Kumar Venkat and Susan Cholette
Inside our soil model
We are seeing a surge of interest in life cycle assessments (LCAs) for determining the carbon footprints of food products. In many food LCAs, attention turns rather quickly to the agricultural practices used to produce the ingredients. This is especially the case with producers of processed food products who source ingredients grown using organic and/or regenerative methods. We summarize here what is in our soil model as implemented in FoodCarbonScope and CarbonScopeData, how we model soil dynamics and what the standards say about all this.
The diagram above captures the flow of key inputs into the soil and the resulting emissions of greenhouse gases (GHGs) as well as sequestration of soil organic carbon. The inputs of interest are synthetic fertilizers, organic fertilizers and soil amendments, and crop residues. The soil model basically tracks, to a first-order approximation, what happens to the nitrogen and carbon in these various inputs. The GHGs that are released as a result of these inputs, as well as due to land management practices and climatic conditions, are nitrous oxide (N2O), carbon dioxide (CO2) and methane (CH4).
How we model soil dynamics
Changes in soil organic carbon
We use the IPCC tier 1 parameters (see discussion in the last section) to estimate changes to soil organic carbon (SOC) stocks in croplands. These parameters are a function of land management practices (tillage method, carbon inputs to soil), climate zone, moisture regime, and soil type. The default time period for stock changes is nominally 20 years and management practice is assumed to influence stocks to a depth of 30 cm. Soil carbon is considered to be in steady state until there is a significant land-use change (such as a conversion from grassland to cropland) or management change (such as a conversion from conventional cropland to organic cropland) that changes the soil carbon stocks.
When such a change occurs, the soil carbon is assumed to reach steady state again after 20 years under the new land-use or management practice. During this 20-year transition period, soil carbon may be increasing or decreasing each year as illustrated below, thus increasing or decreasing the total greenhouse gas emissions and the product carbon footprint during that period. When soil carbon is in steady state, it does not contribute to net emissions in agricultural systems.
Regarding the question of why a steady-state organic farming system does not get credit for the potentially higher level of carbon (relative to a conventional farming system) already accumulated in the soil: The addition of organic matter balances the carbon that is naturally oxidized away from the top layers of the soil, so in an ideal steady-state system, the SOC level remains constant and the net change in soil carbon is zero. LCA models account for this net change and a production system can take credit in the years that this net change > 0. Likewise, if we allowed the organic system to degrade by not adding sufficient organic matter to the soil for a few years, it would result in a new transition with net change < 0, resulting in a higher carbon footprint for those years. So maintaining a steady-state organic system avoids this higher carbon footprint — and that in a nutshell is the benefit of continuously adding organic matter to the soil. That benefit is already reflected in the stable carbon footprint of a steady-state organic system.
The 20-year transition period itself is not a critical parameter. Our tools allow the user to set this to any reasonable value. The more important issue is the magnitude of changes to the SOC stocks over the transition period. There can be wide variability and crop specificity in field measurements of soil carbon stocks, and the tier 1 parameters provide useful estimates of typical SOC changes while avoiding the complexity and data collection effort associated with tier 2/3 models.
Nitrous oxide emissions due to nitrogen inputs
Synthetic nitrogen fertilizers (nitrate, ammonia, ammonium or urea) and organic nitrogen sources (compost, manure or crop residues) produce direct N2O emissions from the soil through the nitrification-denitrification process mediated by soil bacteria. In addition, indirect N2O emissions result from the volatilization and redeposition of NH3/NOx, and through leaching and runoff of nitrogen.
We calculate both direct and indirect N2O emissions based on the amounts of synthetic and various organic nitrogen sources added to the soil each year, as well as due to nitrogen from crop residues and distinguishing between typical agricultural soils and flooded rice fields. We also make an adjustment for legumes to account for the excess ammonium that may leak from nitrogen-fixing root nodules and ultimately escape to the atmosphere as N2O.
Carbon dioxide emissions due to lime and urea
CO2 is released directly due to the application of lime and urea to agricultural soils. Liming is used to reduce soil acidity and improve plant growth. When carbonates such as limestone (CaCO3) are added to soils, they dissolve and release bicarbonate (HCO3-) which evolves into CO2 and water. When urea is applied as a fertilizer, it releases bicarbonate which again evolves into CO2 and water. We calculate these CO2 emissions based directly on the amounts of lime and urea applied each year to a given land area.
Methane emissions from rice fields
Anaerobic decomposition of organic material in flooded rice fields produces CH4. We calculate the annual amount of CH4 emitted from a given area of rice as a function of the crop growth period (measured in days), irrigation method (such as continuously or intermittently flooded), and organic and inorganic soil amendments (such as straw, compost and/or manure).
What the standards say
IPCC guidelines
The IPCC guidelines for national greenhouse gas inventories provide the most detailed and standardized guidance available for calculating soil carbon emissions and sequestration, N2O emissions from soils, and CH4 emissions from soils. Given that the current product carbon footprint standards do not provide detailed guidance on these topics but defer to IPCC in general, our methodology for modeling soil dynamics is based almost entirely on the IPCC guidelines.
There are three tiers of methodology in the IPCC guidelines. We have found the tier 1 methods to be the most appropriate for LCA models used to compute product carbon footprints. Given the time and cost constraints in most LCA projects, as well as the sheer difficulty involved in obtaining country-specific data or field measurements, we use tier 1 methods as a practical default unless higher tier data are readily available.
Here is a brief description of the three tiers:
- Tier 1 methods are designed to be the simplest to use, for which equations and default parameter values (e.g., emission and stock change factors) are provided by IPCC. Country-specific activity data are needed.
- Tier 2 can use the same methodological approach as Tier 1 but applies emission and stock change factors that are based on country- or region-specific data.
- At Tier 3, higher order methods are used, including comprehensive field sampling repeated at regular time intervals and/or GIS-based systems of age, class/production data, soils data, and land-use and management activity data, integrating several types of monitoring.
Product carbon footprint standards
Neither of the two leading product carbon footprint standards, PAS-2050 and the GHG Protocol Product Standard, requires soil carbon changes (due to changes in management practices such as tillage) to be included in the carbon footprint of agricultural products. The default is to exclude it, but both standards provide for ways to include it (in the case PAS 2050, it can be included in accordance with the standard’s supplementary requirements). Both standards do require direct land use change (such as conversion from grassland to cropland) to be included per IPCC guidelines, but indirect land use change is not included in the carbon footprints. Our methodology is consistent with both standards, and we do include the effects of both land management practices and direct land use changes.
PAS 2050 requires non-CO2 emissions (N2O, CH4) from soils to be assessed using the highest tier approach set out in the IPCC guidelines or the highest tier approach employed in the country in which the emissions are produced. The Product Standard does not provide specific guidance on this. Our methodology takes a compromise position here and defaults to using the IPCC tier 1 parameters to calculate these soil emissions.
Carbon registries
The methodology described here generally aligns with the principles developed by carbon registries (for example, the Climate Action Reserve’s Soil Enrichment Protocol) for allocating carbon credits based on soil carbon sequestration.