A metric measuring the global mean surface temperature change at a chosen time horizon, following an emission of a unit mass of a given substance or a change in surface albedo.

A metric measuring the radiative forcing following an emission of a unit mass of a given substance, accumulated over a chosen time horizon.

One or more practices or processes carried out by an operator, or a group of operators, resulting in a permanent carbon removal, a temporary carbon removal through carbon farming or through carbon storage in products, or soil emission reductions through carbon farming where such carbon farming, overall, reduces the emissions of carbon from soil carbon pools or increases carbon removals in biogenic carbon pools.

The spatially delimited area or areas where the activity takes place.

A period during which the activity generates a net carbon removal benefit or a net soil emission reduction benefit, and which is determined in the applicable certification methodology.

In the context of lifetimes, adjustment time is the time scale characterising the decay of an instantaneous pulse input into the reservoir.

Agroforestry includes all types of agricultural systems, in which an agricultural component, either arable land, grassland and/or livestock, is combined and interacts with woody perennials like trees, hedges, or shrubs.

The fraction of total CO₂ emissions (from fossil fuels and land-use change) remaining in the atmosphere.

The proportion of incoming solar radiation that is reflected by a surface or object, sometimes expressed as a percentage but more generally ranging from 0 (all incoming solar radiation is absorbed) to 1 (all is reflected back towards space).

Agricultural Soil Emission.

The amount of agricultural soil emissions under the baseline.

The total amount of agricultural soil emissions of the activity.

Positive or negative evaluation of a behaviour.

Umbrella definition for reference locations where soil, vegetation and management data are collected with sufficient detail and/or continuity to support our understanding of SOC dynamics in agricultural systems.

Living biomass, litter, deadwood, dead organic matter, mineral soils and organic soils as listed in Section B, points (a) to (f), of Annex I to Regulation (EU) 2018/841.

A process involving the adjustment of parameters and constants within a model so that it predicts the measured values more accurately.

Carbon certificates equal 1 ton CO₂-eq that has been removed, avoided or reduced through the implementation of carbon farming practices, which preferably has been verified by an independent auditor. Carbon certificates do not allow companies to offset their emissions (use it in their carbon accounting) and declare itself carbon neutral. Carbon certificates allow companies to contribute to the decarbonisation of their region/country (and make a positive contribution). Carbon certificates can be realised through local, regional or national/domestic carbon farming schemes or initiatives (e.g. Claire, Soil Capital, Label Bas Carbone, Stichting Nationale Koolstofmarkt). Contrary to carbon credits, carbon certificates cannot be traded internationally (own definition).

Carbon credits equal 1 ton CO₂-eq that has been removed, avoided or reduced through the implementation of carbon farming practices, which preferably has been verified by an independent auditor. Carbon credits allow companies to offset their emissions (use it in their carbon accounting) and declare themselves carbon neutral. Carbon credits are realised through international carbon standards (e.g. Verified Carbon Standard, Gold Standard) endorsed by ICROA = the International Carbon Reduction & Offset Alliance. Contrary to carbon certificates, carbon credits can be traded internationally (own definition).

Any practice or process carried out over an activity period of at least five years, related to the management of a terrestrial or coastal environment and resulting in the capture and temporary storage of atmospheric or biogenic carbon in biogenic carbon pools, or in the reduction of soil emissions.

One metric tonne of CO₂ equivalent of certified temporary net carbon removal benefit generated by a carbon farming activity and registered by a certification scheme in its certification registry or, as applicable, in the Union registry provided for in Article 12 of the CRCF Regulation.

Process of transferring carbon from the atmosphere into the soil through plants or other organisms, which is retained as soil organic carbon resulting in a global carbon stock increase of the soil.

A conformity statement issued by a certification body certifying that an activity complies with this Regulation.

An audit carried out by a certification body.

An accredited or recognised independent conformity assessment body that has concluded an agreement with a certification scheme to carry out certification audits and issue certificates of compliance.

The period between a re-certification audit of an activity and the most recent preceding certification audit or re-certification audit of that activity.

An organisation that certifies the compliance of activities and operators with the quality criteria and certification rules set out in the CRCF Regulation.

Avoided energy accumulation in the Earth System or avoided global warming at a given point in time.

An anthropogenic intervention that reduces the sources or enhance the sinks of greenhouse gases.

A sample obtained by combining several individual samples, often to reduce analytical costs or to obtain an average measure.

A range of values, derived from the sample statistics, that is likely to contain the value of an unknown population parameter.

The amount of carbon removals under the baseline.

The total amount of carbon removals of the activity.

Groupings of “similar” crops that could be used interchangeably in the arable LUST when model calibration for the specific crop type and carbon farming practice is not possible.

Process of integrating observational data into models (e.g. process-based) to enhance the accuracy of simulations and reduce uncertainty of predictions while ensuring the consistency between modelled outputs and observed data. Most commonly, this technique is applied inform crop models (e.g., WOFOST, APSIM, SAFYE-CO2), with satellite-derived data to improve estimates of key biogeophysical variables like biomass, yield, and soil moisture. Various assimilation methods exist, including sequential (e.g., Ensemble Kalman Filter), variational (e.g., 4DVAR), calibration-based, and other probabilistic approaches (e.g., MCMC Bayesian inference). While remote sensing data is the primary source of observational inputs, other data, such as CO₂ flux measurements, can also be assimilated to refine model predictions. Assimilation of remote sensing (RS) data allows a more objective accounting of biomass spatial variability (and carbon inputs to the soil) while relaxing the need for some input data (e.g. date of sowing, fertilisation) since the data assimilation process allows to implicitly account for some practices, stresses (nitrogen, water), extreme events (e.g. hail, floods), spontaneous regrowth, etc. impacting vegetation development.

Classification, or stratification, of European land into strata of relatively homogeneous climate, geology, geomorphology, soils, vegetation and land cover.

Refers to the difference between the measured (measurement error) or estimated (model error) value and the true (but unknown) value of the Soil Organic Carbon (SOC) stock.

Agricultural land parcel including its boundaries.

Field scale refers to the level of a field parcel.

The increase in direct and indirect greenhouse gas emissions over the entire lifecycle of the activity which are attributable to its implementation, including indirect land use change, calculated, where applicable, in accordance with the protocols set forth in the 2006 IPCC Guidelines for National Greenhouse Gas Inventories and any further refinement to these 2006 IPCC Guidelines.

A legal entity that represents at least two operators and is responsible for ensuring that those operators comply with the CRCF Regulation.

Insetting occurs when project outcomes are purchased by agrifood companies (within the value chain), which aim to reduce their scope 3 emissions (see section 7.1). This mostly occurs through the payment of price premiums (a higher price per production unit), rather than through the payment for carbon certificates / credits as such (own definition). Insetting claims are only valid under a rigorous set of conditions, including that the reductions and removals involved are additional, not over-estimated, and exclusively claimed. Further, insetting can only be used to claim net zero status to the extent it is “like for like” (see definition carbon neutrality) with any residual emissions.

LULUCF Soil Emission.

The amount of LULUCF soil emissions under the baseline.

The total amount of LULUCF soil emissions of the activity.

Land Use x Soil Type (LUST): MARVIC focuses on 1. arable farming on mineral soils, 2. permanent grassland on mineral soils (>5 year grassland), 3. agriculturally managed peatland, 4. woody crops/agroforestry (including orchards, hedges and small wood lots).

Can be decomposed into stochastic (random) measurement error and systematic measurement error. Stochastic error refers to the random noise (small-scale fluctuations), it has to do with how accurate any single measurement is and can be reduced by more replicates. Systematic measurement error is a consistent bias (e.g. miscalibration or systematic sampling bias), which shifts measurements up or down. It affects accuracy and cannot be reduced by mere replication; it requires improved calibration, instruments or protocols. Measurement error can also be thought to have a third component called the representativeness error which is the mismatch between where/how you measure and what the true quantity actually presents, however this can be framed as a sampling design issue.

Soil that is not organic soil.

The smallest difference between population parameters can be detected with a given level of confidence and power.

Definition of the conceptual model describing the system of interest, that identifies the processes to be modelled and defines the mathematical relationships and parameters that simulate real-world processes.

For a model to be considered suitable for usage in a Monitoring, Reporting and Verification (MRV) system over an area of interest, it must be suitably calibrated and validated for the LUST(s), pedoclimatic areas, carbon farming (CF) practices, and crops of interest in that area. The model domain is defined as the combination of LUSTs, pedoclimatic areas, CF practices, and crops that the model has been suitably calibrated/validated on.

Can be decomposed into input-data error and process (structural) error. Model input-data error arises because model inputs (e.g. initial conditions, weather drivers, soil properties, parameters) have measurement errors of their own which propagate through the model. Process (structural) error arises when important processes are missing, oversimplified, or misrepresented in the model and can be defined as the remaining error if all inputs were to be known perfectly. Similar to measurement error, model error can also be thought to have a third representativeness error component which can be framed as the wrong domain of applicability.

Process of setting the model state variables to reflect the initial state at the start of the model simulation. Usually conducted during a so-called model “spin-up phase”.

Empirical models represent observed relationships between Soil Organic Carbon (SOC) changes and environmental, or environmental and management practices. Can be regression-based or can implement machine-learning techniques when solid domain knowledge is available.

Soil process models that simulate Soil Organic Carbon (SOC) dynamics from Soil Organic Matter (SOM) turnover, considering the effects of climate, soil factors, land use, and management (e.g. C-TOOL, Roth-C, ICBM).

Ecosystem models that also integrate feedback from multiple soil-plant-atmospheric processes (e.g. STICS, DAYCENT, Landscape DNDC, APSIM).

The set of parameter values chosen are deemed suitable for model use in the specific domain of interest.

Uncertainty in model predictions, as determined from comparison to direct measurements. Measurements used to determine model prediction error must be the same as those used to validate the model.

Ensuring a model's transparency, openness, and pre-existing testing to verify its use without calibration and validation. If a documented, version-controlled, and publicly available model has been successfully tested in scientific publications for a specific Pedoclimate × Crop category, it can be used directly.

Testing the model in a new domain to verify if it can be used/is valid without further calibration/validation by assessing its performance on a new set of data.

All the steps required for preparing and running one or more models to produce model predictions. This includes input data curation, model configuration and parametrisation, initialisation, spin-up, and post-processing.

A period during which the soil emission reduction or storage of carbon is monitored by an operator or a group of operators, which covers at least the activity period, and which is determined in the applicable certification methodology.

In an Monitoring, Reporting and Verification (MRV) system, sequence of actions that are directly linked to running a model (or multiple models) using a modelling workflow.

An event or series of events resulting from natural processes, such as wildfires, storms, extreme climate events, or biological and insect outbreaks, that lead to significant alterations in the structure and functioning of ecosystems.

Net removal of CO₂-equivalents of greenhouse gases from the atmosphere.

LSE_baseline – LSE_total + ASE_baseline – ASE_total – GHG_associated > 0

Reducing GHG emissions (including through avoided emissions), or increasing GHG removals through activities external to an actor, in order to compensate for GHG emissions, such that an actor's net contribution to global emissions is reduced. Offsetting is typically arranged through a marketplace for carbon credits or other exchange mechanism. Offsetting claims are only valid under a rigorous set of conditions, including that the reductions/removals involved are additional, not over-estimated, and exclusively claimed. Further, offsetting can only be used to claim net zero status to the extent it is “like for like” (see definition carbon neutrality) with any residual emissions.

Automated workflow that connects data (e.g. land cover, weather or soil property maps, statistics, farm data, remote sensing data) and models via automated workflows to quantify and assess accuracy of soil organic carbon stock changes, carbon budget components (e.g. biomass, CO₂ fluxes) and soil GHG emission. An OPC can contain one model (e.g. a soil model using databases for biomass production input) or several models (e.g. a soil model taking biomass data input from a crop model driven by satellite data, several soil models providing an ensemble of simulations).

Any natural or legal person or public entity that operates or controls an activity, or to whom decisive economic power over the technical functioning of the activity has been delegated; in the case of a carbon farming activity, ‘operator’ means a farmer as defined in Article 3(1) of Regulation (EU) 2021/2115, any other manager of an activity in a terrestrial or coastal environment, a forest owner or manager as defined by national law, or a competent public entity.

The benefits that would have been obtained if the input resources had been used in the second most beneficial activity.

Organic soil as established in accordance with the 2006 Intergovernmental Panel on Climate Change (IPCC) Guidelines (‘2006 IPCC Guidelines’) or under national definitions accepted under greenhouse gas (GHG) inventories reporting.

A layer of soil that primarily consists of partially decomposed plant material, that has accumulated under conditions of waterlogging.

A time period over which the total peat layer would disappear in the absence of the activity.

Geographical areas of relatively homogenous soil type and climate conditions.

Perceived ease or difficulty of performing a behaviour, influenced by factors such as expected obstacles, access to resources, and previous experience.

Sustained storage of the sequestered carbon in the long term (>1000 years).

Plot scale refers to the sub-field level.

Attribute of policy that systematically reduces conflicts/trade-offs and promotes synergies between and within different policy areas to achieve the outcomes associated with jointly agreed policy objectives.

The degree to which repeated measurements or estimates of the same quantity under the same conditions yield similar values; the closer the values, the higher the precision. High precision means low dispersion (low variance) of the error term. Precision is determined by stochastic (random) error and is conceptually independent of accuracy: for instance, measurements can be precise but systematically wrong, or imprecise but, on average, unbiased.

A measure of the global mean surface temperature change at a chosen time horizon resulting from a Project Intervention.

A measure of the radiative forcing resulting from a Project Intervention, accumulated over a chosen time horizon.

Running an OPC for a Project over a specific Project area, under specific farming conditions, and using relevant data layers (i.e. using a Monitoring, Reporting and Verification (MRV) tool across one or multiple farms).

Project scale refers to the level of the project, which typically includes a set of parcels from a group of operators.

Private costs are those that accrue to private agents (households, farms, firms), public costs are those that accrue to governments at different levels.

Concept used to quantify the change in energy balance of the Earth’s atmosphere. Various factors contribute to this change, such as concentrations of GHG and aerosols, or changes in surface albedo and solar irradiance.

An audit carried out in the process of renewing a certificate of compliance issued by a certification body.

New activity period.

The number of units in the sample is referred to as the sample size.

The process of selecting units from a population to estimate characteristics of the whole population.

The method or plan used to select sampling units from the population.

The geographic coordinates or area where the sampling unit is located.

The specific point or area within the study region where a sample is collected.

A sampling unit is an element or a group of elements considered for selection in some stage of sampling.

Scope 3 emissions are a consequence of the activities of the company but occur from sources not owned or controlled by the company. Some examples of scope 3 activities are extraction and production of purchased materials; transportation of purchased fuels; and use of sold products and services.

A system that combines, on the same plot, woody perennials (trees, hedges or shrubs), and annual or perennial crops (cereals, protein crops, etc.), with joint management of both the tree and crop components.

A system that combines, on the same plot, woody perennials (trees, hedges or shrubs), a pasture or other forage cover, and the grazing of animals (most often ruminants), managed in a way to produce both timber/trees and livestock.

A broad class of lower dimensional models of the energy balance, radiative transfer, carbon cycle, or a combination of such physical components.

An increase in Soil Organic Carbon (SOC) stock at a given unit of land, starting from an initial SOC stock or compared to a business-as-usual value (does not always result in climate change mitigation or carbon sequestration in soils).

An anthropogenic intervention to reduce Soil Organic Carbon (SOC) losses compared to a business-as-usual scenario.

The size of the Soil Organic Carbon (SOC) pool (i.e., SOC stock or SOC content).

Indicates tendency or respondents to bias their answers in a survey to appear in a more favourable light.

The reduction of net greenhouse gas emissions from biogenic carbon pools as listed in Section B, points (e) and (f), of Annex I to Regulation (EU) 2018/841 or the reduction of greenhouse gas emissions from the IPCC source category of agriculture, subcategory 3.D agricultural soils, as determined pursuant to Regulation (EU) 2018/1999, where the relevant activity, overall, reduces the emission of carbon from soil carbon pools or increases carbon removals in biogenic carbon pools.

One metric tonne of CO₂ equivalent of certified net soil emission reduction benefit generated by a carbon farming activity and registered by a certification scheme in its certification registry or, as applicable, in the Union registry provided for in Article 12 of the CRCF Regulation.

A measure of the statistical accuracy of an estimate, indicating the variability of the estimate due to sampling.

The set of variables that determines the state of the dynamic system at the start of the project (T0).

The process of dividing a population into subgroups (strata) that are more homogeneous.

Perceived social pressure to perform or not perform a behaviour.

The length of time after which the positive radiative forcing due to increases in CH₄ emissions at a restored wetland is overtaken by the cumulative negative radiative forcing due to CO₂ uptake.

CR_baseline – CR_total – GHG_associated > 0.

Model to predict and explain human behaviour. It suggests that intentions are central to performing a behaviour. Intentions are influenced by: 1. Attitude towards the behaviour, 2. Subjective norm, 3. Perceived behavioural control. Actual behaviour is contingent on volitional control and the availability of opportunities and resources.

The state of incomplete knowledge about the value of a quantity or the magnitude of an error. Statistically, uncertainty is represented by a probability distribution over the quantity (or over its error), reflecting which values are considered possible and how likely they are. Uncertainty can be summarised by measures such as standard deviations, confidence or credible intervals, and prediction intervals. This measure corresponds to the UNC (uncertainty deduction factor) in the CRCF annexes to the Delegated Act (draft carbon farming methodology, January 2026).

The process of evaluating the performance of a model with regard to measured values, demonstrating its satisfactory performance in terms of suitability and characterisation of model prediction error.

The presence of real differences in the true value of a quantity across space, time, or units. Variability is a property of the system, not the measurement device. It is typically summarised by the variance or standard deviation of the true values. More data improve our characterisation of variability but do not reduce it. When measurements are precise there will be little uncertainty about variability in the sample.

The height of a phreatic water surface relative to the soil surface.