Properly accounting for indirect land use change

If Treasury uses GREET to model SAF lifecycle greenhouse gas emissions, it must modify it to be both similar to CORSIA and to fully account for indirect land use changes 

Statutory overview 

Section 45Z requires that lifecycle greenhouse gas (GHG) emissions for SAF be determined using either “(I) the most recent Carbon Offsetting and Reduction Scheme for International Aviation [(CORSIA)] . . . or (II) any similar methodology which satisfies the criteria under section 211(o)(1)(H) of the Clean Air Act.” Clean Air Act (CAA) section 211(o)(1)(H) defines lifecycle GHG emissions to include the “aggregate quantity of greenhouse gas emissions (including direct emissions and significant indirect emissions such as significant emissions from land use changes).”  

In other words, if Treasury uses a methodology to calculate the SAF lifecycle assessment (LCA) other than CORSIA, the plain text of 45Z requires that methodology to both be “similar” to CORSIA and properly account for all direct and significant indirect emissions, including those from indirect land-use changes (ILUC), in accordance with the CAA. 

The GREET model is used for carbon accounting for several tax credits included in the Inflation Reduction Act, including the 40B SAF tax credit. Treasury created a specific version of GREET to use for the section 40B SAF tax credit, called “40BSAF-GREET.” But 40BSAF-GREET satisfies neither of the requirements listed above. To the extent Treasury opts not to use CORSIA, it must modify GREET for 45Z to ensure that it is both similar to CORSIA and fully accounts for ILUC in accordance with CAA section 211(o)(1)(H). 

Indirect land use change values for 40BSAF-GREET are inconsistent with other models 

First, 40BSAF-GREET is not “similar” to CORSIA in “how [it estimates] the indirect emissions attributable to biofuel policy, particularly on emissions from indirect land-use changes . . . triggered when more acres are put into biofuel production.” Specific and substantial differences between CORSIA and 40BSAF-GREET include: (1) 40BSAF-GREET relies solely on the GTAP-BIO model to estimate indirect emissions, while CORSIA uses both GTAP-BIO and GLOBIOM to account for differences in substitution effects¹; and (2) CORSIA amortizes indirect emissions over a 25-year period, while 40BSAF-GREET amortizes indirect emissions over a 30-year period, resulting in 16.7% lower indirect emissions.  

These differences in methodology between CORSIA and 40BSAF-GREET became apparent in the final guidance for the section 40B SAF tax credit, where Treasury used ILUC values for corn and soy that were significantly lower than the values used in CORSIA, therefore not satisfying the “similarity” to CORSIA requirement. Modeled “indirect effects” values in 40BSAF-GREET are approximately 11.1 g CO₂e/MJ for corn and 16.2 g CO₂e/MJ for soy,² which are 37-55% lower than the CORSIA ILUC defaults of 25.1 g CO₂e/MJ for corn and 24.5 g CO₂e/MJ for soy. As the graph below demonstrates, the ILUC values used in 40BSAF-GREET are outliers among several different models. 

Figure 1. This plot compares the ILUC values for corn and soy in the 40BSAF-GREET model compared to other models, including the EPA RFS 2010 (used for implementing the RFS), CA LCFS GTAP-BIO 2015 (used for implementing California’s LCFS), CORSIA GTAP-BIO 2019 and CORSIA GLOBIOM (both used together to determine CORSIA’s LCA values). The ILUC values in 40BSAF-GREET are clearly the lowest compared to any other model and are not “similar” to those in the CORSIA models. 

Final 45Z guidance must include updated ILUC numbers from those in 40BSAF-GREET to comply with the Clean Air Act  

In addition to not being similar to CORSIA, 40BSAF-GREET does not properly account for ILUC under CAA section 211(o)(1)(H). Before Treasury published finalized guidance for section 40B, EPA explained in a letter to Treasury that, in the context of 40B, “the only methodology that the EPA has determined satisfies the CAA section 211(o)(1)(H) criteria is the methodology, modeling, and analysis the EPA developed for its 2010 Final Rule regarding the RFS program and has applied in subsequent RFS rulemakings.” EPA further explained that, while “developers of the GREET model have added indirect land use change emissions into the model . . . the current version does not include all of the significant direct and indirect emissions that the EPA determined were necessary in 2010.”  Indeed, in the context of the RFS, EPA uses indirect emission values of 27.8 g CO₂e/MJ for corn and 33.6 g CO₂e/MJ for soy, which are 52-60% higher than the indirect emission values estimated through 40BSAF-GREET. 

Upon finalization of the section 40B guidance, EPA published a second letter concluding that 40BSAF-GREET had been updated to appropriately consider significant indirect emissions. But while 40BSAF-GREET includes ILUC emissions values, as the above graph illustrates, they are still precipitously lower than EPA’s own corn and soy ILUC values used in the RFS program, which is governed by the same statutory definition of lifecycle GHG emissions as both 40B and 45Z. Treasury’s 40B ILUC values are therefore doubly insufficient, satisfying neither the requirement to be based on a “similar” methodology as CORSIA nor allowable under CAA section 211(o)(1)(H). Treasury must correct these errors in its 45Z guidance by using ILUC values that are closer to CORSIA and EPA’s RFS values. 

See CATF’s full comments on 45Z for more information.