1. Underestimates

EPA’s deceptively low number for agricultural greenhouse gas emissions often leads policymakers and journalists to focus on other sectors when considering climate change mitigation strategies.

First, while the impact on climate change for most sectors of the economy stems almost entirely from their production-related greenhouse gas emissions, with agriculture one must also consider the impact of land use. The land footprint of other sectors is insignificant in relation to their emissions and therefore is not considered in EPA’s greenhouse gas inventory. But agriculture’s land footprint is the dominant part of the impact. The use of land for growing crops or raising livestock means that agricultural land—62% of the contiguous United States—cannot be used for other purposes, including those that could have a very different climate impact. Most agricultural land before development was grassland or forest land, which both stored and annually sequestered large amounts of carbon. This lost sequestration capacity of agricultural land is a very real climate impact of agriculture, although one rarely considered. If this impact is included, the total annualized climate change impact of agriculture is approximately 50% bigger than the total agriculture sector emissions in the EPA inventory.³⁷

As one group of scholars explained: “Restoration of native ecosystems, including forests, is a land-based option for atmosphere carbon dioxide removal.

50

currently occupied by animal agriculture is comparable in order of magnitude to the past decade of global fossil fuel emissions.”³⁹

Similarly, other scholars have noted that “standard methods for evaluating the effect of land use on greenhouse gas emissions systematically underestimate the opportunity of land to store carbon if it is not used for agriculture.”⁴⁰ They note that “typical lifecycle assessments, which estimate the [greenhouse gas] costs of a food’s consumption, only estimate land use demand in hectares without translating them into carbon costs. Other [life cycle assessments] consider land use carbon costs only if a food is directly produced by clearing new land ….” A better approach would be to add to the production-related greenhouse gas emissions the “quantity of carbon that could be sequestered annually if [that land] were instead devoted to regenerating forest [or grassland].”⁴¹

Many already acknowledge this opportunity when they note the capacity of U.S. agricultural land to sequester carbon.⁴² In many cases, the land has this capacity to increase carbon stored in vegetation and soils currently because earlier agricultural activities have significantly depleted what had been previously stored prior to cultivation.⁴³ Thus in reality, there is a need to restore land to its pre-agricultural condition to repay this debt before interpreting sequestration as an additional opportunity. While now generally discussed as a future sequestration opportunity (often in the context of proposed payment or offset schemes), this can also be seen as legacy harm in need of repair.

51

countries like the United States, the carbon opportunity cost contributes as much to climate change as all fossil fuel and cement emissions together.⁴⁴

Second, EPA includes on-farm fuel combustion such as for tractors or direct heating in the industrial sector; on-farm electricity for irrigation pumps, cooling, heating, ventilation, and other needs in the electricity sector; and soil carbon lost from conversion of forest or other nonagricultural land to farmland in the consideration of land use. Thus, these emissions are not included in EPA’s calculations for the “agriculture” sector. Nor does EPA’s agricultural tally include emissions related to aquaculture and fisheries, which provide significant amounts of our food.⁴⁵ On-farm fuel combustion in 2018 contributed about 40 MMT CO₂ eq.,⁴⁶ as did the indirect emissions of on-farm electricity use, while land annually converted for agricultural use released 56 MMT CO₂ eq.⁴⁷ All told, these additional elements of agriculture’s greenhouse gas emissions increase the sector’s share to about 11%. This total does not include upstream and downstream food system emissions such as emissions associated with the manufacture of fertilizer (discussed below in Chapter VIII and itself adding at least one-half percent of total U.S. greenhouse gas emissions), refrigeration and transport of food, and managing food waste, which, if included, would bring the U.S. food system’s total carbon footprint much higher.⁴⁸ At the global scale, as noted above, approximately one-third of all greenhouse gas emissions are attributed to the food system.⁴⁹

Third, calculating agriculture’s climate change contribution is also complicated by the fact that, unlike the energy and transportation sectors, which emit primarily CO₂ as fossil fuels are burned, crop and livestock greenhouse gas emissions consist largely of N₂O and CH₄.

52

climate impact implicates fundamental policy choices. N₂O, largely released as a result of fertilizer that is applied but not taken up by crops, is a particularly potent greenhouse gas, with an average global warming potential of 265-298 times that of CO₂ over 100 years.⁵⁰ Whether a calculation uses the lower or the higher number of that range for N₂O’s global warming potential creates about a 10% variation in its relative contribution to climate change.⁵¹

Additionally, a 2016 study found that the Intergovernmental Panel on Climate Change’s (IPCC’s) Fifth Assessment Report underestimated CH₄’s global warming potential by 20%-25% because its methods did not take into account the absorption of shortwave radiation by CH₄, among other factors.⁵² The study’s author estimates that the IPCC’s Sixth Assessment Report may revise CH₄’s 100-year global warming potential to 35 or higher.⁵³

Calculating CH₄’s global warming potential is further complicated by the fact that methane breaks down relatively quickly compared to N₂O or CO₂. The global warming potential of methane is about 84-86 times that of CO₂ over 20 years.⁵⁴ EPA, however, uses a longer time horizon for calculating the global warming potential of CH₄, reducing the relative impact of agriculture’s total emissions by more than half. Instead of determining the CO₂ equivalent of CH₄ by comparing the two gases over a 20-year time span, EPA’s report follows the IPCC’s Fourth Assessment Report in using a 100-year time span. This significantly lowers CH₄’s global warming potential since CH₄’s potency declines relatively quickly. As a result, EPA’s estimate assumes that CH₄ has only 25 times the radiative impact of CO₂.⁵⁵ The IPCC’s Fifth Assessment Report, however, not only increased the 100-year global warming potential of CH₄ to 28-34 times that of CO₂,⁵⁶ but also supports the use of a 20-year timescale for measuring the impact of emissions from agriculture.⁵⁷ While a 100-year time period for CH₄ is still commonly used in scientific discussions, policy debates increasingly use a 20-year period

53

54

due to the urgent need to reduce CH₄ emissions over the next 10-30 years.⁵⁸ For example, New York’s Climate Leadership and Community Protection Act requires use of the 20-year time frame for analysis and policy development, which time frame increases CH₄ share of the state’s total greenhouse gases by 3.4 times.⁵⁹ If EPA had calculated agricultural emissions using a 20-year time horizon, its estimate would nearly double, from 619 to 1,216 MMT CO₂ eq.