The European Commission’s methodologies to certify carbon farming activities in the EU are now worse than the original draft presented back in 2024. The EU must not allow this to pass.

The European Commission will soon adopt the methodologies to certify carbon farming activities in the EU under the Carbon Removals and Carbon Farming (CRCF) Regulation, which have gone from bad to worse. These will become law after a two-month scrutiny period by the European Parliament and the Council of the European Union. 

By sending the methodologies back to the Commission’s drawing board, the Parliament and Council can fix one of the most crucial elements of these methodologies, the mandatory co-benefits criterion.

Moving beyond carbon 

Article 7(1)(f) and 7(2) of the CRCF Regulation obliges all carbon farming activities to, at least, generate co-benefits for the protection and restoration of biodiversity and ecosystems, including soil health and the avoidance of land degradation. This is because the natural context in which carbon farming activities occur demands a holistic approach incorporating benefits beyond sequestration. 

Given the vulnerability of the stored carbon and the high risks of reversals, the main benefits for many so-called carbon farming activities are actually ecological rather than climatic. Despite these clear advantages, this fundamental element has been watered down in the course of the four drafts of the methodology. 

The first two versions, shared with members of the Carbon Removals Expert Group in October 2024 and April 2025, were referred to as ‘draft elements’ and involved three separate documents, one for each of the three carbon farming practices (soil carbon in mineral soils and agro-forestry, peatland rewetting, and tree planting). 

The subsequent two versions were draft Delegated Acts that grouped all practices, with many provisions, such as on additionality, liability and sustainability, streamlined across all three. The third version was shared in January 2026 (this version was subject to an EU public consultation), and the supposedly final version in April 2026. For comparative consistency, this article will focus on the provisions related to soil carbon in mineral soils.

Circular logic

The first version offered a quantitative or results-based assessment of the  mandatory co-benefits criterion, referring to two of the three indicators listed under Article 11(2) of the Nature Restoration Regulation (NRR). These indicators are the stock of organic carbon in cropland mineral soils and the share of agricultural land with high-diversity landscape features. According to that version of the methodology, any activity showing an improvement in one of these two indicators would be considered compliant, adding that since sequestration in mineral soils would, by definition, improve the stock of organic carbon, compliance was automatic. 

This was particularly problematic as it meant that the very activity the scheme was aiming to credit was also being used to prove biodiversity benefit. In other words, sequestering carbon would automatically comply with biodiversity enhancement because it sequesters carbon. In addition, the extent to which NRR indicators could be integrated into a CRCF system was questionable, given that NRR is a national-level policy with different monitoring cycles.

Still, basing the mandatory co-benefits assessment on indicators, also supported by the ESABCC, was a good idea. CMW therefore recommended looking into indicators that could measure the number, type and role of species in soils. This was not taken on board by the Commission.

Immeasurably problematic

The second draft involved the first weakening of the text. This version kick-started the departure from a quantitative to a qualitative approach to the mandatory co-benefits assessment, which removed the need for concrete measurements based on indicators. It also introduced a discrepancy between how the CRCF measures carbon and how it treats and measures co-benefits. 

Although measuring, monitoring, reporting, and verification (MRV) by a third party was mandatory for climate benefits, the same level of results-based rigour was no longer needed for the mandatory co-benefits, despite these being as important, if not more so than potential climate impacts. 

The second version also introduced a series of options. As seen below, operators could choose between one of the following three options:

1. The activity has been demonstrated in peer-reviewed scientific literature to have beneficial impacts for biodiversity protection or restoration.
2. The activity corresponds to at least one of the types of measures provided in the typology of measures and is consistent with restoration within the meaning of Article 3(3) of the NRR and with the relevant national restoration plan.
3. The activity is qualitatively demonstrated to contribute to improved condition or maintain good condition of a habitat type or a habitat of species of Union interest, using established methodologies such as those under the Habitats and Birds Directive, or the EU-wide methodology on mapping and assessing ecosystem condition for other ecosystems.

The first option is particularly worrying, as it allows any scientific literature to be used for compliance purposes. It failed to introduce criteria on permissible literature, for instance, requiring specificity to the conditions of the project, and to prevent the results of the literature from being abused or misinterpreted by the operator. It also ignores the impact of local conditions and the fact that results can vary significantly over place and time.

The European Commission interprets the second option as a qualitative measure, even though Article 3(3) of the NRR explicitly requires the use of indicators. 

Meanwhile, the third option is very vague, as it simply refers to the methodologies of the Habitats and Birds Directive or the EU-wide methodology for mapping and assessing ecosystem condition for other ecosystems, without clarifying how the rules should be operationalised in the context of the CRCF. 

Clearer loopholes

The third draft continued to allow for only one of the three options presented in the previous version and entailed a definitive departure from the quantitative approach. The reference to the NRR went from “is consistent with restoration within the meaning of article 3(3)” to “actively or passively assists in the ecosystem ‘restoration’, as defined in article 3(3)”, weakening the link to the entire definition of restoration, and consequently, the need for indicators. 

As to the first option, the reference to scientific literature was slightly expanded to include examples of such literature, but continued to lack explicit safeguards to prevent the use of literature that is not representative of the operator’s project. 

The link to the Habitats or Birds directive was slightly refined to include management plans and “other implementing measures” under the Birds directive, which correspond to the ecological requirements of the natural habitats as described in Article 6 of the Habitats directive. Despite offering slightly more clarity, the implementation and transposition to a CRCF context remained unclear. Reference to the methodologies under those directives or the EU-wide methodology on mapping and assessing ecosystem conditions for other ecosystems was maintained. 

A la carte

The final version reveals yet another weakening of the text:  while a single operator can seek certification for a variety of practices listed (non-exhaustively) under the soil carbon in mineral soils methodology, the assessment for the mandatory co-benefits only needs to be carried out for one of them. Proving, through the weak options presented above, that the mandatory co-benefits requirement is met for one of their practices is sufficient. 

This is particularly problematic as eligible practices also involve rotational grazing of nitrification inhibitors, which pose clear risks for biodiversity, soil health and land degradation. Thus, operators could be using nitrification inhibitors, yet only carry out a mandatory co-benefits assessment for cover crops, and still pass the test. 

A slight improvement came with the peer-reviewed literature in that a reference to ‘similar pedo-climatic conditions’ (i.e. soil and atmospheric conditions) was introduced, limiting the risk that just any literature is used. Still, issues with the interpretation of the results remain.

Why the weakening?

The entire CRCF process has been torn between the conflicting demands of minimising the administrative burden and maximising the environmental benefits. Yet, during this current mandate, deregulation has trumped environmental quality. In a carbon farming certification context, this trade-off is inherently flawed since sacrificing stringent and robust methodologies will come at the expense of environmental rigour, amongst other things. 

How many times will the mandatory co-benefits criterion be weakened? Clearly, far too many. Measurable biodiversity results through the use of indicators is crucial, particularly in a context where demand is uncertain and investors are willing to pay a premium for these added benefits. The strength of these activities lies in their capacity for ecosystem restoration. This should be exploited rather than introducing flexibilities that will make it harder, if not impossible, to determine whether a particular practice has in fact delivered on its co-benefits. 

It is regrettable that the Commission has taken a central component of the carbon farming methodologies and treated it as an afterthought. Around two-thirds of the EU’s agricultural soils are depleted, biodiversity is deteriorating, with agriculture being on track to become one of the highest polluting sectors in the next decade. As it stands, the CRCF is unlikely to improve any of this. 

Soon, the EU institutions will have to decide whether or not to endorse the carbon farming methodologies. They should reject this current framework, which risks doing more harm than good.