How we prepare an EIA: a real report, page by page30 pages of an environmental impact study built on the platform, with a voice that walks you through them.Browse the reportEU project · FORWARDSIn development
Ask the ForestWard Observatory in plain language.
ForestMind is an AI agent that gives natural-language access to the Observatory's forest data. Ask a question about a stand, a region or a species. The agent picks the datasets, reads them through tools and answers with numbers, units, maps and the list of datasets it used.
The ForestMind interface is not public yet. A first version is planned for May 2027, with sign-in. We will link it here when it opens.
Funded by the European UnionWhat you get
Who it is for
No GIS skills and no scripting needed.
A question about a stand, not a script.
Example: “Produce a forest management plan report for my 95 ha stand.”
Screening across a region, on demand.
Example: “Which parts of the region show the largest disturbance footprint?”
Evidence they can read, with sources.
Example: “Which datasets support this figure, and from what year?”
Explore what the Observatory holds without writing queries first.
Example: “What is the dominant tree species within 5 km of Hyytiälä?”
How it works
The ForestWard Observatory, run by the Swedish University of Agricultural Sciences (SLU) inside the FORWARDS project, serves forest data on tree species, biomass, disturbances and field inventories. ForestMind is a conversational layer on top of it.
The chat shown is an illustration with invented numbers. It is not a real answer and not the ForestMind interface.
What the agent does
ForestMind is being built in stages. Here is what each capability is, and where it stands today.
A question in plain language becomes a query on the Observatory. The answer comes back with numbers, units, a map and the datasets used.
A complex question is split into steps, run in parallel and returned as a report: an HTML document today, PDF later, with maps, charts, methodology and data provenance.
Recurring analyses run on a schedule, with alerts that arrive by email when a threshold is crossed.
Target question · Ask
What is the dominant tree species and average biomass within 5 km of the Hyytiälä ICOS super-site?
Example request · AnalysePlanned
Produce a forest management plan report for my 95 ha stand aligned with the EU Forest Strategy 2030.
What it reads
ForestMind reads only what the Observatory publishes. Coverage differs by dataset: some layers cover all of Europe, others one country or one site. The agent says when a question falls outside what is covered.
Species maps across Europe.
Biomass, height, stem volume and other forest attribute maps for Sweden (national forest attribute maps), plus stand age and site index for pine and spruce.
Where and when forests were disturbed.
Measured field plots and individual trees.
Monthly gross primary production for the Krycklan catchment, 2001–2020.
Modelled forest development from 2024 to 2100, on a 50 km grid.
Built on experience
Since 2016, 3Bee has worked with satellite data, IoT sensors and citizen-science records for nature monitoring. ForestMind builds on that experience and applies it to the Observatory's forest data.
How it is built
Figures come from the data through tools, never from the model's memory, and each answer names its datasets.
If data is missing or out of date, the agent is built to say so and give a partial answer with the caveat, rather than a number it cannot back.
Users are told that answers are AI-generated and shown the sources behind them.
The language model runs on an EU-hosted service under a data processing agreement.
ForestMind reads the Observatory through its published APIs and data files, read-only. It changes nothing there and needs nothing installed on it.
Planned for release under Apache 2.0 on GitHub, with the derived datasets on Zenodo (DOIs, CC BY or CC0), by the end of the project.
Availability
The public interface is expected in spring 2027, by May 2027 at the end of the project, together with the open-source release. Until then ForestMind is in development and testing.
The project
Funding
ForestMind (grant G-01-2026-1) is supported by the European Union through FORWARDS (Horizon Europe Project No. 101084481), in grants to third parties managed by the European Forest Institute.
FORWARDSHorizon Europe GA 101084481EFI grant G-01-2026-1
Funded by the European Union. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or the European Research Executive Agency (REA). Neither the European Union nor the granting authority can be held responsible for them.
Part of
ForestMind · FORWARDS
Write to the team at XNatura and we will get back to you.
Your choice on measurement and advertising.
This site needs no consent to work: it runs without profiling. With your consent we also measure how the pages are used, and attribute a request to the campaign that brought you here. You can change your mind whenever you like. Read the cookie policy
