There’s a significant shift taking place for food companies in relation to agricultural emissions and carbon accounting, and with it comes the need for more accurate Land Use Change (LUC) measurement.
The Land Sector and Removals Guidance (LSR) from the Greenhouse Gas Protocol (GHG) and the Science-Based Target Initiative’s Forest, Land and Agriculture Guidance (FLAG) have played a role in the shift. Companies that grow, manufacture or sell food products are impacted by FLAG, and its announcement sparked a growing wave of conversation and reevaluation around measuring emissions at the agricultural production stage.
The HowGood team has fielded numerous questions from partners and clients on the topic of land use change. This article outlines the methodology behind our first-of-its-kind offering for measurement in the Latis platform, along with supporting information to help build knowledge of challenges and opportunities.
From Land Management to Land Use Change: An Overview
The focus throughout the food industry has previously been on measuring on-farm emissions. Until now, there was really no standardized way of estimating land use change, but with the guidance now necessitating a shift to set targets against, account for, report on and reduce LUC, new methods are necessary.
HowGood has introduced a scalable and practical approach that’s available directly within the Latis platform. With agricultural production making up 80% of the total emissions for food products, the HowGood team has always had a deep focus on accurately measuring the impact of these emissions.
Now, with SBTi guidance to set targets for FLAG emissions, we have aligned our methodology with the LSR guidance for more granular accounting of agricultural emissions.
The traditional method of using direct land use change (dLUC), which relies on specific farm data over the last 20 years, isn’t scalable across all ingredients and products within a company’s portfolio. The HowGood approach employs statistical land use change (sLUC) modeling. This method not only aligns with the latest FLAG reporting requirements but also offers scalability, ensuring that food companies can comprehensively measure the carbon footprint across their entire product portfolio.
What is the Land Use Change Metric?
Land use change is one of three categories of emissions within FLAG guidance, along with land management and carbon removals.
While deforestation may be the most frequently talked about component of land use change, there are additional categories, such as soil draining and forest degradation. Overall, the land use change metric involves identifying conversions that have taken place from one ecosystem to another.
Statistical Versus Direct Land Use Change
With the new guidance in mind, it’s important to understand how the metrics for statistical land use change (sLUC) differ from those of direct land use change (dLUC).
dLUC is a measurement of how much land has been converted for agricultural use over the past 20 years, on a company’s own lands or on any land that’s specific to your supply chain. It requires that you obtain granular data for all of your suppliers’ land usage, knowledge of exactly where all of their farms are located, where and how they’ve been growing which crops, etc. This information would be needed for every farm from which you source crops.
sLUC is measured at a regional or jurisdictional level. Because the suppliers and farms in a supply chain will not all have the information needed for dLUC, sLUC provides an alternative way to estimate land use change.
The Land Sector Removal (LSR) guidance includes direction on how to calculate (sLUC), and also provides some choice for flexibility:
Time Discounting Factor
To measure using sLUC, companies begin with the total land use change due to agriculture within a region. This can be at country-level – total land use change within the United States, for example. The standard time period to look at is 20 years per LSR guidance. So, you consider all of the land use change that has occurred within that period. This is the first choice to be made, and companies can choose linear or equal. Equal means you discount equally across the years since the change occurred. Linear is the method we chose because it apportions the highest impact to the year the LUC occurs, and a lesser amount each subsequent year for 20 years. This aligns with the method that SBTi used when developing the FLAG target setting methods.
Product Allocation Factor
Product allocation factor (PAF) also starts with the whole land use change number at the regional or national level, then allocates all of the crops that were grown within that area. There are two different options for this: the shared responsibility method and the product expansion method.
The shared responsibility method looks at all the crops raised within that region, and identifies the percentage of the agricultural footprint that they occupy. So for a crop that has 50% of the total agricultural land footprint, in that year it gets 50% of the land use change after time discounting.
The product expansion method looks only at products that have expanded their land footprint within that year and how much of the land of the expanded footprint that that crop took up. For example, if the land area was expanded 50 hectares, and one crop was responsible for 10 hectares of that expansion, then it would get 20% of the land use change allocated to it.
Note that LSR draft guidance states the PAF should be computed for individual years using either a shared allocation or product expansion allocation method. HowGood’s methodology uses shared allocation instead of product expansion.
Types of Emissions
In addition to the above factors, the guidance states that emissions due to land conversion as it relates to above and below ground biomass, soil carbon stock changes, and dead organic matter must be calculated. For example, HowGood included factors for soil drainage and pasture.
LSR guidance states that the time discounting factor and product allocation factors must be viewed for every year within that 20 year time period. This starts to have a big impact when modeling scenarios, especially when combined with the differing options of using either the linear or equal time discounting factor.
HowGood’s Latis uses sLUC and is designed with in-app tools that enable this level of modeling in order to support informed decision-making and abatement strategies.
Benefits of the Shared Responsibility Approach
The HowGood team tested several scenarios when calculating sLUC and chose the shared responsibility approach.
Here’s what was found:
- Shared responsibility better represents the need to efficiently use existing land, whereas using the crop expansion approach doesn’t represent low yield crops that occupy large portions of a land area.
- Nationally reported data on crop land occupation is not reported consistently for all nations. Missing years of data can be imputed, but these choices on how to impute have implications on the associated LUC for the country.
For example, the Food and Agriculture Organization (FAO) reported pasture area in Brazil is mostly imputed data. Using these figures gives a consistent approach to land occupation but when the values were applied in the product expansion approach, the result was an allocation of zero LUC to pasture in Brazil in 2001-2006 and 2018 -2021. This doesn’t match other evidence of LUC in Brazil during these years.
HowGood’s findings demonstrated that due to the quality of national data, the product expansion allocation method produced inconsistent results year over year. When data wasn’t reported in one or more years, data was then imputed using the most recently reported official figure. Although imputed yearly data makes sense, the values remain static for some years when official figures aren’t reported. If a country and commodity showed an increasing trend in LUC in prior years of official figure reporting, that trend could very likely continue if official figures continued to be reported. A product expansion approach for national data of varying quality could be better used as a single allocation factor over all years in the assessment, rather than computing on a per year basis.
- The shared responsibility approach used in tandem with linear discounting still penalizes crops with expanding land occupation, as the more recent years of LUC are more heavily weighted by the time discounting along with the larger share of agriculture area occupied by these expanding crops.
Note again that shared responsibility and product expansion are both allowed by the GHG Protocol draft LSR guidance. The approaches begin with the same data – the total land use change emissions in a region or jurisdiction. PAF only impacts the share of the region’s LUC allocated to the product. Both approaches are therefore sensitive to changes. For example, adding a new commodity to a growing region or removing one altogether would impact both in these ways:
- The land area occupied would be considered an “expansion” for that crop.
- The land area occupied would become part of the total land area and receive its share of land occupation under the shared responsibility approach.
As sLUC becomes more widely adopted and final draft LSR guidance is released, it’s expected that there will be changes in outcomes.
The Need for New Research and Methodologies
There has been no previous standard for measurement of Land Use Change. The limitations of on-farm emissions and direct land use change methodology quickly become apparent when you try to scale at the level needed for reporting.
For example, with dLUC you would use your primary data for a specific farm over the last 20 years. While you may have that data for one or two farms, or for one or two crops, it’s unlikely that you will not be able to do that for all ingredients across your products. It’s not scalable.
Another example of note is that LUC for animal products is driven largely by their feed. HowGood used country-specific typical animal diets to calculate the LUC for each animal product produced within a country. Because some firms who offer LUC datasets previously didn’t provide values for animals/feed, we built our own. The LUC associated with soy, palm and pasture within the diet was used.
Additional issues leading to the need for new solutions include lack of previously existing data granularity, regional variations, and outdated data.
These examples and challenges combined with the evolving regulatory landscape and increased stakeholder awareness pushed the need for the development of new methodology.
HowGood’s solution is the first-of-its-kind and is enabled by its existing vast database on agricultural emissions. Because it utilizes statistical land use change it’s scalable across entire product portfolios, which is now necessary for companies to report progress against FLAG.
Because SBTi and LSR allow choices and LSR offers more variation, some of the larger decisions for modeling sLUC were built and pressure-tested for what HowGood’s research team believes are the most viable scenarios. HowGood’s methodology takes into consideration each of the following:
- Data Assessment – Data sets were identified that have the best coverage to support the choices noted above
- Scientific Consideration – The focus here is on existing standards and conventions, for example, the draft LSR – which was written by international experts and is currently being vetted through the draft process
- Moral and Ethical Consideration – The methodology includes interplay between what the data allows to be represented and what we found to be the most fair representation of impacts, taking into consideration the constraints on land available for agriculture.
Managing Land Use Change Alongside Land Use Management Emissions in Product Carbon Footprints
Land Use Management and Land Use Change emissions are typically the highest contributors to a product’s carbon footprint. This is why HowGood’s Latis provides visibility into FLAG emissions at both the material and product levels. By understanding how FLAG emissions contribute to a product carbon footprint, you can explore increasingly innovative product opportunities.
Knowing where you source your ingredients and where the raw commodity was likely grown is the first step in understanding and managing your emissions at these stages.
The ability to identify products, materials and vendors with the highest FLAG emissions empowers you to strategically engage suppliers and target collection of primary data for the highest impact materials. This can begin relationships that make it easier to work together to implement reduction scenarios. It could also reveal that a supplier’s on-farm practices vary from what’s common in the area, resulting in a different total emissions calculation and possibly changing which reduction strategies will make the biggest impact.
From those starting points, Latis can then be utilized to model reduction scenarios and identify abatement strategies to bring the most efficient and meaningful impact reduction to FLAG emissions.
HowGood is an independent research company and SaaS data platform with the world’s largest database on food product sustainability. With more than 90,000 on-farm emissions factors for food ingredients, HowGood helps leading brands, retailers, suppliers and restaurants to measure, manage, and communicate their environmental and social impact.
