Pro Farmer Editors
After conducting a full review of the RFS Notice of Proposed Rulemaking from EPA, the Renewable Fuels Association (RFA) says they have some very serious concerns. Namely, the RFA says many of the models, data sets, and other pertinent pieces of information used by EPA for its lifecycle analysis are not publicly available, preventing third parties like the RFA from replicating and validating EPA's work. "Such transparency is absolutely critical so that constructive comments and peer reviews can be completed effectively," says the group.
Specifically, EPA's inclusion of the highly uncertain theories of international land use effects and indirect emissions in their greenhouse gas emissions calculation for biofuels makes the need for access to how EPA connected the dots that much more imperative, says RFA.
These concerns have prompted the RFA to send a letter to EPA requesting more information and access to the entire EPA modeling framework. The whole letter is available here.
The following are key points raised by the RFA:
Based on the information currently provided in the Notice of Proposed Rulemaking, Regulatory Impact Analysis, and supporting and related materials posted to docket EPA-HQ-OAR-2005-0161 as of May 29, 2009, third parties are unable to replicate the results of EPA's biofuels lifecycle modeling for proposed changes to the RFS program. While certain inputs, outputs, and documentation for independent models are made available on the docket, stakeholders are unable to assemble these disparate components in a way that would reasonably allow replication of the EPA modeling process.
It is important that stakeholders have the ability to replicate EPA's results for several reasons: 1.) to allow stakeholders to understand the structure of the modeling framework and the impact of certain parameters on the modeling outcomes; 2.) to validate the accuracy and precision of EPA's analysis; 3.) to allow stakeholders to perform independent analyses using alternative assumptions and inputs.
RFA's specific requests include:
- EPA should provide public access to the FAPRI system of models as configured for the RFS analysis. This should include all assumptions, inputs, price and demand elasticities used, outputs for all scenarios and sensitivity cases, and other pertinent information.
- EPA should provide public access to the FASOM model as configured for the RFS analysis. This should include all assumptions, inputs, price and demand elasticities used, outputs for all scenarios and sensitivity cases, and other pertinent information.
- EPA should provide the version of the GREET model it used for the RFS analysis along with all assumptions and inputs that differ from GREET defaults. Further, the ASPEN-based model used by EPA should be made available with documentation describing how the model was used to generate 2022-era performance assessments.
- EPA should provide detailed documentation describing exactly how all of the various models used (FAPRI, FASOM, GREET, ASPEN, etc.) were integrated and what adjustments were made to each to enable amalgamation. For example, there is no documentation available on how the emissions data provided by Winrock Corp. was applied to the FAPRI model results. A thorough description of how all of the inputs and outputs from each model were reconciled and harmonized is absolutely essential.
- EPA should provide the results of formal uncertainty analysis (if such an analysis was conducted) including confidence intervals and probability distributions. Uncertainty analysis is necessary so that decision makers can reasonably understand the degree of confidence they can place in the model results.