We combine earth observation foundation models with rigorous counterfactual science to build more credible forest carbon baselines.
Belian is the local Sabahan name for the Bornean Ironwood — a remarkable tree with a lifespan exceeding 1,000 years and exceptional carbon storage capacity.
Forest conservation and carbon projects face a fundamental challenge: how do you credibly estimate what would have happened without the conservation intervention? Getting counterfactuals right is essential for carbon markets to deliver real climate impact.
We're building new approaches to this problem, bringing together state-of-the-art geospatial foundation models with rigorous causal inference methods.
Science-backed counterfactuals
Foundation model powered
Like the Bornean Ironwood, we're committed to long-term thinking and lasting impact.
We're sponsoring ML4EO again this year. Here's why the machine learning for earth observation community matters for carbon markets — and what it still gets wrong about pixels.
The carbon market has an integrity problem. But while the industry obsesses over which biomass map to trust, the real uncertainty is in the carbon baseline.
Combining cutting-edge technology with established scientific methods for credible baseline estimation.
Leveraging the latest advances in Earth observation AI to understand landscape-scale patterns of forest change and condition.
Applying rigorous statistical methods to estimate counterfactual outcomes—what would have happened without the conservation intervention.
Deep domain knowledge in forest carbon monitoring, project assessment, and the challenges facing credible baseline estimation.
We'd love to hear from researchers, developers, and project teams.
