and turns them into something a real farmer would find useful and understandable.
What good looks like
- 1UsefulnessWould a real farmer actually use it?
- 2ClarityUnderstandable to a non-expert?
- 3IntegrationHow well does it combine separate data?
- 4FeasibilityCould it grow beyond the weekend?
- 5CreativityDid they see what others didn't?
Five criteria, equal weight.
04 — The data
The open data you'll get
All sources verified against the providers' own pages while preparing this programme.
| Dataset | What it is | Licence |
|---|---|---|
| Copernicus / Sentinel | ESA's satellite gateway — crop-health, land cover, flood and land-surface data. | Free · quotas |
| Met Office MIDAS Open | UK weather-station records — temperature, rainfall, sunshine, wind, soil temperature. | OGL · free account |
| Defra data | Soils, water, field boundaries and land cover via the Defra platform on data.gov.uk. | Open Gov Licence |
| AHDB market intelligence | Market prices and sector benchmarking for real-world context. | Free / levy account |
Open data repository on GitHub (opens external site) No account needed to browse
Take part in 01
Sunday 13 June 2027. 100 places, free to enter, open to all ages — some ages may need parental consent. Bring a team or form one on the day — no prior coding needed.