Taheripour et al. recently posted comments on their websites about our peer-reviewed study published in the Proceedings of the National Academy of Sciences (Lark et al. 2022). In their commentary, the authors question several components of our work, which they conclude “resulted in overestimation of the GHG emissions of corn ethanol.” We find Taheripour et al.’s conclusions to be unsupported and based upon several misunderstandings and misinterpretations of our methods and results. To help clarify, we offer the following replies which correspond to the 9 major points of Taheripour et al.:
1. The land use changes we identify specifically represent the conversion of pasture and Conservation Reserve Program (CRP) lands to crops and thus are likely to cause a large carbon debt upon conversion. In our study, we used USDA National Resources Inventory (NRI) data–not the Cropland Data Layer (CDL)–to estimate the types, amount, and regional locations of land converted to crop production due to corn ethanol and the RFS (see Lark et al. 2022, SI Appendix, ln 375-393). The USDA NRI dataset specifically identifies land as pasture or CRP, and such areas are known to result in a substantial carbon debt upon their conversion to cropland (Guo and Gifford 2002; Gelfand et al. 2011; Sanderman et al. 2017; Spawn-Lee et al. 2021). Taheripour et al. quote and appear to conflate our methods for estimating water quality impacts (Lark et al. SI ln 696) with our methods for identifying land transitions (SI ln 375) and also seem to misinterpret our use of satellite-based data, which we use only to infer at a higher resolution the biophysical characteristics of converted lands for our water quality and greenhouse gas modeling (SI ln 555). While the USDA NRI data allow us to identify the type, amount, and regional locations of land use changes, we do not know the exact location of those land use changes. Therefore, we used satellite-based data on land conversions only to help parameterize our water quality and greenhouse gas models at sub-regional levels.