Scope 3 FAQ

Scope 3 is a greenhouse gas accounting term that can be understood as a company’s supply chain emissions. Emissions are “Scope 3” when they are emitted upstream or downstream in a company’s supply chain from sources that a company does not own or control (Source: GHG Protocol).

For many companies, Scope 3 emissions include on-farm practices used to produce commodities like corn, soy, cotton, wheat and others. Implementing regenerative practices including cover crops, reduced tillage and reduced nitrogen can reduce Scope 3 emissions in the industry.

In agriculture, Scope 3 emissions have gotten a lot of attention because many companies do not own farmland, but have a significant part of their overall carbon footprint occurring on-farm as Scope 3 emissions. In addition to this, farmland has been identified as a large potential carbon sink, making it have an even higher potential for climate mitigation.

The last few years have seen a large number of net-zero commitments (more than a third of the global market capitalization has set SBTi targets, many since 2020), including companies sourcing agricultural products. New standards are being proposed in the land sector to help achieve those commitments. Rather than only focusing on Scope 3 emissions, new standards are proposed to take advantage of the potential of agricultural land to remove carbon and be accounted as Scope 3 removals. We have also seen a significant commitment from the biggest players like Nestle, Pepsi, Danone, General Mills, and others to source from acreage following regenerative practices. Companies that work in the agricultural supply chain are committed to reaching net zero via regenerative agricultural practices.

Getting started first means identifying the standards you will use as your guideline, defining the geographic boundaries of your supply shed, and getting a baseline of your emissions.

Check out the CIBO infographic, The Corporate Agriculture Net Zero Journey, for an overview of the steps corporates need to take to set and meet their Scope 3 agricultural emissions reduction targets. We invite you to contact CIBO to find out how the CIBO Impact platform can support your company throughout the process, including baselining, reducing and reporting on Scope 3 emissions.

In offsets, the carbon benefits of regenerative ag practices are separated from the crop and sold to neutralize emissions from any source. In Scope 3, the positive carbon impact and the crop are kept together inside the supply chain. Corporations account for that benefit when they are estimating the carbon footprints of their products or progress towards their Scope 3 emissions reduction or removal targets.

No, the practices can be the same, but the accounting for the carbon benefits and the structure of the carbon programs are different.

The answer to this is: “It depends.” The goals of Scope 3 programs are to get/keep farmland acreage following practices that enable resilient supply chains, rather than creating carbon offsets. This may result in a different variety of eligibility criteria for Scope 3 programs, but much of this evaluation is underway now as standardization for Scope 3 continues. Eligibility for carbon programs is often determined by third party protocols or standards, and we should expect a diversity of Scope 3 program standards, just like in the offset market.

The question of how much traceability is required is still under evaluation by Scope 3 standard-setting bodies. Many companies do not source from the same fields each year, and many fields produce different crops each year. In addition to this, we will continue to see more crop rotation, double cropping, and intercropping as regenerative farming practices are adopted to benefit soil health. Due to the complexity of the current agricultural supply chain, standard-setting bodies are deciding if companies need to know where the crops they purchased were each step of the way, or if other “chain-of-custody” models could enable low-risk accounting (and thus, corporate investment) without requiring significant changes in how these supply chains operate.

Companies can still deploy and deliver results in a Scope 3 program even when they don’t know the specific farms from which they source. They do this by using a supply shed approach. Supply sheds are the larger geographic regions from which the commodities are sourced. A supply shed may be comprised of watershed districts, counties, states, or even custom boundaries such as all farms within a 250 mile radius of a processing facility. CIBO Impact has the unique capability to deliver reliable Scope 3 insights, monitoring, and reporting for supply sheds in addition to individual farms and fields.

CIBO can support any variety of incentive programs for regenerative agriculture, including Scope 3 inventory and emission reduction claims. CIBO is available to be a design partner with Scope 3 customers to meet their needs. CIBO is an active partner and reviews and participates in (when possible) the development of Scope 3 standards with organizations including Greenhouse Gas Protocol, Value Change Initiative, and the Science Based Targets Initiative. We listen to what our customers are trying to achieve and work with them to design and implement programs that make sense for their specific goals and standards to which they report.

Yes! CIBO can model absolute emissions and carbon intensity at the supply shed level on an annual basis. When implementing a Scope 3 program with CIBO, we support establishing a base period or baseline recalculations to help establish a starting carbon intensity for a supply shed. This enables an apples-to-apples comparison of outcomes from a Scope 3 program.

CIBO has developed and validated crop models for corn, soy, cotton, and wheat, and for the cover crops rye and hairy vetch.

We can use these models to estimate crop performance in several functional groups:

• Select C4 grasses like millet and switchgrass
• Select C3 grasses like oats and barley
• Select Legumes like field peas and chickpeas 

Additional crop models are being actively developed and will be available in the future.

Crop yield figures in CIBO Impact are modeled using our crop simulation engine, SALUS. Many variables and parameters that determine crop yield are included in the model in order to create a reliable simulation of real-world conditions. The model inputs fall under four main categories:

• Soil properties (soil pH, soil texture, soil organic matter, etc.)
• Weather (temperature, radiation, precipitation)
• Management (planting date, planting density, amount of nitrogen applied during the season, etc.)
• Characteristics of the specific crop variety grown (e.g. in the case of corn, relative maturity)

Using these inputs and soil hydrological and biogeochemical, and crop physiological relationships contained in our model, CIBO is able to simulate the daily development and growth of the crop at the parcel level. Yield is the mass of the economic part of the crop at simulated maturity.

To ensure that the model outcomes are accurate, the model is trained and independently validated (i.e., using datasets different from the ones used during model training) using a wide array of datasets ranging from simple yield datasets to more complete datasets in which different crop growth aspects (e.g., phenology, biomass, yield, etc.) are measured several times during the season. If the model error is too large relative to the error in the ground truth data, iterative improvements are made to the model to increase model skill and prediction confidence.

To learn more about the complex simulation process underpinning CIBO Impact’s yield model, check out this blog post from one of our leading crop scientists, Kofi Dzotsi, PhD.

Practice history and changes are determined via AI-driven computer vision that analyzes publicly available satellite images of land to identify the tillage type (conventional, conservation or no-till), cash crop and cover crop presence. Cash crop history may also be based on the Cropland Data Layer (CDL) produced by the USDA National Agricultural Statistics Service (NASS).

TL;DR:

CIBO has performed an extensive and independently reviewed model validation and bias analysis report on our systems. The results clearly demonstrate that CIBO’s approach delivers the highest degree of confidence and accuracy.

Details:
CIBO believes a modeling approach to carbon accounting is critical for achieving scale across millions of acres and varied farming practices. All models must be proven to be effective and reliable in order for enterprises to gain the confidence required to put their trust in a model-based carbon accounting approach.

As part of the process of registering a project with Verra under the VM0042 protocol “Methodology for Improved Agricultural Land Management,” CIBO has undergone a model validation process to quantify the model bias and prediction uncertainty of our cropping system model SALUS (Systems Approach to Land Use Sustainability), which is integrated into the CIBO Impact platform. During this assessment, datasets from diverse sources including peer-reviewed papers and multi-institutional research networks were curated and used to replicate in the model the field experiments described by the data. Model outputs resulting from simulations using those experiments are then compared to soil carbon measurements reported in the data.

The results of CIBO’s model validation process showed low model bias (-0.07 tCO2e/acre/year) when comparing model results to the soil sampling data sets across all practice changes and crop functional groups. In other words, a credit figure calculated using CIBO’s model would, on average, be about 0.07 credit less than we would expect from the manual soil sampling data. CIBO’s model bias is also significantly smaller than the error associated with soil sampling and laboratory instruments used to analyze soil carbon.

The results of our study showed a model error standard deviation (prediction uncertainty) of 0.49 tCO2e/acre/year. While the model bias is the mean of model errors across studies, this standard deviation is an estimate of model uncertainty and should be interpreted as the average distance separating the same model errors and the bias. A small standard deviation means the model errors are consistent across studies, and a large value means the model errors differ greatly among studies. This number is used to ensure that we properly quantify the uncertainty of a carbon project as a whole. The prediction uncertainty is one variable that Verra will use to calculate an uncertainty deduction from issued credits and ensure that the project does not issue more carbon credits than are actually generated.

CIBO is extremely pleased with the model validation results, as they show we can be confident that the SALUS model will produce results that reflect the real world and that it can be relied upon to not overpredict carbon credits. At the same time, this model validation report was just the beginning. We are engaged in a continuous process to ensure that the model continues to produce results that are accurate and reliable for both carbon credit markets and Scope 3 programs.

CIBO chose to register a project first with Verra, but most of the major registries have similar requirements for model validation. We believe our experience is broadly applicable to other registries and to the voluntary Scope 3 standards that are emerging.

To see the complete results and an in-depth explanation of CIBO’s model validation process, check out our science team’s paper Confidence in Carbon Modeling.

To learn more about the role of crop modeling in scaling regenerative agriculture, check out this GreenBiz article by CIBO co-founder Bruno Basso, PhD.