Review of the U.S. Farmers and Ranchers in Action White Paper on Building a Scientific Roadmap to a Carbon-Negative Agricultural System (2024)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Appendix A Specific, Editorial, and Minor Chapter Comments CHAPTER 1: DEFINING THE NEED FOR ACHIEVING A CARBON-NEGATIVE AGRICULTURE Specific Comments on the Text L5: Please include a word or two on the “demand.” L7–8: Provide a reference for “over the past 10,000 years, human activities have transformed the Earth’s….” L17–18, 19: Update greenhouse gas (GHG) emissions data with the latest Intergovernmental Panel on Climate Change (IPCC) report, here and elsewhere. L19: With respect to the need to “reduce annual emissions of long-lived GHGs (CO2 and N2O),” why is methane (CH4) not included? Although it has a shorter half-life, studies say its control is important. Also, should “annual emissions” be net annual emissions? L20: Include a definition for the concept of what you mean by a “meta-system” which you also refer to in L36–37. L24: Be more specific on how buildings can be considered a sector that emits GHGs. L27–29: Update GHG emission data with the latest U.S. Environmental Protection Agency (EPA) report, here and elsewhere. L28: The phrase “after land sequestration” might be better as after increased land sequestration. It might be good to discuss nature of emissions from historic land use change and the possibility of reversing some. L50–53: Can the daily/annual cost of a nutritious diet in the U.S. be added here? The text should be more direct if it is saying that it would not be desirable to increase food costs when reducing GHGs. L54–60: Here the text reports 2020 and 2021 grain and other product yields and later introduced some forecasted yields, omitting the year for which they are forecasted—maybe the year 2022? Please update the information. L65–67: Please clarify if the net cumulative emissions are global or domestic. L68–71: Is this text comparing total annual anthropogenic GHG emissions from agriculture in 2017 with a global food system GHG emissions? The discussion included on L72 forward suggests that. Consider revising this part, as GHG emissions from agriculture will most likely be included in the food supply chain systems calculation. L77–80: This paragraph misses a chance to talk about interrelatedness of agriculture sectors when it states “For example, cereals are responsible for the highest fraction of global total land use GHG emissions (47.4%), while chicken is only responsible for 0.6% (Table 1, after Hong et al., 2021). This does not mean all humans should only eat chicken.” It could go on to explain that chickens eat cereal crops and make a reference to life cycle analysis. L87: Is the “food system” referenced the same as the one stated on L40 as the “food supply chain metasystem”? It will be better to stick to one term, and if shortened to “food system,” added to L40. L88: The text notes that “The total proportion of GHG emissions from the food system has decreased 10% since 1990 (Crippa et al., 2021).” But that paper notes that the reason the proportion has decreased is because emissions in other sectors have grown. You might indicate if agricultural emissions have decreased or grown slower, for more context. 15 

Review of White Paper on a Scientific Roadmap to Carbon-Negative Agriculture L90: You discuss “rapid increase of electricity and other energy sources” without reference or context. What are those? Solar? Wind? Manure? It could influence if emissions are net positive or negative. L92: Having noted that the rapid increase of electricity is a challenge for emerging economy agricul- tural production, the text could use a few more words to make the “inter-relationship between emerging renewable energy and agricultural GHG emissions” clearer. L99: A reference is needed for “CH4 has a GWP of about 30 over a 100-year time span (GWP100).” L102–111: The text notes that CH4 has 10-year lifetime, so why present results only in terms of Global Warming Potential (GWP) over 100 years (GWP100)? As pointed out in L103, the shorter lifetime of CH4 means that it has a much higher Global Warming Potential over 20 years (GWP20) (about 80 according to the International Energy Agency) than GWP100, indicating that near-term mitigation of methane (CH4) emissions will have a greater impact on slowing climate change during the next two decades than will mitigating carbon dioxide (CO2) or nitrous oxide (N2O). Slowing warming during the next two decades is crucial to slow positive feedback to climate change from natural sources, such as GHG emissions from permafrost and wetlands, so expressing results as GWP20 in addition to GWP100 would help readers un- derstand how effective CH4 mitigation could be. This is particularly important for agriculture, because live- stock is the largest single source of global anthropogenic methane emissions. In addition, some discussion of GWP* (a specific treatment of the GWP for short-lived, cyclical CH4 emissions) is in order, so that readers become better informed about this controversial topic. L110: The text should confirm which timespan or basis is used for the CO2-equivalents (CO2-eq) used throughout the white paper. Do you use the GWP100 or something else? L112–116: This paragraph needs additional explanation: “Globally agriculture would have to se- quester between 14 and 18 Gt CO2-eq to be GHG negative out of the total emissions of 52 Gt CO2-eq. For the U.S. agricultural sector, total GHG sequestration would have to reach approximately 0.60 Gt CO2-eq to offset all emissions from all agricultural systems.” Additionally, what exactly is meant here by “GHG sequestration”? This may be confusing to many readers, so it would be good to clarify if it means soil carbon (soil C) sequestration as well as reductions of emissions of CO2, CH4, and N2O. The nomenclature needs more clarity here and consistency throughout the document. In addition, as argued later in the white paper, sequestration will hit a new equilibrium or saturation point, so the assumption that sequestration will outstrip emissions in the long run is suspect. L119–120: “The global boundary for annual GHG emissions was estimated to be 9.8 Gt CO2-eq, a more than 43 Gt CO2-eq reduction from 2015 baseline emissions. Total agricultural emissions are about one-third of the target emissions level.” The juxtaposition of “boundary” with “target” is confusing. Perhaps global maximum is a better word than boundary. The reduction of GHG emissions that agriculture can provide needs to be more clearly stated relative to the C negative (below net-zero) goal. You seem to indicate that if agriculture went to net zero that would reduce emissions by 14 Gt CO2e (one-third of 43?) leaving 29 Gt CO2e above that which must be reduced. This needs to be explained and written more clearly. L126–134: While the last chapter of the white paper makes the case that neutral or negative emissions are plausible for the U.S. under an ambitious adoption scenario, it does not follow that this is true globally as put forward in L126. Indeed, L113–116 show that the sequestration would need to be about 25 times greater (14–18 Gt vs. 0.6 Gt) for the globe vs. the United States. The statement “Sequestering enough carbon to offset all anthropogenic emissions appears to be plau- sible” goes too far, as “all anthropogenic” means all sectors. Also, the text is opaque about whether this situation would be economically possible or would persist for 20 years. As referenced in the text, Sadatshojaei et al., 2021 suggests far-reaching requirements to achieve de- sired or necessary levels of soil C sequestration, and there are several publications here that give optimistic assessments of soil C sequestration potential. Rather than present the upper limit (“as high as 2 Gt per year”) this report should present a more balanced view. For example, working group III of the IPCC AR6 con- cluded with medium confidence that the mean estimate for the global technical mitigation potential of cropland C sequestration is 1.9 (range 0.4–6.8) GtCO2-e yr−1, but with only 0.6 (range 0.4–0.9) GtCO2-e yr−1 likely to be realized at market prices of USD100 per ton CO2-equivalent (IPCC 2022a). This range is 16 

Appendix A not much different from Sadatshojaei’s reported range, but the “likely” value is toward the lower end, which should be acknowledged here. IPCC, 2022. Climate Change 2022: Mitigation of Climate Change. Contribution of Working Group III to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change. [P.R. Shukla, J. Skea, R. Slade, A. Al Khourdajie, R. van Diemen, D. McCollum, M. Pathak, S. Some, P. Vyas, R. Fradera, M. Belkacemi, A. Hasija, G. Lisboa, S. Luz, J. Malley, (eds.)]. Cambridge University Press, Cambridge, UK and New York, NY, USA. doi:10.1017/9781009157926. L137–148: The process categories of “soil carbon sequestration, nitrogen and water use efficiency” do not include agriculture as a provider of energy through biomass, and wind and solar energy, which may hold greater potential for GHG emission reductions than what is listed. Shouldn’t enteric fermentation and manure management be considered process categories? Do animals need some mention in the prior paragraphs above as a possibility where reductions are possible? There might also need to be talk about calorie and protein sources as some advocate diet changes away from animal proteins. The statement about enteric emissions is vague, as much is known about some aspects of these emis- sions and less about others. What, specifically, “is not well understood at the global level”? L62–164: This document may only provide a “roadmap” in a very general sense. Roadmaps are more useful when they identify key bottlenecks or impediments to specific paths that have been identified to achieve particular goals and suggest ways to overcome or avoid those impediments. In addition, it is not clear that that the white paper provides a prioritized list of needs for both research and implementation. The white paper mostly focuses on the research needed before implementation can move ahead at scale. A better articulation of the purpose could help identify the conclusions that should be drawn at the end and will also allow for readers’ expectations to be in line with what they will receive if they read the entire report. Table 1: It should be noted that Table 1 only includes land use emissions for agriculture. Also, the total (24.8 % of total anthropogenic GHG emissions) in the table is much larger than figures cited in the text. Inclusion of wood products accounts for only part of the difference. Editorial and Minor Issues L36: Editorial: A new paragraph is needed, starting with “The global food supply….” L40: Change “The global food metasystem” to “The global food supply chain metasystem”. In that way, the reader can be related to the previous paragraph. L41: Misses a reference. L43–44: Should the text be has at the end L43? Delete the word “expansion” in the next line? Also, it is redundant to have “global” and “to all parts of the world” in the same phrase. L46: What does it relate to the 8.9% of people? Is it related to the global population? If yes, include here and add a reference to the latest global human population report. L49–50: Update the numbers to the latest United Nations or Food and Agricultural Organization re- ports. L54: Odd word—robust. L55: Demand tightened—what is that? L62: Odd phrase—”technological co-option.” L57: When are coarse grain and rice yields projected to reach record yields? Also, add a reference. L72: These differences—Do you mean these ranges? L76: “Should be interpreted with caution”— A suggestion is to add a general reason why in the phrase and expand on it later. Not sure you need to use “global climate catastrophe.” Perhaps better to be less alarmist and say this would be failure of the Paris agreement and its goals. 17 

Review of White Paper on a Scientific Roadmap to Carbon-Negative Agriculture CHAPTER 2: CHALLENGES AND OPPORTUNITIES FOR CLOSING THE ROW CROP YIELD GAP Specific Comments on the Text L17: You focus on “genetic potential of a crop cultivar or hybrid” but climate change adaptation could suggest the need for attributes other than those involved in yield, such as varieties that have bigger roots or resilience characteristics, lower fertilizer demands, etc. The focus on yield gap is limiting in this white paper that focuses on emissions reduction and sequestration. L28: The text notes that there is “the opportunity to utilize county and state level yields available from the NASS databases” but that farm level yield records are often not available. The uses and limitations of these records should be better developed in the text. L30: Since the text is exploring what is there in the literature to assist the process of yield gap analysis, it might consider the economic concept of frontier production functions and the statistical explanations of deviations from those. L33: When you discuss “crop model assumptions” you might also mention such models often do not have pests or much modeling of extremes and other forces. L47: This chapter refers to “carbon neutral systems” but Chapter 1 focused on carbon negative. Per- haps low-C emitting might be a better term. Regardless, the terminology should be consistent between chapters. L60: The text seems to focus on CO2 sequestration but what about reducing other CO2e emissions as opposed to just C capture. L64: This section should highlight how improvements in crop yields have had tradeoffs in root-to- shoot ratios, and how this impacts the carbon neutrality of the whole system. This needs to be recognized as a challenge. L65: Regarding “genetic variation and test[ing] transgenic approaches for improving traits that deter- mine yield potential (Yp) and enhance resilience to abiotic and biotic stresses,” there may well be plants with traits not necessarily connected to yield specifically but have other advantages in adapting to climate change. L80: Just above L75 you have an equation that involves three terms none of which are defined in terms of absorption of CO2 from the air with the carbon being stored in the plant. But that in the next sentence you say something about “efficiency of carbon assimilation.” It would be useful to crosswalk the equation terms with the text. L90: You introduce an important point when you say, “Some strategies to improve photosynthetic efficiency have the added theoretical benefit of improving nitrogen and water use efficiency.” But more generally since the thrust of this whole paper is on reducing net emissions it would be useful to say that is important that one should not increase the emissions of a crop from other sources (like fertilizer) or at least say something about increasing the carbon efficiency of the crop by reducing emissions per unit of product produced. L163: Low carbon or carbon negative or carbon neutral? L166: “Genetic improvement continues to increase the Yp of most cultivated crops”—crops such as? L167: Provide a brief list of improved management practices that can increase or attain the yield potential of the most cultivated crops. L170: There is some inconsistency being introduced here with statements like “For instance, manage- ment accounted for 44–77% of yield variability of winter wheat in the US. Great Plains” whereas earlier in the document weather was said to be responsible for most of the yield gap. L201–210: The explanation of the effect of time of planting relative to outcomes in the fall is a little confusing. Isn’t earlier planting recommended to exploit the key peak solar radiation in summer to match it to the time that the crop needs for maturation? Should there be some discussion here that climate change is lengthening the growing season, potentially moving the onset of the planting season to earlier in the year? What do the tradeoffs of fall versus summer planting mean overall? 18 

Appendix A L212–217: Maximum yield is not the only objective of farmers. Double cropping, while not producing the maximum yield of the crop, may nevertheless help increase income. L228: Define soil quality or update the phrase with the more commonly accepted term, soil health. L230: Please explain what you mean by “suppression of microorganisms that could be a sink for crop photosynthate,” and how this could benefit carbon neutrality. L231: Given the stated purpose of the document it would be nice to include more context on the statements like “These factors will become the foundation for a carbon neutral agriculture because each of these factors increase the ability to store carbon in the soil and remove it from the atmosphere.” In a more non-technical sense what the text is saying is yield improvement leads to larger amounts of C stored in the plant and probably the roots and larger amounts of organic material being returned to soils. I think more of a line needs to be drawn between improving yields and improving soil C. L261: If possible, it would be good to reference a couple literature reviews and perhaps treatments like the ones Pete Smith et al. did: Smith, P., D. Martino, Z. Cai, D. Gwary, H. Janzen, P. Kumar, B. McCarl, et al. Greenhouse Gas Mitigation in Agriculture. 2008. Philos Trans R Soc Lond B Biol Sci 363, no. 1492 (Feb 27 2008):789-813. https://doi.org/10.1098/rstb.2007.2184. L303–318: Bring all this information about yield increase, nutrient use, etc., into the context of reduc- ing net emissions. L322: Yield maximization is not all “Producers must ensure adequate but not over-fertilization to optimize production.” Many producers use over-fertilization as a risk management method, and many do not strive for the highest level of yields but do care about profitability. This chapter takes a narrow perspec- tive on farmer decision-making. L381: “The expectation of these products is to close the yield gap while minimizing environmental losses by synchronizing nutrient release with crop uptake and soil water availability.” Is it clear that these products address the yield gap? They may lower the cost of providing the fertilizer without changing the yield gap at all. L410: The statement that “adjustments in the animal-to-land ratio will be required to distribute carbon widely to balance nutrients” needs elaborating in a way that the general audience can understand it. L411: One needs to be careful about cost here as the weight of the amount of material that must be applied to achieve a given amount of nutrient application with manure could be hundreds of times more than with chemical fertilizers and this generally reduces the affected service area to be a fairly narrow one around the availability of supplies of manure. This would also have consequences for the energy use em- ployed in hauling the manure to the application location. L412–435: This paragraph should be moved to before the description of the 4Rs. The section on C neutrality briefly mentions the role of soil organic carbon but could elaborate on specific practices or sys- tems that contribute to increasing soil C. Providing more insights into the practical implementation of C- neutral practices would be valuable. L420: What about profits? L421: The statement that research on “evaluating increases in soil C have focused on no-tillage prac- tices and the associated increase in soil surface residue” is a somewhat narrow crop perspective. There also has been soil C research related to the effect of perennials versus annuals, and the effect of going from crops to trees or grasslands. L426: The nitrogen use for carbon enhancement was subject to mental dialogue years ago where Rat- tan Lal was suggesting this approach and William Schlesinger wrote some comments indicating the emis- sion effects of manufacturing the nitrogen and the associated nitrous oxide made this a more complex prob- lem: 19 

Review of White Paper on a Scientific Roadmap to Carbon-Negative Agriculture Schlesinger, W.H. 2000. Carbon sequestration in soils: some cautions amidst optimism. Agricul- ture, Ecosystems & Environment 82 (1-3): 121-127. https://doi.org/10.1016/S0167-8809(00)002 21-8. L435: It would be nice to include the concept of profits here in addition to costs, time, and environ- mental impacts of nutrient management. L452: The paper states that management can be responsible for a much larger proportion of the yield variability than genotype in cropping systems with large yield gaps but above there were statements that the yield gap was small in irrigated fields and, in reference to a paper by Hatfield, that it was mainly caused by weather variations. There needs to be more clarity in the text with respect to the factors of variations. L457: There is a mention of genetic improvement for “resilience to environmental stresses” but this section of the report did not build much of the case for research needs on this, though it is a major emphasis of the seed companies. There is also no mention of pests and how climate change maybe exacerbating their negative impact on yields. L467: The conclusions section makes the case that a priority should be on rotations with new crop species and legume cover crops “(i.e., coupling carbon and nitrogen cycles) as these result in the largest benefits to the system” but there is not a significant discussion of this up to this point in the chapter. L478: Field experiments and remote-sensing information are important. What about crop modeling? Editorial and Minor Issues L10: The wording “places crop performance into a biophysical and statistical evaluation of historical yields” and, in particular, the word “places” does not fit well here. Do you somehow mean allows one to analyze crop performance in such frameworks? L188: The hierarchy of the subheadings are confusing. They seem to be at the same level as the major headings. L244: Perhaps insects and diseases also? L255: It would be nice to allude every now and then to profit as opposed to pure yield maximization. L291: The authors might mention data mining efforts where one takes such information and does some form of regression to see where there are critical factors that influence the outcomes and allow one to develop hypotheses on factors meriting further investigation. L303: There seems to be a denominator unit missing here like per acre or perhaps hectare. L308: After “4R management” add as the application of the right nutrient rates…. When something very specific is mentioned like 4R nutrient management it might be good to include a reference. The target audiences for this will be people besides those on the inside in crop agriculture. L320: Some words are missing in “soil test P or K needs exist.” L350: “Fused”? L360: “Fertilizer applications are matched more closely” seems to ignore time delayed application forms. L379: Be consistent throughout the chapter and white paper in using symbols for elements. L391: Cost efficiency may also be a factor as may strategies like fertigation. L473: It would be better to say coupled with data mining and statistical analysis. CHAPTER 3: SOIL CARBON SEQUESTRATION ON U.S. AGRICULTURAL LANDS Specific Comments on the Text L8: This section is called Soil Carbon Management Research but is more background information than a research discussion. The section could use some stage-setting sentences for the chapter and the white paper by stating that: This chapter discusses one way of adding a carbon-negative contribution to general agricultural emissions by sequestering more carbon in soil. 20 

Appendix A L9: The amount of the planet’s land area dedicated to agriculture needs to be reconciled with the Chapter 1 info, which says 50% (L34). L12: Please check the reference, as the information is more likely from Sanderman et al (2017, PNAS). L14: Do all these activities, everywhere, continue to deplete soil, or should the statement be qualified, as “in places” or “on net”? L16: Is urban encroachment a factor to include? L18: Consider including in here information of historical anthropogenic soil erosion in the Midwest USA, as an example from Thaler et al. (2022). Thaler, E. A., J. S. Kwang, B. J. Quirk, C. L. Quarrier, and I. J. Larsen. 2022. Rates of Historical Anthropogenic Soil Erosion in the Midwestern United States. Earth’s Future 10 (3):e2021EF002396. https://doi.org/10.1029/2021EF002396. L19: Protecting stocks does not help “drawdown atmospheric CO2 concentrations” but it does help avoid emissions. L21: Although large reductions in emissions are needed, it is possible to emit at very low levels and still stabilize climate. L24: Storage can also be in trees or biochar. L26: Negative carbon technologies….to include carbon capture and storage as another strategy beside soil. L28: Soil C sequestration has been a major area of research focus for some time. L44: Which countries have the desired inventory system and what can the U.S. learn from their suc- cesses? Are they primarily small, wealthy, western European countries or are there other examples too? How can small country success be scaled up to a continental-scale country like the U.S.? L64: You might say something about examining the challenges and maybe their alleviation. L71: Rather than say “best” management practices (BMPs), you might say altered, as these practices target C sequestration specifically. Given current economic signals one might argue that farmers are doing best management for them. Again, there is a difference between best carbon management and best profita- bility management. L114–119: This is theoretically true, but do we really know if carbon saturation has ever been achieved and can we reliably estimate what a soil’s saturation potential is? L123: Should the discussion also include adding more stable carbon forms like biochar? L129: Please define what is meant by regeneration in this context. L142–144: Here the text lists the 6 soil health principles which were proposed by Natural Resources Conservation Service (NRCS) (https://www.nrcs.usda.gov/conservation-basics/natural-resource-con- cerns/soils/soil-health). Why is General Mills referenced? The list should be reorganized and listed as “un- derstand agroecosystem context” first, followed by the others. Also, check the Figure 2 legend, and change “ecosystem” to agroecosystem. L173: No-till is not always more profitable, tends to be more risky, has a long history of technical scientists saying it is better but depends critically on weed treatments that are beginning to see herbicide resistance. It does lower energy costs and manages water better that may explain places it has been adopted. Famers like John Bennett in Saskatchewan do not want to adopt due to risk but rather want to lease for a time to avoid risk. Bennett, J. E. and D. Mitchell. 2002. Emissions Trading and the Transfer of Risk: Concerns for Farmers. Bennett, J.F. A Commentary on Marland, G., B.A. McCarl, U.A. Schneider, Soil carbon: Policy and Economics. in Carbon Sequestration in Soils: Science, Monitoring, and Beyond. Proceedings of the St. Michaels Workshop. edited by N.J. Rosenberg, R.C. Izaurralde, E.L. Malone, pp. 168- 170, St. Michaels, MD, December 3-5, 1998. 21 

Review of White Paper on a Scientific Roadmap to Carbon-Negative Agriculture Klemme, R. M. 1985. A Stochastic Dominance Comparison of Reduced Tillage Systems in Corn and Soybean Production under Risk. American Journal of Agricultural Economics 67(3):550-556. https://EconPapers.repec.org/RePEc:oup:ajagec:v:67:y:1985:i:3:p:550-556. L201: You might add the words carbon-stable in describing biochar. L212-214: Some discussion is needed of where the biomass feedstock for biochar production would come from to scale it up for a nationally or globally significant effect on soil C sequestration. Due largely to limited feedstock, Schlesinger, and Amundson (2018) suggest that a realistic estimate of net C seques- tration using biochar on croplands is 0.18 GtC. Schlesinger, W. H., and R. Amundson. 2019. Managing for soil carbon sequestration: Let’s get real- istic. Global Change Biology 25(2):386-389. https://doi.org/10.1111/gcb.14478. L214: Carbon pricing is mentioned here for the first time in the white paper. This should appear earlier along with some review on social cost of carbon. L224: Sometimes there can be too much cover, for example, excess corn stover can interfere with no- till (Babcock 2008). So, this is double edged sword. Babcock, B. A. 2008. Breaking the Link between Food and Biofuels. Center for Agricultural and Rural Development (CARD) Publications 08-bp53. L216–225: The subtitle for this section puts residue retention first, but the rest of the section discusses mainly cover crop affects. Make the distinction between residue and cover crops more prominent, consid- ering that excess residue can make no-till difficult in some cases. L225–235: It would be worth expanding information about cover crops and their impact on soil or- ganic C accrual. Although numbers from several meta-analyses are presented, this chapter should dive into the conditions under which cover crops are successful (time, management, and rotations), some of which are briefly mentioned on L239 and after. L235: You might mention yield effects of these actions, for example, a paper by collaborators with David Lobell. Others may have done literature reviews on effects of residue retention. Seifert, C. A., G. Azzari, D. B. Lobell. 2018. Satellite detection of cover crops and their effects on crop yield in the Midwestern United States. Environmental Research Letters 13(6):064033. https://doi. org/10.1088/1748-9326/aac4c8. https://dx.doi.org/10.1088/1748-9326/aac4c8. Kaye, J. P. and M. Quemada. 2017. Using cover crops to mitigate and adapt to climate change. A review. Agronomy for Sustainable Development 37(1):4. https://doi.org/10.1007/s13593-016-0410-x. https://doi.org/10.1007/s13593-016-0410-x. Lobell, D. and N. Villoria. 2023. Reduced benefits of climate-smart agricultural policies from land- use spillovers. Nature Sustainability 6:1-8. https://doi.org/10.1038/s41893-023-01112-w. L263: Good to see grazing lands mentioned here (and surprised it is not mentioned in earlier chapters). However, grazing lands might be very different than those used for cropping, from the perspective of inte- grated crop and livestock systems, so that need to be distinguished clearly in the text. L276: On grazing there are some major literature reviews you could cover by H. Wayne Polley and colleagues that go beyond this chapter’s coverage of the topic. Joyce, L. A., Briske, D. D., Brown, J. R., Polley, H. W., McCarl, B. A., Bailey, D. W. 2013. Climate change and North American rangelands: Assessment of mitigation and adaptation strategies. Range- land Ecology & Management 66(5):512-528. 22 

Appendix A L276: Is there a need for any mention of manure here? L277–281: Integrating animal and crop production systems, including grazing of cover crops, is a central tenet of regenerative agriculture, but there appear to be major impediments to its adoption. Many of today’s farmers don’t have expertise in managing livestock and are reluctant to take on those risks. The premium niche market for meat grown from regenerative agriculture is rather limited. Some discussion of the reasons that this seemingly attractive idea is not being widely adopted in the U.S. would be helpful for readers who are not aware of these issues. L301–311: The text needs to include references that can support all these claims. L307–311: Perhaps there isn’t sufficient room in this section to expand on the socioeconomic chal- lenges and cost of adoption of new practices, but it will be worth discussing those issues, and mentioning the cost-share programs farmers have currently available, e.g., Natural Resources Conservation Service, Practical Farmers of Iowa, industry. L311: The text should add that yield and thus profits may be less but that carbon sales could provide a source of income. L324–325: This would be a good place to provide metrics on the status of cover crop adoption in the U.S., no-till, and biochar, as well as the socioeconomic challenges that prevent farmers from rapid adoption. L325: Isn’t this section missing issues about profitability? Risk? Carbon payments? Monetizing se- questration benefits? L376–381: This part of the text would be improved by adding information about acres under no-till and acres under cover crops. When dealing with no-till it might be worth covering concepts of continuous versus periodic, as much of the Midwest is no-till soy followed by conventional corn and this compromises sequestration. L384–388: Try to avoid using the word “aggressive” or be more specific about its meaning here. Your modeled results might be cross-referenced with data from long-term field experiments published by the Center for Advanced Bioenergy and Bioproducts Information (CABBI) and the Great Lakes Bioen- ergy Research Center (GLBRC). The data you presented here as ‘aggressive scenario’ and ‘moderate’ are not clear in terms of which parameters you considered when running these land conversions, nor did you mention which locations or land sizes you chose to run COMET-Farm. There is a need to provide the basic info about these simulations. Perhaps it can be assumed that you used the best management practices for an average farm size and presented them as a summary in Tables 2 and 3, but it is difficult reconciling this table with the main text. This section should be revised in a way that lets the reader know how these simu- lations were run. There is a plethora of good work from GLRBC and CABBI about cropland conversion to bioenergy fields, and they distinguished the impact between an annual biofuel crop and a perennial bioenergy crop. I suggest reading: Gelfand, I., S. K. Hamilton, A. N. Kravchenko, R. D. Jackson, K. D. Thelen, G. P. Robertson. 2020. Empirical Evidence for the Potential Climate Benefits of Decarbonizing Light Vehicle Transport in the U.S. with Bioenergy from Purpose-Grown Biomass with and without BECCS. Environmental Science & Technology 54(5):2961-2974. https://doi.org/10.1021/acs.est.9b07019. Mosier, S., S. C. Córdova, G. P. Robertson. 2021. Restoring Soil Fertility on Degraded Lands to Meet Food, Fuel, and Climate Security Needs via Perennialization. Frontiers in Sustainable Food Systems 5. https://doi.org/10.3389/fsufs.2021.706142. Sanford, G.R., R.D. Jackson, Y. Rui, C. J. Kucharik. 2022. Land use-land cover gradient demonstrates the importance of perennial grasslands with intact soils for building soil carbon in the fertile mollisols of the North Central United States. https://doi.org/10.1016/j.geoderma.2022.115854. 23 

Review of White Paper on a Scientific Roadmap to Carbon-Negative Agriculture Zhang, X., T. Lark, C. Clark, Y. Yuan, S. LeDuc. 2021. Grassland-to-cropland conversion increased soil, nutrient, and carbon losses in the U.S. Midwest between 2008 and 2016. Environmental Research Letters 16:054018. https://doi.org/10.1088/1748-9326/abecbe. Grace, P. R. and G.P. Robertson. 2021. Soil carbon sequestration potential and the identification of hotspots in the Eastern Corn Belt of the United States. United States. https://doi.org/10.1002/saj2.20 273. https://www.osti.gov/servlets/purl/1804296. Kasanke, C. P., Q. Zhao, S. Bell, A. M. Thompson, K. S. Hofmockel. 2021. Can switchgrass increase carbon accrual in marginal soils? The importance of site selection. GCB Bioenergy 13:320-335. https://doi.org/10.1111/gcbb.12777. Martinez-Feria, R., and B. Basso. 2020. Predicting soil carbon changes in switchgrass grown on mar- ginal lands under climate change and adaptation strategies. GCB Bioenergy 12:742-755. https://doi. org/10.1111/gcbb.12726. https://ui.adsabs.harvard.edu/abs/2020GCBBi..12..742M. Spiesman, B. J., H. Kummel, R. D. Jackson. 2018. Carbon storage potential increases with increasing ratio of C(4) to C(3) grass cover and soil productivity in restored tallgrass prairies. Oecologia 186(2):565-576. https://doi.org/10.1007/s00442-017-4036-8. Tiemann, L. K., and A. S. Grandy. 2015. Mechanisms of soil carbon accrual and storage in bioenergy cropping systems. GCB Bioenergy 7. Bandaru, V., R.C. Izaurralde, D. Manowitz, R. Link, X. Zhang, and W. M. Post. 2013. Soil carbon change and net energy associated with biofuel production on marginal lands: a regional modeling perspective. Journal of Environmental Quality 42(6):1802-14. https://doi.org/10.2134/jeq2013.05.0171. L418: For how long would these increases occur until a new equilibrium is reached? L420–422: Although the last chapter of the white paper (Chapter 9) puts various components into context and sums things up, it would be helpful to the reader of this chapter to put the estimate of 100–200 MMT CO2eq into perspective by comparing it to the 0.6 Gt target value indicated in L116 of Chapter 1. Soil C sequestration will get us only 1/6th to 1/3rd of the way to that target. It would also be helpful to harmonize units (MMT vs. Gt) throughout the document. L420–426: How will climate change affect these potential gains and losses of soil C stocks? It seems likely that a warmer (and in some places wetter) climate would result in higher rates of soil organic matter (SOM) decomposition, making the hoped-for soil C gains more difficult to achieve. This is speculative, but the reader should be made aware of this source of uncertainty. L427: Can all of this be put into perspective relative to achieving carbon neutral or carbon negative agriculture? Is it feasible for soils to offset fertilizer, N2O, plus equipment plus enteric CH4 plus CH4 from manure? Also, there was a 2 Gt offset number in chapter 1. Do you come anywhere close to that number? L429: The remark “We have progressed toward a robust understanding of the mechanisms controlling soil organic carbon (SOC) stabilization and storage, as well as the key role of SOC in ecosystem” might be prefaced with words like the scientific community has. As worded it could be taken that as saying this chapter has described this progression. L432: “Best management” might be qualified as best carbon management as income also would be involved in what an economist would call best management. L442: You have not described several huge challenges – incentive design, measurement and monitor- ing, additionality, permanence, leakage, uncertainty, transactions costs, stimulating adoption, and the non- stationarity introduced by climate change among others. L444–450: These potential soil C sequestration estimates should also be put into perspective with total U.S. greenhouse gas (GHG) emissions. The point here is that every little bit helps, but the last sentence 24 

Appendix A could be misconstrued to mean that the soils could be a large part of the solution to reducing total national emissions. However, at most they can be only a relatively small part of national emission reductions across all sectors, maybe one-third of the 10% contribution of agriculture to total U.S. GHG emissions? Table 1. Suggest expanding this table and adding information from the remaining regenerative agri- culture practices suggested as strategies to increase soil organic carbon accrual. Also, Table 1 only includes two articles per practice. Also, Poeplau and Don (2015) is cited incorrectly. Table 4. This is a great table that should be part of the main text. Suggest adding “co-development of knowledge with farmers” somewhere in this rubric. Perhaps is it most apt for the row on socio-economic barriers because it is important to make sure that “outreach” is not presented as a one-way download of knowledge from “experts” to farmers, but rather a two-way exchange and co-development of knowledge. However, engagement of farmers in all types of research would likely be valuable because many of them have ideas about technical aspects as well as socio-economic aspects of innovation. Perhaps this concept belongs in the main text as well as here in a table. Editorial and Minor Issues L22–23: Either not use acronyms for carbon or be consistent with the use of acronyms throughout this chapter and the whole white paper. L52: The authors probably should reference these initiatives. L133: There is a sentence fragment here. L134: “What should happen” depends on farmer goals not purely carbon sequestration. L139: How about profitability? L386: Be consistent with the use of GHG units. Change CO2e to CO2-eq, as in Chapter 1 on L67. L398–400: This is missing a reference. CHAPTER 4: CLIMATE MITIGATION AND NITROGEN AGRICULTURAL OPPORTUNITIES TO ABATE NITROUS OXIDE (N2O) EMISSIONS Specific Comments on the Text L23–29: The introductory text to this chapter presents quite different statistics than those given in Chapter 1. The figure of 24% must include forestry, but this is omitted from consideration in Chapter 1. L30: This chapter states yet another “take” on the overall goal, this one “net zero carbon emissions”. The authors of each chapter need to agree on the goal and use a consistent term in each chapter. L34: Is the 33% contribution from N fertilizer due to the CO2 emissions during the Haber-Bosch synthesis process? This needs better explanation. L40: This sentence makes a definitive conclusion, which might be qualified with the word “potential.” It should also say something about its contribution to whitepaper’s roadmap. L42–50: This is a great description of Global Warming Potential (GWP) and a good explanation of using CO2 as a reference gas when accounting for the total amount of GHGs. This concept might be moved to the introductory chapter (Chapter 1) or used as an example for a glossary. L51–52: The Intergovernmental Panel on Climate Change and Environmental Protection Agency have reported various GWP100 values for nitrous oxide (N2O), including 310, 298, 273, and 265 (see https://ghg protocol.org/sites/default/files/ghgp/Global-Warming-Potential-Values%20%28Feb%2016%202016%29 _1.pdf and https://erce.energy/erceipccsixthassessment/). It might be best to pick one (273 is the most recent AR6 estimate) and be consistent (as both 273 and 300 are used in one sentence here). It might be worthwhile to explain that there are a variety of estimates in the literature, and they are all near 300. L58–59: The rate of increase of N2O concentration in the atmosphere is also increasing. It was 1.3 ppb/yr in 2020 and 2021, up from 0.96/yr. for the 2010-2019 average. L107: Does “enters agricultural systems via precipitation” mean via rainfall or some other pathway? 25 

Review of White Paper on a Scientific Roadmap to Carbon-Negative Agriculture L124: When the text says, “which is why IPCC national greenhouse gas inventories are by default based on a simple percentage of nitrogen inputs” One could add assuming soil N is a direct result of N applications. This is not uniformly true as for example the U.S, inventory comes out of DayCent which is not just using a simple ratio. L171–174: Is there more than one study showing high N2O emissions from legume cropped fields? How common is this and how confident are we of this finding? Isn’t it contrary to the IPCC protocols for calculating Tier 1 emissions of N2O from legume cropped fields? L212: It seems that few farmers actually reduce N application rates even if they use enhanced effi- ciency fertilizers (EEFs) and inhibitors. It should be emphasized that convincing risk-adverse farmers that they can apply less nitrogen is a very big challenge, even if EEFs and inhibitors are shown to be effective at what they do. L222–227: Explain why it often takes 10 or more years to see the impacts of adoption of no-till. L251–254: Could nitrogen use efficiency (NUE) be overestimated, which would explain high N2O emissions? How well do we know biological nitrogen fixation (BNF) inputs? Underestimating BNF would cause overestimation of NUE and would support more N2O emission. L269: The term “soil quality” is used here but not defined. Other chapters use term soil health. L286: What is meant by “circular agricultural systems”? Consider adding a small explanation. L313: Isn’t over-fertilization used by farmers as an insurance policy against weather variations, which is their justification for applying extra N fertilizer? Perhaps you could comment on this. L328: Any comment on the cost and analysis needs of this? L332: The reduced emissions from using bioenergy crops as feedstocks for bioenergy versus other energy sources could be mentioned here. L354: Don’t certain combinations of weather conditions cause large fluxes of N? Can anything be done to identify those conditions and provide some moderating factor? L380: Could comments on N2O emissions control from grasslands, such as changing the forage spe- cies or fertilizer, when used, be added? L382: Regarding downstream ecosystems –Are “downstream” systems included in “carbon nega- tive”? All comes back to a clear definition, as noted earlier. L383: Clarifying text might simply be added that N applications that do not exceed plant needs would also help modulate N2O. L384: This is the first time the term “reactive nitrogen” has been used but is not defined. L393: Other chapters listed research needs here. What research needs are there for nitrogen manage- ment? Editorial and Minor Issues L26: “N2O producers”? L130: “To define” might be better expressed as to observe. L132: What does equilibrate mean here? L193: What do you mean by “(below)”? L197: This phrase is odd: “(see later)”? L239: There is a need to be consistent when referring to crops and adding their scientific name. In this line, wheat, rice, and maize were not accompanied by their scientific name in the same way as in L167 switchgrass and poplars were. L385: Be consistent with the use of either acronym like “‘NUE” or the whole term “nitrogen use efficiency.” Figure 1. Include percentages within each category of your doughnut chart. Figure 4. Some information here is illegible. More explanation is needed in the figure legend and the manuscript text to draw the reader into this complex figure. The visual units are small and hard to read, so many readers may simply skim past this figure. 26 

Appendix A CHAPTER 5: ANIMAL PROTEIN PRODUCTION CHALLENGES AND OPPORTUNITIES Specific Comments on the Text L7: This backgrounding on food demand is relevant to more than just animals and might be introduced to a greater extent in Chapter 1. L10: The text encourages a focus on “Innovation in climate-negative systems” but why not low-GHG. Has “climate-negative” been defined? L11: The verbiage here is not very consistent with the literature, in the sense that “climate pollutant” is not commonly used—maybe radiative forcing GHGs. Rather than refer to “short-lived climate pollutants” perhaps the background for animals and CH4 can be laid out. The sentence “Sharp reduction of short-live climate pollutants results in a negative climate state” needs better explanation. Reducing additions to at- mosphere is different to removing carbon to achieve a negative state. L16: It might be useful to provide global numbers for context, as other chapters of the white paper have done. L22: Add milk, eggs, to meat and fiber. L24: The climate negative concept is difficult to understand in the context of animal agriculture. Low emitting or lower emitting makes more sense, although rangelands for grazing may provide some seques- tration, but it is doubtful it could outweigh emissions. L28: Climate negative would seem to mean not just reduction a reduction but also added sequestration more than offsets emissions. L34: Clarify that it is the U.S. beef supply. L36: Not sure why it is important to say that beef is the most segmented supply chain as one could say chickens and hogs are close. More to the point this is just not needed. L38: The poultry and swine industries have also consolidated. L39: The sentence should clarify that beef livestock consolidation is less persistent. Livestock is broader than just beef (dairy, chicken, and hogs are more consolidated than crops). L46: Are seedstock producers the same as breeders? L49: “Ways to mitigate emissions from intensive finishing feedlots focus on enteric methane, manure, and feed crops and will be discussed in the dairy section.” This suggests feedlots are discussed in dairy section, but they are not. L54: In the U.S., grazing is about twice as large as crops and is about equal to forest land. Given the more U.S.-centric nature of this white paper perhaps U.S. numbers should be introduced for consistency with other chapters in the white paper that provide U.S. and global information. L59: Are fire and disturbance also important plus type of vegetative cover and grazing intensity? L62–64: Avoiding the loss of rangelands—This sounds like not disrupting soil C, but how does that contribute to achieving carbon negative values? L63: When the chapter says “avoiding the loss of rangelands or conversion to other uses is most important to maximize C stocks and achieve carbon negative beef systems” those are not carbon negative actions. To get there, you need a net increase in soil C not avoided loss. L73–76: “The most important practices that increase long-term carbon storage in rangelands and con- tribute to climate negative beef systems are 1) avoiding conversion, 2) restoring cultivated or degraded lands, and 3) practicing adaptive livestock management.” Should climate negative be changed to carbon negative? Also, the next section of the text highlights the important role of forage species in this system, but forage selection is not part of that list. It wasn’t clear if that is an explicit part “practicing adaptive livestock management.” L76: The text doesn’t address anything on vegetative mix, fire, or adding organic material. See Joyce et al., 2013. 27 

Review of White Paper on a Scientific Roadmap to Carbon-Negative Agriculture Joyce, L. A., D.D. Briske, J.R. Brown, H.W. Polley, B.A. McCarl, D.W. Bailey. 2013. Climate change and North American rangelands: Assessment of mitigation and adaptation strategies. Soci- ety for Range Management 66(5):512-528. L79: “To sequester more C in grasslands, the particulate organic matter pool (plant inputs) must in- crease.” One might add, relative to the rate of decomposition (as mentioned in the soils chapter). L83–86: The point comes across that balancing carbon storage and N2O emissions is a complex sys- tem and concept, but the text seems vague about whether legumes are good for soil C or not, and under what circumstances they might be. Can this be clarified? L86–91: If you are saying that N2O emissions fueled by the Biological Nitrogen Fertilization (BNF) from legumes offset the soil C sequestration, chapter 4 should be cross referenced here. Likewise, the last sentence of the paragraph should cross reference chapter 3. L93–98: You cite two different studies by Cusack et al., but are they consistent? Is 0.28 MgC ha-1/yr of C sequestration in the 2017 study roughly equivalent to the 46% reduction in GHG emissions reported in the 2021 study? Does a pattern emerge? What does this say about the consistency of LCA studies with one another? L94: To be consistent with other data given, including a conversion from Mg ha-1/ yr to Tg/yr would be a helpful comparator. L104: Is “the next few decades” a common time horizon across all the chapters of the white paper? L114: While dairy farms have decreased cows per farm has increased. Herd size and total dairy num- bers might be added relating to effects on total GHG emissions. L118: Should the Food and Agriculture Organization (FAO) report on Livestock’s Long Shadow be referenced? L131: If this is where you cover feedlot animals and enteric emissions, should not this section include both dairy and beef cattle? L138: Different breeds of animals, faster growth rates, and other animal yield improvements have also been mentioned as approaches to reducing methane emissions of animals. L141–148: These are rather old citations. Hasn’t the science of livestock feed and emissions pro- gressed significantly since 2008–2013? L146: Why does the control diet show a decrease, wouldn’t the control diet be the baseline? L150–156: Does the effectiveness of these feed additives decline over time? L158: Hogs should be in this section as a lot of GHG trades under the Clean Development Mechanism involved manure management for hogs. Feedlots should be mentioned. L161: Other manure management systems—provide an example or two of other systems. Earlier, the text said that feedlots would be covered under dairy. L165–166: “Capturing methane in an anaerobic digester and combusting the gas effectively reduces methane emissions and manure organic C content.” This reduces methane but would it result in fewer CO2e? L173: Not sure what is meant by addition of “as well as fertilizing the crops and forages.” Perhaps the sentence needs restructuring. L183: Under feed, there should be a discussion about feed conversion rates and implicit emissions in feed used, for example, cattle use about 7 lbs. of feed per lb. of gain and hogs, 4 lbs., chickens, 1.5 lbs., and fish 1 lb. One could argue that shifting protein forms can reduce GHG emissions. Some argue that having less meat in the human diet would enhance mitigation. I think this section sidesteps the larger point of emissions in raising the feed for the animals. L184: Should not beef cow-calf herds be included in this discussion as they are much more of a graz- ing entity? L190: Seems as though the overall size of pork production and the share of pork in the diet should be mentioned here. 28 

Appendix A L195–196: With respect to the cradle-to-grave carbon footprint of U.S. pork—This kind of data would be good to provide for other animal species/systems, too. L199: The text notes that the effects of warmer regions influence on “driving greater nitrogen volati- lization.” Please add whether that is from the manure or elsewhere. L200: Emissions from feed production should be covered in an all-livestock manner in this chapter rather than production system by production system with unequal treatment in each. L205: It might be useful to make a cross reference to chapter 6 on energy; also, wind and solar are relevant here. L219–222: What fraction of animal protein could be supplied by synthetic amino acids? This could be the biggest emissions reduction for several types of livestock if a significant fraction of crop production for crude protein in feed could be avoided, thus avoiding broadcasting of N fertilizers for so much high-N corn and cultivation of leguminous crops to produce proteinaceous feed. L251: The text mentions a diet formulation tool “where regional environmental life cycle inventory information was incorporated in feed formulations.” Such tools might include the social cost of carbon in calculating feed costs. L259: With respect to “consumer demands for shifting management” – Is this a reference to organic and cage free demands? L266: Animal welfare in poultry section seems overemphasized relative to other animal systems. Se- lection indices are changing to include welfare measures such as leg strength, group behavior. L296–297: “Therefore, management…” should be restated as feed and grazing management…. L300: Increasing lambs per ewe could also raise animal welfare concerns. This is not to suggest that more attention to animal welfare concerns be added to this chapter, only noting here that the coverage of animal welfare concerns in this chapter is largely focused on poultry and should be more balanced. L312–318: Shouldn’t it be acknowledged in a simple declarative sentence that livestock production will never have net negative GHG emissions, so that puts the burden on row crop agriculture to sequester enough soil C to offset the livestock emissions? Reducing livestock emissions is critically important but acknowledge the limits of going to zero or into the negative. L316: The 70 to 80 percent from cow-calf seems too high (see the Environmental Protection Agency’s GHG inventory which has different shares). L328: The greatest opportunity for poultry and pork production lies in improved yield for feed—not sure what this means, is it better crop production to reduce the carbon cost of feed? Editorial and Minor Issues L48: Some confusing wording: “In this section, we focus on increasing C storage and sequestration, which are more applicable to extensive operations.” That is one focus of the section but enteric and manure are also covered. You might also mention links to feed supply. L54: Delete the stray s after C. L108: Bad wording: “Milk production… has seen tremendous improvement in … milk production.” L145: Up to a 9%. L149: Perhaps us could instead of “can”? L151: Add (3NOP) next to the full word. L153: Should Asparagopsis be spelled out here? L207: Define hoop barns. L234–235: Does this mean poultry meat is lower in protein than ruminant meat? L248: Including less ruminant—less ruminant in the feed? Does this mean less bone meal or blood meal? L325: “Methane in milk production”—perhaps use from? L328: “Yield for feed production”—perhaps increased feed efficiency or yield per unit feed consumed. 29 

Review of White Paper on a Scientific Roadmap to Carbon-Negative Agriculture CHAPTER 6: CHALLENGES AND OPPORTUNITIES FOR ENERGY AND EFFICIENT ENERGY USE Specific Comments on the Text L13: Does the chapter need to provide total U.S. energy consumption numbers? L17–19 and Fig. 2: Does this energy consumption include both crop and animal production in the definition of “agriculture” energy use? L20: Energy that increases yields and reduced labor is also linked to fertilizer and inputs like pesti- cides. L27: Should not food supply and energy use be used herein rather than acute risks and changing climate? One way to think about it is that the changing climate goes toward a systematic force putting pressure on the cost and usage of energy as well as modifying the production systems and causing needs to adapt. L30: Since this is the energy chapter, the authors might consider including adding a major section describing agriculture as playing a vital role as a supplier of alternative energy through liquid fuels, feed- stocks for electricity, and land for wind and solar. L34–38: What is the definition of a “farm” in this context? Is this only for row crop farms, animal production farms, or mixtures of crop and animal production systems? L37–38 suggest row crop ag only, but L51–54 seem to include animal production systems. What is the scope of this chapter and when do specific sentences refer to only crop, only animal, or both? L36: It would be good to clarify if the model on whole farm emissions omits treatment of processing in the whole supply chain plus transport. L38: It might be good to include irrigation water supply, but maybe not including transporting in the list of on farm activities. It certainly is a major part of the supply chain and may be substantially involved with transporting farm equipment from location to location, but it is more related to input supply and output demand. On the output side one might also talk about energy use and processing. L43: What about fertilizer, fossil-based chemicals, and irrigation water supply in the list of inputs L51: That there is energy use implicit in all the things that you list might be added parenthetically to the sentence. L53: The supply chain is more than just national security and domestic welfare. It is also important for international relations, avoiding conflict and more generally, global welfare, particularly given the U.S.’s large role in international agricultural trade. L60: There is substantial natural gas-powered irrigation pumping as well. L63: Should qualify this sentence “Direct energy is responsible for 60 percent of agricultural energy use and 40 percent from indirect energy consumption” as this may exclude processing and transport. L70: Do “all” non-leguminous crops use much nitrogen fertilizer? L76: It would be good to add words to the chemical symbols like for potassium and phosphate ferti- lizers. Figure 3 is very informative. L85–87: The authors might consider the potential use of green ammonia for on-farm fuel, especially if green ammonia production is decentralized so that it is produced on the farm or in local farmer coop facilities. L87: The sentence “Increasing nitrogen use efficiency will reduce impacts from GHG emissions and nitrogen losses to the environment” might be reworded as this topic is covered elsewhere in this white paper in at least two places. This chapter should bring out the energy aspects as opposed to worrying about the nitrous oxide. Figure 4: This is informative, but an explanation of color coding is needed. Color-code the two Y axes or at least indicate in the legend which color bar goes with which axis. L126: Some of this is covered elsewhere in the white paper. But this section might discuss alternative fertilizer sources manufactured with renewable energy such as hydrogen-based ammonia manufacturing. 30 

Appendix A L141: This section might add a discussion about lower energy-requiring crops, increased irrigation efficiency, and irrigation pumping powered by renewable energy sources. L142: This section is titled Renewable Energy Production and Uses but it doesn’t discuss wind and solar and agrivoltaics, which are potentially big opportunities for agriculture. L153: This paragraph might reference the photosynthesis improvement discussion in chapter 2. L158: It isn’t clear to what this sentence is referring. L159: Is the liquid biofuel industry really rapidly maturing or just stagnant? The production of corn ethanol has been stable, and there seems to be a small growing segment in biodiesel, but the only growth in the cellulosic ethanol has come largely from landfill gas. L165: Don’t landfills contribute significantly to renewable natural gas (RNG)? L172: Consider adding additional references from the large literature segment on indirect land use and/or the topic of carbon leakage in addition to the Field et al. paper. Austin, K. G., J. P. H. Jones, C. M. Clark. 2022. A review of domestic land use change attributable to U.S. biofuel policy. Renewable and Sustainable Energy Reviews 159:112181. https://doi.org/10.10 16/j.rser.2022.112181. https://www.sciencedirect.com/science/article/pii/S136403212200106X. L179: This section is describing what is known as bioenergy production with carbon capture and sequestration (BECCS) and should be referenced. L193–200: The discussion on resilient crop production systems raises the question of which cropping systems might be most vulnerable and how to manage these risks. Figure 5: More information is needed in the legend and/or the text for the reader to understand this figure. Identifying the location would be helpful, but more importantly, what is the reader supposed to see in these images and how would these images help farmers manage their land? L230–232: Linking to Chapter 3 would be appropriate here. Editorial and Minor Issues L11: The authors might reword “transform fossil energy into food” as this does not literally happen. It is certainly a vital input and yield enhancer. L40: Perhaps “encumbered” is not such a good word to use. Embodied might be better. L134: A better word than “encumbering” is needed. L148: “Emitted” doesn’t seem like the right word here. Maybe incident upon. CHAPTER 7: CLIMATE IMPACT AND MITIGATION POTENTIAL OF U.S. FOOD LOSS AND WASTE Specific Comments on the Text L1: The title is misleading as the chapter only deals with one aspect of food loss and waste (FWL). Beginning the chapter with “The global production and consumption of food pose one of the greatest threats to our environmental resources and the planet’s wildlife” is very provocative and doesn’t have much to do with the content of the chapter or the white paper. Another way to frame it, more neutrally, is that food production strongly competes for land resources and releases a number of pollution streams. L26–28: “Food waste and loss are significant contributors to climate change, producing an estimated 10% of global emissions and 18–24% of the food system’s GHG emissions….” This statement can’t be correct as it is greater that the entire GHG output of agriculture cited in Chapter 1. This also does not square with L58 “This high level of FLW creates significant climate risk for the U.S. and is now responsible for four percent of the country’s GHG emissions.” Something is missing here to explain this discrepancy. Maybe the issue is how FWL is attributed to different sectors? It would be useful to identify the basic constituents of these figures, whether the whole life cycle or just emissions from wasted food. 31 

Review of White Paper on a Scientific Roadmap to Carbon-Negative Agriculture L32: There is a likely inconsistency with other estimates here. If food waste is 4% of total anthropo- morphic GHGs and 16% of agricultural emissions, then agriculture must contribute about 30% of total emissions but is not reported to be that high earlier in the white paper. For perspective it would be good to have some U.S. estimates. L43: This introduction should somehow link the main topic to carbon, arguing that managing waste can propel the agricultural and food system toward a new state of low carbon emissions. L57: The estimate that food waste is 4% of U.S. greenhouse gas emissions is also likely to be incon- sistent with numbers used in other chapters of the white paper. Some estimates for agricultural production are 7% so food waste and loss would be over 50% of all agricultural emissions. Again, the scope of contri- butions could be defined. L58: Another way to look at it is that carbon in decomposing produce is carbon that was drawn from the air earlier in the year so would net out as is commonly computed when evaluating biofuels. It would be good to focus more about carbon used in fertilizer, fuel etc. in growing and raising that unused food plus the need for extra land in growing the crops for feed. L63: The text here might link food waste to CH4 and talk about aerobic and anaerobic decomposition. L68: Would recycling the organic material and food waste now placed in landfills reduce methane and cause more of emissions to be CO2 gaining in GWP-based climate forcing? L77: Would recovery of food currently lost release any land from production? L77: What about food waste at consumer level? It is not clear if this link is part of the discussion. L105: Is there much loss in animal feed systems? There is not a case for that not built in the text above. L120: What tools are needed to catalyze action from the Stewardship Index for Specialty Crops (SISC) Food Loss Metric. The text seems to say this type of measurement is need, but then it explains that the SISC exists. Maybe the text could argue need for more implementation or for the creation and widespread distri- bution of SISC information in many contexts coupled with carbon footprints. L123: So SISC is not a single valued metric but seems to be a family of metrics—this needs more clarity in the text. L128-136: What incentives do farmers have to participate in the Food Loss Metric program? L142: What exactly is being suggested here for widespread adoption—a requirement, an extension program, consumer information, producer information? A statement is needed in the text. L144: The text says “FLW is again not a point-source pollution” but this wasn’t pointed out earlier, so the word “again” is confusing. L146: Designing out all loss and waste if a high bar to reach for but perhaps reduce loss and waste and recycle the nutrients etc. might be more achievable. L147: Provide examples and references for the text that says, “proven solutions exist.” L149: WWF should be spelled out. Also, add a little more text on nature of successes achieved in GHG and food waste terms. Figure 2: It might be good to classify and discuss some of these strategies to reduce FLW. Not all the short labels in the table are self-explanatory. For example, please give more information on what efforts are underway or proposed to reduce portion sizes. Since this is at the top of the list for potential emissions reduction, perhaps it merits more attention in the text. Does this mean portion sizes in meals at home or at restaurants or both? It can be assumed that the approach to each would be different. It might also be useful to express the emissions reduction estimated in terms of percentage of all FLW emissions to complement the CO2e numbers. L161-165: Similar estimates were already given in an earlier paragraph, so this text seems redundant. L168: Please provide any available estimates and a reference relative to “row crop commodity loss in field, transport, and storage may be very low.” Is there similar information on animal feed? L172: The temptation to save time and energy by using an excerpt from another publication is under- standable, but it results in some redundancies within this chapter, so really should be revised to improve flow, reduce redundancies, and make sure that all imported topics are relevant to the present objective. 32 

Appendix A L177–187: What are the impediments to directing more food waste to animal feed? How can these impediments be overcome “using 21st century technology and practices”? Please unpack that for the reader. L189–213: These research results are in a section on future research but the research needs being advanced here need to be clearer. Most of the material could be integrated into the earlier text. L252: This section does not get to the research needs. L237–246: A table showing pros and cons for both growers and retailers could be helpful. Specifi- cally, what are the economic or other barriers to adoptions of these types of contracts? L257–276: This text doesn’t make clear how soil health is necessarily linked to reducing post-harvest loss or how adopting regenerative agriculture affects food waste? This is a nicely crafted argument for regenerative agriculture, but it needs to explain further a link to the food waste topic of this chapter. There might be added a more direct argument that reducing food waste reduces land needs and as such the asso- ciated emissions plus associated externalities. L294–295: Was the Bill Emerson Good Samaritan Food Donation Act enacted by Congress or is it still pending? If enacted, when, and why hasn’t this campaign already been launched? L332–334: Are the food hubs that are “already in place” run by governments, NGOs, or private sector? What is the vision for how they should be supported and managed? L350: A lot of these policies have been put in place. Any success stories? L366–368: The readers of this document would probably appreciate learning a bit more about this issue of defining compost products and how that affects their use. This is an interesting topic but presented too briefly to really understand. L370–373: How well did this program work from 2018 until it expired? What lessons can be learned from it? Is reauthorization currently being negotiated in the 2023–2024 Farm Bill? L391: The conclusion section begins with “As the climate changes and the food system shifts in in- creasingly unpredictable ways, reducing FLW as part of a government overall approach to mitigate climate risk and food insecurity will be imperative.” At least part of this statement is an opinion not necessarily a fact and besides, it does not directly relate to the purpose of the whole white paper. It would be better to frame this topic as a carbon lowering action which has substantial co-benefits. L397: The “nearly one-fifth” of U.S. cropland, fertilizer, and agricultural water used to grow food that is wasted is from an estimate of fresh produce and may overestimate for all crops and livestock. It would be helpful to provide a little more detail for that estimate. L412: The statement on increasing “the transparency of loss measurement,” suggests that loss is now measured but is not transparent. From a slightly different perspective, one might argue more for increasing the prevalence and monitoring of losses accompanied by dissemination of information. L415: One somewhat related fact is that, for years, the value of manure has been advanced, but eco- nomics often shows that the hauling cost of applying it as a fertilizer is much greater than chemical coun- terparts when the location of application is a substantial distance from the source. The issue of material assembly for conveyance and movement to service area may raise substantial costs and preclude move- ments to more distant areas due to cost and in turn limit applicability in regenerative ag. Also, there may be some risk issues that food waste supplies are only available under certain circumstances so it cannot be relied on and would not be worth the investment in facilities to spread the food waste that are only operated occasionally. Editorial and Minor Issues L172: In this document, the text should probably not refer to the study as “our”. L230–235: Check for grammar and typos: “all of which ADD”; “systems for fresh produce ARE”; “nott” vs. not. L399: “Radical transparency”—jargon! 33 

Review of White Paper on a Scientific Roadmap to Carbon-Negative Agriculture CHAPTER 8: ECONOMIC AND POLICY RESEARCH CHALLENGES Specific Comments on the Text L7: It is hard to know if the Biden administration will still exist much beyond the publication of this white paper, so perhaps this chapter should be revised to give it life regardless of who wins the next election (reference to Biden also appears elsewhere, such as L243). L8: The goal of the Biden administration was net zero versus net negative. L14: It would be good to add some references to these absolute statements. L16: Substantial consensus does exist on many potential policies and programs that could result in a reduction in the rates of emissions but not on their total effectiveness. Pete Smith has led at least 3 lit review efforts and there have been others. See reference in this document in comment on L366 of Chapter 2. L18: The statement “Sustained investment is needed to integrate climate science and technical poten- tial” misses the need for understanding economics, transactions costs, and behavioral responses in the pol- icy development process. L30: In might be mentioned that a “larger sustainable development challenge” encompasses main- taining food and fiber supply for a growing population. L35: “Net Negative Emissions (NNE)” is yet another way the overall goal of the white paper has been expressed. NNE is apparently all emissions, including particles, all gases, liquids, solids? This should be defined. L35 and after: No other chapter of the white paper has a preview of findings; this should probably be made consistent throughout. L49: An argument might be made for a total climate impact that also folds in adaptation and mitigation along with effects. L62: Consumer reactions/influence are not discussed in other chapters of the white paper as a possi- bility. Perhaps this needs more context if it is to be included here. L83: The exclusion of agriculture from the Kyoto Protocol was due to more than perceived scientific uncertainties about soil and biomass carbon sequestration. Other reasons included food importance, perma- nence, uncertainty, additionality, leakage, nonpoint pollution and more. L132: Irrigated rice has not been mentioned in this white paper before this chapter, nor have defor- estation/afforestation concerns been discussed. Land use change is mentioned but perhaps not so compre- hensively. L152: There are a number of topics missing from Table 1 including some significant ones like regional heterogeneity of impact of strategies (See Murray et al., 2005); resource competition among strategies (See competitive economic potential discussion in McCarl and Schneider, 2001); and three types of responses that include sequestration, emissions reduction, and emissions offset (the latter is bioenergy) laid out by McCarl and Schneider in Choices and what is now AAEP. Murray, B., B. Sohngen, A. Sommer, B. Depro, K. Jones, B. McCarl, D. Gillig, B. DeAngelo, K. Andrasko. 2005. Greenhouse Gas Mitigation Potential in U.S. Forestry and Agriculture. U.S. Envi- ronmental Protection Agency, Office of Atmospheric Programs. McCarl, B. and U. Schneider. 2001. Greenhouse Gas Mitigation in U.S. Agriculture and Forestry. Science 294(5551):2481-2482. https://doi.org/doi:10.1126/science.1064193. https://www.science. org/doi/abs/10.1126/science.1064193. McCarl, B. and U. Schneider. 1999. Curbing Greenhouse Gases: Agriculture’s Role. Choices: The Magazine of Food, Farm, and Resource Issues 14(1):131673. https://EconPapers.repec.org/RePEc: ags:aaeach:131673. 34 

Appendix A McCarl, B. and U. Schneider. 2000. U.S. Agriculture’s Role in a Greenhouse Gas Emission Mitigation World: An Economic Perspective, Review of Agricultural Economics, Agricultural and Applied Eco- nomics Association 22(1)134-159. L160: Several additional studies could be referenced here, such as the latest USDA Office of the Chief Economist report on (Jones and O’Hara, 2023); (McKinsey & Co., 2009); (Murray et al., 2005); (McCarl and Schneider, 2001), and several bioenergy studies out of the Research Triangle Institute. Jones, J., and J.K. O’Hara (Eds). 2023. Marginal Abatement Cost Curves for Greenhouse Gas Miti- gation on U.S. Farms and Ranches. Office of the Chief Economist, U.S. Department of Agriculture, Washington, DC. McKinsey & Company. 2009. Pathways to a Low-Carbon Economy: Version 2 of the Global Greenhouse Gas Abatement Cost Curve. https://www.mckinsey.com/~/media/mckinsey/dotcom/client _service/sustainability/cost%20curve%20pdfs/pathways_lowcarbon_economy_version2.ashx#:~:text =curve%20shows%20the%20range%20of,2030%20time%20horizon. L167: “A key insight from the early research was to distinguish the technical potential for greenhouse gas mitigation—i.e., the amount that was scientifically and technologically possible—from the economic potential—i.e., the amount that farmers would do given their economic circumstances, capabilities, and motivations.” In short, farmers need to make money to survive and won’t do things that sink farm econom- ics. This information was mentioned by McCarl and Schneider (2001) along with a third type of potential— the competitive economic potential. L177: Cost might also be defined in terms of implementation cost and opportunity cost. For example, the direct cost of afforestation might not be so much, but the value of land is lost to other enterprises. L181: This could be good place to discuss the social cost of carbon and market values. In 2022, EPA raised its estimate of the social cost to $190 per ton; previously it was $50 and many transactions in the few existing trading markets have been $30 or below. L186: It is advisable to add a reference to statements like “In additional to these technical and eco- nomic factors, research also has shown that farmers’ willingness to make climate smart management change….” Considering the target audience for the white paper and what is appropriate in terms of the level of scientific content, there are many statements in this chapter that provide supporting references. L189: The profitability of taking action, considering the opportunity cost of resources, should be more prominently mentioned to help make the case for needed incentives. Antle and McCarl (2002) addresses that issue. Antle, J.M. and B.A. McCarl. 2002. The Economics of Carbon Sequestration in Agricultural Soils. In Tietenberg, T. and H. Folmer (Eds.). The International Yearbook of Environmental and Resource Economics 2002/2003: A Survey of Current Issues. 2002. Edward Elgar Publishing. L200: With respect to additionality, two other considerations are that 1) continued use of carbon- building practices by farmers who had already been using them do not add more carbon than what would have been added by them anyway, and 2) almost all farm groups came out with the position that all farmers should be paid, even they had been employing carbon sequestering practices before incentives were avail- able. L205: On the poorly designed policy side of the equation, there is also the risk that farmers may reverse adoption and plow it up in anticipation of being to enroll and be eligible for benefits. There are more careful discussions of these permanence, etc. issues in the book, that Gordon Smith led (Willey and Chameides, 2007) and in a monograph led by Mac Post (W.M. Post et al., 2009). Also see Ogle et al. (2023) and Fei and McCarl (2023). 35 

Review of White Paper on a Scientific Roadmap to Carbon-Negative Agriculture Willey, Z. and B. Chameides. (Eds.). Harnessing Farms and Forests in the Low-Carbon Economy: How to Create, Measure, and Verify Greenhouse Gas Offsets. Durham, NC: Duke University Press. 2007. 229pp. ISBN 9780822341680. U.S. $60, paper. Post, W.M., J.E. Amonette, R. Birdsey, C.T. Jr. Garten, R. C. Izaurralde, P. M. Jardine, J. Jastrow, R. Lal, G. Marland. 2009. Terrestrial biological carbon sequestration: science for enhancement and implementation. McPherson, B. P., E.T. Sundquist. (Eds). Carbon Sequestration and Its Role in the Global Carbon Cycle. Geophysical Monograph Series 183. American Geophysical Union. Devon, UK 73-88. Ogle, S. M., R. T. Conant, B. Fischer, B.K. Haya, D.T. Manning, B.A. McCarl, T.J. Zelikova. 2023. Policy challenges to enhance soil carbon sinks: the dirty part of making contributions to the Paris agreement by the United States. Carbon Management 14:2268071. https://doi.org/10.1080/175830 04.2023.2268071. Fei, C.J. and B.A. McCarl. 2023. Agricultural Soils and the Quest for Net Zero Emissions. Choices: The Magazine of Food, Farm, and Resource Issues, Agricultural and Applied Economics Associa- tion 38(4). L213: There is a deeper treatment of permanence in several places in the literature and that raises additional issues. Reversibility is certainly one as covered here but also the nature of annual increments (they decrease as soil reaches an equilibrium, as discussed in Chapter 3 in this white paper). Another issue is in maintaining carbon after a new equilibrium is reached. Do we keep paying to maintain the carbon? Finally, there is the issue of whether the practice can be maintained in the face of, for example, weed re- sistance to glyphosate. (See Kim, McCarl, and Murray, 2008). Kim, M-K., B. McCarl, B. Murray. 2008. Permanence discounting for land-based carbon sequestra- tion. Ecological Economics 64(4):763-769. https://EconPapers.repec.org/RePEc:eee:ecolec:v:64:y: 2008:i:4:p:763-769. L214: While old papers by Babcock et al. on set-asides used the term slippage, it is and always has been termed leakage or indirect land use in the climate literature. There are some papers on this in a carbon setting. See Murray, McCarl and Lee (2004) and Wear and Murray (2004). Murray, B., B. McCarl, H-C. Lee. 2004. Estimating Leakage from Forest Carbon Sequestration Pro- grams. Land Economics 80. https://doi.org/10.2307/3147147. Wear, D.N. and B.C. Murray. 2004. Federal timber restrictions, interregional spillovers, and the im- pact on U.S. softwood markets. Journal of Environmental Economics and Management, Elsevier 47(2)307-330. L221: Important issues missing here include: Uncertainty in amount, difficulty of measurement, ways to measure combining modeling, field observations and remote sensing, transactions cost and pay for prac- tices versus carbon. Much of this is covered in Ogle et al. (2023) or in papers by Mac Post and others. (References above). L244: Consider changing “economic” to socio-economic. Indeed, this chapter is mostly missing dis- cussion of the social science aspects of farmer decision making that are not entirely or even mostly eco- nomic in nature. Farmer age, ideology, training, influence of neighbors and communities, and trusted sources of information are all crucial factors that need to be called out with respect to research. See Buck and Palumbo-Compton (2022) for a discussion of the needed social science research. 36 

Appendix A Buck, H.J. and A. Palumbo-Compton. 2022. Soil carbon sequestration as a climate strategy: what do farmers think? Biogeochemistry 161(1):59-70. https://doi.org/10.1007/s10533-022-00948-2. https://doi.org/10.1007/s10533-022-00948-2. L254: This consumer approach is not covered anywhere else in the document and might merit more coverage here or somewhere. There are current ideas in the literature about ecolabeling for carbon footprint or for low carbon goods and older ones about the GHG value of diet shifts away from meat. L263: The chapter might mention the demise of the Chicago Climate Exchange (CCX) trading system under voluntary market solutions. L265: There are other policy approaches that could have an impact: R&D, liability assignment, taxes on other high emitting goods, allowing agriculture to provide offsets into markets. The 2010 National Acad- emies report, America’s Climate Choices, has a wider array. America’s Climate Choices Panel on Limiting the Magnitude of Climate Change. 2010. Limiting the Magnitude of Future Climate Change. National Academies of Sciences, Engineering, and Medicine. 2010. Washington, DC: The National Academies Press. https://doi.org/10.17226/12785. L274: Some of Stavins’ lessons-learned papers and arguments might be added here. Schmalensee, R., R. N. Stavins. 2017. Lessons Learned from Three Decades of Experience with Cap and Trade. Review of Environmental Economics and Policy 11(1):59-79. https://doi.org/10.1093/reep/rew017. L290: There are several papers that address a carbon tax versus cap-and-trade approaches, such as Stavins (2022) and a literature review by Cropper and Oates (1992) that spans the possibilities. Stavins, R. 2022. The Relative Merits of Carbon Pricing Instruments: Taxes versus Trading. Review of Environmental Economics and Policy 16:000-000. https://doi.org/10.1086/717773. Cropper, M.L. and W. E. Oates. 1992. Environmental Economics: A Survey. Journal of Economic Literature 30(2):675-740. http://www.jstor.org/stable/2727701. L304: The research needs, such as “most likely some combination of regulatory and incentive-based mechanisms” have been the lifeblood of agriculture for many decades. This is “What we need to know”? L307–308: “Currently we do not know which agricultural sectors can realistically achieve NNE, or whether agriculture overall can achieve net negative emissions”. This statement seems to go against the very premise of the white paper. L350: This section might talk about the World Climate Research Programme Coupled Model Inter- comparison Project (CMIP) as a source of simulations. L354: The Shared Socio-Economic Pathways is a scenario tool that could be mentioned here, though more information is needed on agriculture. L365: Is the USDA Regional Economic Assessment Project model still active? L368: This is a quite incomplete literature review of integrated assessments and misses big efforts that have addressed the agricultural role in moving toward net zero. Here are some: The Forestry and Agricultural Sector Optimization Model with Greenhouse Gases (FASOM-GHG), described in McCarl and Schneider (2001) (referenced previously) addresses all components of the forestry and agricultural system except food waste. Murray et al., 2005 (referenced previously) includes forestry. Ogle et al. 2016, adds in nitrous oxide management: Ogle, S. B. McCarl, J. Baker, S. Del Grosso, P. Adler, K. Paustian, W. Parton. 2015. Managing the nitrogen cycle to reduce greenhouse gas emissions from crop production and biofuel expansion. Mit- igation and Adaptation Strategies for Global Change 21. https://doi.org/10.1007/s11027-015-9645-0. 37 

Review of White Paper on a Scientific Roadmap to Carbon-Negative Agriculture Wang et al., 2021, added in voluntary programs: Wang, M., B. McCarl, H. Wei, L. Shiva. 2021. Unintended Consequences of Agricultural Participation in Voluntary Carbon Markets: Their Nature and Avoidance. Complexity 2021:9518135. https://doi.org/ 10.1155/2021/9518135. Groups working on this include the Research Triangle Institute (RTI), Uwe Schneider in Hamburg, Burt English and others at the University of Tennessee, Petr Havlik at IIASA, an early USDA piece on the Kyoto protocol and the PNNL joint climate group at the University of Maryland. Across a lot of this there is a recent literature review in Fei and McCarl, 2023: Fei, C.J. and B.A. McCarl. 2023. The role and use of mathematical programming in agricultural, nat- ural resource, and climate change analysis. Annual Review of Resource Economics. John Reilly also has done some work on this in EPA, and so has Tom Hertel in GTAP. This also misses the EPA and USDA efforts to develop marginal abatement curves, referenced earlier, along with some efforts by Jason Jones and others at ICF to do cross model comparisons. This line of work has linked in Century and Epic along with econometric models on crop performance plus swat hydrology. L421: Using “trackable” to describe agriculture’s role in net negative emissions is an overstatement. Given the non-point emitting, heterogeneous nature of agriculture, lack of monitoring, and apparent policy preference for paying for practices not for results, it can be argued that most of this will never be trackable. It is possible that these activities can be sampled and estimated but fully trackable is a reach. (Some things like manure management come close to trackable but others, not so much.) L447: “In summary, the key research opportunity we have identified in this chapter is to build research program to design an effective climate policy for agriculture that will support the national NNE goal.” But NNE needs to be defined by the authors. Now all is reduced to policy. The authors need to consider the driving forces to change producer’s behavior, not “policy” as the panacea. L457: The authors should consider the role of competitive potential. For example, McCarl and Schnei- der (2005) show that if you only consider soil sequestration then you get much more economic potential than if you add in land-use change and biofuels, because as the carbon price gets higher, they displace conventional production. This is also in a paper by Lee, McCarl and Gillig (2004) with shows different roles of items over time. Probably also in Murray et al., 2005. Lee, H‐C., B. McCarl, D. Gillig. 2005. The Dynamic Competitiveness of U.S. Agricultural and Forest Carbon Sequestration. Canadian Journal of Agricultural Economics/Revue canadienne d’agroecono- mie 53(4):343-357. https://EconPapers.repec.org/RePEc:bla:canjag:v:53:y:2005:i:4:p:343-357. L377: Will more and better data needed to improve models help with increasing adoption of practices and technologies by producers? Will “more big data” have any impact beyond better models in academia? L458: Related to the issue of uncertainty and contracting, most agricultural outputs, like soil carbon and yield, have variable effectiveness from year to year. Kim and McCarl argue that longer-term spatially diverse contracts would be better in cap-and-trade systems to manage that variability. The same is likely true in policy as the need is to focus more on aggregate outcome than farm by farm outcomes. Kim, M-K. and B. McCarl. 2009. Uncertainty discounting for land-based carbon sequestration. Jour- nal of Agricultural and Applied Economics 41:1-11. https://doi.org/10.1017/S1074070800002510. Editorial and Minor Comments L93: The Paris development is more commonly called the Paris agreement, not the accord. L221: This material needs references. 38 

Appendix A CHAPTER 9: CONCLUSIONS AND SUMMARY OF PRIORITY RESEARCH NEEDS Specific Comments on the Text L11: The introduction states that “to achieve GHG negative agriculture in the U.S., cumulative emis- sions must be reduced by at least 0.6 Gt CO2e per year.” A central problem with the ability to support this statement is the lack of a defined system boundary in the white paper. Does this include algae blooms in the Gulf of Mexico or producing tractors to maintain production? More than the EPA reference of emissions for agriculture is needed for the sake of the reader, to show what exactly is meant by “U.S. agriculture” and to avoid any misunderstandings about how this goal will be achieved. The statement belies the sequestration increases and emission intensive product offsets that must be realized to meet this goal. Mathematically, this goal could all be achieved without any change in emissions (which is not a practical reality), but to address climate change CO2e sequestration gains will be needed along with the offset of emission intensive fossil fuels through bioenergy, wind and solar, plus building materials. L16: This sentence states that the white paper has identified the technologies and practices that demon- strated the highest reduction of carbon emission and the highest potential for implementation. The white paper missed some the highest-potential technologies, however, such as bioenergy and agriculture-based wind and solar energy. L18: Closing the yield gap (Chapter 2) and regenerative agriculture (Chapter 3) are not the whole picture to reducing emissions or increasing sequestration. Technical improvements in genetics, pest issues, animal breeds, drought tolerance, water requirements, etc. are also important. L42: Achieving an average increase in storing 1.5 tons per hectare across all of U.S. agricultural soils seems excessively optimistic, as one size does not fit all. Note that this is about half of the number men- tioned in the introduction. It would be useful to compare this estimate with the 0.6 Gt CO2e requirement on L11 in the chapter to say what percent of that emissions reduction goal can come from here. Adding some material on how long this level of soil storage could be continued would be valuable. It would be useful to add some perspective on equilibrium or saturation here and in Chapter 3. The climate science community has long talked about agriculture as a bridge to the future so incorporating some of that perspective would be valuable. (See discussion in Ogle et al., 2023, mentioned in the last section, or McCarl and Sands (2007). McCarl, B. and R. Sands. 2007. Competitiveness of terrestrial greenhouse gas offsets: Are they a bridge to the future? Climatic Change 80(1-2):109-126. L46: It is not clear what control functions refers to here. Is this some mix of regional potential and influencing factors? L51: This makes an important point with respect to the variation in sequestration capacity of different lands and moving away from one size fits all. This paragraph is reminiscent of estimates that made early on in considering soil C sequestration. Subsequently, big differences in technical and economic potential were found that make the assumption of adoption everywhere tenuous. Additionally, the early optimistic papers on practices and results have been tempered by findings that show some practices can lead to nega- tive outcomes. L50: What about cost and prospects for practices really working and being adopted by farmers? It is important to be cautious because overoptimistic statements run the risk of falling short of expectations, and losing credibility, including in trading schemes. L53: Striving for such accuracy of 100 m2 may not be cost-effective, relative to the pursuit of a mix of field measurement, modeling, precision agriculture knowledge of responses, and remote sensing. Accu- rate aggregate estimates based on larger scale sampling and supporting modeling, data and remote sensing are needed. 39 

Review of White Paper on a Scientific Roadmap to Carbon-Negative Agriculture L56: “Translating model output to decision support information will require developing risk-based assessment methods useful to producers in real time.” This statement reveals an on-farm bias. What about support to policy makers asserting compliance with government and international targets, as well as partic- ipants like Microsoft in cap and trade who might one day need to prove compliance. L60: Not sure you need to assert “Nitrogen fertilizer is the most energy intensive and GHG emitting component of modern agriculture….” Driving a tractor and drying grain are likely more intensive and emit- ting per unit. Perhaps it could be said that it is a large or the largest component of emissions and one that can be managed without asserting its relative intensity. L65: The text here says that soil management practices are 94 percent of U.S. nitrous oxide (N2O) emissions. This should be referenced here and included in chapter 4 on managing N2O, for consistency. This is certainly an overestimate. See Xu et al. 2021, who show estimates of agriculture’s contribution to U.S. N2O emissions more on the order of 50%: Xu, R., H. Tian, N. Pan, R. L. Thompson, J. G. Canadell, E. A. Davidson, C. Nevison, W. Winiwarter, H. Shi, S. Pan, J. Chang, P. Ciais, S. R. S. Dangal, A. Ito, R. B. Jackson, F. Joos, R. Lauerwald, S. Lienert, T. Maavara, D. B. Millet, P. A. Raymond, P. Regnier, F. N. Tubiello, N. Vuichard, K. C. Wells, C. Wilson, J. Yang, Y. Yao, S. Zaehle, F. Zhou. 2021. Magnitude and uncertainty of nitrous oxide emissions from North America based on bottom-up and top-down approaches: Informing future research and national inventories. Geophysical Research Letters 48(23):e2021GL095264. https://doi. org/10.1029/2021GL095264. https://agupubs.onlinelibrary.wiley.com/doi/abs/10.1029/2021GL0952 64. L68–69: Hard to accept this view of adoption of precision agriculture. Do we really think this would happen on all farmland especially for the smaller farms and even in the middle of a large field? L62–77: What are the prospects for precision agriculture being adopted on all farmland, especially smaller farms (an average farm is 500 acres), or even in the middle of a large field? Also, it would be helpful to express some of the reduction amounts from precision management adoption in terms of the percent of the total target for reduction as many readers will not carry Tg numbers in their minds. L78: Similar to carbon, a mixture of observation, modeling, data synthesis and remote sensing is likely to be needed for the N2O. L90–92: Reductions in electricity may not be needed if it were possible to move more to wind and solar and maybe manure generated methane. L92: Diesel is not generally a target of substitution with cellulosic biofuel (ethanol), but more bio- diesel. L98: The discussion about energy raises the issue of why off farm sales of farm-derived energy is not well covered in the white paper. Some studies assert that farm-derived energy could be the largest potential agricultural contribution to GHG mitigation in the long run, for example, Murray et al. (2005), previously referenced. L102: The focus on the crop yield gap undersells other technological advances (as noted in comments on chapter 3). The advances noted could be an even broader discussion of technology that could be brought to bear. L105: It is an overstatement to say “The potential reduction in GHG emissions from these activities was provided” as there were some estimates for some alternatives but not an all-encompassing set of num- bers. L113: If reduction in crop emissions is credited to crop production to avoid double counting, is not the estimate that animal agriculture accounts for 39% of total agricultural GHGs double counting? The paragraph does not appear to be internally consistent. L121: “Feed use efficiency with current technologies could reduce GHG emissions by 25% across all animal production systems” is hard to imagine. Maybe feedlots and industrial hogs and poultry but is it realistic to think this applies to all the small cow calf producers? Is aquaculture part of this assessment? 40 

Appendix A L128: The reductions estimated for improved grazing and high adoption estimates push the limits of credibility, given the low input nature of most grazing. L134: Is manure management from grazing operations that feasible? L146: The food waste summary is more credible. But as changes in demand are raised as an influenc- ing factor, it should be described in greater detail with respect to GHG reduction benefits and reduced obesity. In addition, depending on the shift, it might lead to reduced acres in production, or to less beef in the diet, meaning fewer direct emissions and fewer feed emissions. L148–177: This risk analysis of GHG emissions reduction potential could be made part of Chapter 3, which presents the potential that medium and high adoption rates of regenerative practices may have in reducing the carbon footprint of agricultural production. Otherwise, it could be a stand-alone chapter where authors could expand on the ten best ag practices to reduce GHG emissions in medium and high adoption rates. To keep them consistent, there is a need to justify why those were considered the best and connect them to the other chapters. L168: The assumptions on adoption strains credibility. It is not clear if Monte Carlo modeling im- proves confidence if the underlying assumptions are flawed. L170–174: Regarding “The high adoption scenario…”, can there really be confidence in this scenario without actual research into (possible) adoption by producers, and their motivation (or de-motivation)? Which could be argued is mostly based on income. For example, the simple “assumptions” by Department of Energy’s (“Billion Ton Study”) on the total amount of bioenergy crops that could be produced were meaningless since farmers cannot afford to be tied down for many years in a given crop (as is required in a 50-mile radius or so to feed a hypothetical bioenergy facility at scale over its lifetime). The risk is just too high. So, a “billion tons” disappears quickly if no producer will take the contract, leaving shuttered cellu- losic energy facilities like Abengoa in Kansas and elsewhere. Table 9-1: Cellulosic Biofuels Production Max Adoption column. Feedstocks can’t be transported further than about 50 miles to a processing facility (economics), so these estimates can’t be supported. It is not logical to use nationwide numbers. The feedstocks must be grown around (hypothetical) energy plants for 20–30 years (plant lifetime) consistently. Several cellulosic energy plants have closed. Table 9-1: This list does not include some topics described earlier in this chapter and adds things not said above, such as more on energy that is discussed in the preceding paragraphs and less on research directions for narrowing the yield gap. Figs 9-1 and 9-2: The revised Figs 9-1 and 9-2 now have values on the X axis that agree with the text of the original version (L168–174), which wasn’t the case in the document that the committee initially received. However, the locations of the values “5.0%, 90.0% and 5.0% along the tops of the graphs are still confusing. The 90.0% value appears just above and slightly to the right of the left-most vertical line, but that line should be clearly identified as the 5.0% line. The 90.0% value should be centered midway between the two solid vertical lines, indicating that the space between the two lines is the 90% interval. L184: “The next four most effective practices—which included cellulosic biofuel production.…” Cel- lulosic biofuel has proven itself to be unsuccessful in the marketplace, and there are no operating commer- cial facilities, after quite a few years of trying and significant taxpayer investments. Where was the case made (quantitatively, with evidence) in this white paper that the economics and socioeconomics (rejection of contracts by farmers) of cellulosic ethanol have suddenly changed? L199: Although the Monte Carlo analysis is more optimistic, it is difficult to accept the assumptions regarding adoption and scope. What about research to determine how to motivate farmers to do any of the research results from the other bullets? There is much-needed research on socio-economic understanding of farmer decision making. None of the technologies listed will have an impact on GHG emissions if they aren’t adopted by farmers. See Buck and Palumbo-Compton (2022) (referenced previously in comments on Chapter 8) for a discussion of the appalling paucity of socio-economic studies. Otherwise, as far as the list of research and innovation priorities, as noted earlier there are no signs of success for cellulosic ethanol, as the (socio) economics of production don’t align. Similarly, converting from corn to herbaceous biomass crops would need economic studies before it would become a priority. As 41 

Review of White Paper on a Scientific Roadmap to Carbon-Negative Agriculture for on-farm solar and wind, a quick run of the numbers for such capital investments on a farm might show it might be challenging to be an economic success. Finally, where is reducing enteric methane in the list? L207–208: Complete agreement on this point! L212: The text says “soil carbon would need to remain sequestered for 100 years to achieve the GWP goals: the well-known problem of permanence” but the rates of accrual would also need to persist across all 100 years, so the question of a saturation point comes into play. L212–213: It should be acknowledged that GWP20 is just as important a metric as GWP100, and more so for impacts during the next two decades. It would weight CH4 mitigation as more impactful on that time frame. L213–215: See Crow and Sierra (2022) for a method that accounts for gradual decomposition of car- bon sequestered in soil and its release to the atmosphere as CO2. They argue that credits for soil C seques- tration could be adjusted based on knowledge of soil C turnover times for a particular soil. Crow, S.E. and C.A. Sierra. 2022. The climate benefit of sequestration in soils for warming miti- gation. Biogeochemistry 161(1):71-84. https://doi.org/10.1007/s10533-022-00981-1. L213: The committee agrees that “A more appropriate approach might be a mass balance for each category of gas (CO2, CH4, N2O).” L220: “The decision was made to move forward with GWP.…” This needs also be said in the intro- duction, to show the limits and reliability of the study, and clearly state assumptions. This is not a conclu- sion, but a premise, and a debatable one, with tradeoffs for its use in the overall analysis. Figures 9-3 and 9-4: In the original submission, these figures showed “Contribution to Variance” which is interpreted to mean an indication of uncertainty of the effect, not its efficacy. But the text on L180– 182 indicated that these figures show ranking by effectiveness. Perhaps the uncertainty of the effect in- creases with its estimated efficacy? If this is the case, that should be explained. In the revision sent to the committee following a discussion with the authors, Figs 9-3 and 9-4 now indicate that the inputs are ranked by “effect on output mean” but they show bars going to both left and right, still suggesting that they are showing variance and not necessarily the midpoint size of the effect. The legend doesn’t explain the color codes. Perhaps a more appropriate graphic format would be a waterfall diagram, where the size of the bar indicates the parameter’s effect on the predicted mean and an error bar shows the uncertainty of its effect. Waterfall diagrams help show the individual and cumulative effects of parameters. Figure 9-4: One might indicate some sort of error bar, and it would not be assumed that the underlying actual physical carbon flow data would to be better than plus or minus 10%, maybe even 15%? So that limits the need for detailed statistical analysis. Editorial and Minor Issues L6: “…and reduce emissions” of what? Particulates? Add GHG before emissions. 42