Isometric has certified the Model and Measurement Based Quantification of Enhanced Weathering Module, which sets out how models can be used alongside direct field measurement to quantify removals under the Enhanced Weathering in Agriculture Protocol.
Certification follows a 30-day public consultation, with feedback from academics, project developers, buyers, and non-profits. For the full details on how Isometric addressed that feedback, you can read the consultation summary.
The module provides an alternative quantification approach and does not replace the existing measurement-based approach, which remains available to all projects. All other protocol requirements continue to apply, including feedstock characterization, environmental safeguards, life cycle assessment, and independent verification.
Why models
In December 2024, Isometric issued the world’s first independently verified Enhanced Weathering (EW) certificates to InPlanet. Since then we have issued EW certificates to eight further projects, from pilots through to commercial-scale deployments.
EW has the potential to remove billions of tonnes of carbon dioxide from the atmosphere every year. It will only do so if Monitoring, Reporting and Verification (MRV) can scale alongside it.
Today’s EW projects quantify removals primarily through direct measurement. Soil samples are taken across project fields, analyzed in the lab for changes in cation chemistry, and the weathering signal is statistically separated from natural soil variability. This works, and it is a large part of why EW certificates cost what they do.
Intensive sampling is expensive and has a practical ceiling. Soils are heterogeneous, and the weathering signal a project is trying to detect is often smaller than the background variation. Detecting it reliably requires sampling densities that work for pilots but not across the hundreds of thousands of hectares over which EW needs to be deployed to reach gigatonne scale.
Models provide a solution. They enable projects to concentrate intensive measurement at representative monitoring sites and extrapolate across the wider deployment area, and they are among the most actively explored routes to EW operating at a climate-relevant scale.
Zeke Hausfather, Climate Research Lead at Frontier, an advance market commitment to purchase over $1 billion in permanent carbon removal by 2030, said: “Creating a path toward more model-enhanced crediting will be a key part of scaling enhanced rock weathering. But we need to ensure that path starts with the use of modeling side-by-side with current measurement-intensive approaches, and only reduce, but never eliminate, the need for representative sampling as models are proven out by real-world data. Credible models must also be transparent and open to the broader scientific community to assess and improve.”
What the module requires
Without adequate validation, models can systematically over or underestimate removals in ways direct measurement would catch. Over the past year, Isometric’s Science Team worked through what rigorous model-based EW quantification requires in practice. The certified module sets out the standards for accuracy and performance a model must meet before it can be used to issue certificates, how direct measurements should anchor it, and how to handle the uncertainty of extrapolating from monitored to unmonitored ground.
The module draws on input and consultation from EW experts, data scientists, geochemists, and stakeholders from across the EW ecosystem, as well as participants in our workshop at ERW 2025. The requirements are grounded in three principles:
- This is a measurements and models approach. Direct field measurement remains the foundation and models extend its reach. Projects qualify under this module by showing that their model performs against real-world measurements, at validation and again at every reporting period. There is no route to certificates that bypasses field data.
- The bar for certification is high. The module sets strict requirements on training data, cross-validation, uncertainty propagation, area-of-applicability checks, and ongoing re-validation.
- Proprietary models are held to the same standard. They must pass review by an independent expert panel covering scientific basis through to model code, and the findings of each review will be published.
The best place to start is our companion explainer, which walks through the module and its requirements. You can read the full module here.
