Ruminant density increases over time in all GLOBIOM scenarios, with the greatest increases in healthier U.S. diet scenarios . Since cattle are smaller at the point of slaughter in these scenarios with lower beef productivity, it is possible to increase the number of animals per unit of land as each animal requires less feed. The constant and lower ruminant density scenarios track the current declining trend of grazing intensity in the U.S. This lower intensity trend may also continue if consumer demand for pasture-raised and finished beef continues to rise. These changes increase the amount of land required by cattle and provide lower bound estimates for pastureland reduction . Lower ruminant density, despite dampening the effect of reduced beef consumption, does result in total emission reductions . For land use change CO2 emissions, constant and lower ruminant density closely resembles results for GLOBIOM scenarios, whereas higher livestock productivity Calculator sensitivity scenarios estimate greater land sequestration compared to GLOBIOM scenarios , suggesting that livestock productivity values may play an important role in determining the sustainability of reducing beef consumption. As expected, changes to either beef productivity or cattle density do not impact cropland or crop-related emissions. Despite changes in productivity and ruminant density, overall land use efficiency of beef production are very similar across GLOBIOM scenarios and increases over time . Increasing rates of beef productivity increase this efficiency by 35–53% in 2050, and reducing ruminant density reduces this efficiency by 21–28%. With healthier diets, domestic production and consumption of livestock and corn decline significantly,hydroponic bucket and the U.S. increases its export share. Consumption, production, import, and export patterns for the top three livestock and crop commodities explain the overall trends in land use and emissions .
Consistent with healthier diet scenario assumptions, U.S. per-capita consumption of beef declines by approximately 50%, while poultry and pork consumption fall by more than two-thirds . Under Healthy U.S. scenarios, corn and soybean consumption fall with lower demand for livestock feed and dietary shifts away from oil seeds and sugar . Corn and soybean consumption are constant to slightly increasing for the U.S. Yields and Healthy ROW scenarios, with the latter being driven by lower prices as the rest of the world shifts to healthier diets. Wheat consumption increases up to 14.6% in 2050 as diets shift to a higher proportion of cereals. Production changes for livestock and corn mirror shifts in consumption, but soybean and wheat production vary considerably across scenarios . Soybean production changes range from a slight decrease to an increase of up to 27% under Healthy U.S. and U.S. Yields. The variability in soybean production differences is primarily driven by yield changes, while wheat production differences are driven primarily by diet assumptions . U.S. wheat production declines relative to the baseline under the Healthy ROW scenario as dietary changes in other regions cause cereal production to increase outside the U.S.However, meat exports increase for all other scenarios, except Healthy ROW , relative to the Baseline . Reduced demand outside of the U.S. causes global prices to fall, reducing the competitive edge that U.S. beef production has in other scenarios. Exports for corn, wheat, and soybeans also increase for all combinations with Healthy U.S. and U.S. Yields. A shift to healthier diets in the U.S. shifts the final disposition of crop production to a higher volume of exports, while higher productivity enhances U.S. comparative advantage. In scenarios where the U.S. moves to healthier diets, by 2050 the U.S. shifts from a net importer of beef to a net exporter.
We find that healthier diets in the U.S. can complement sustainability and climate change goals, both directly and indirectly. Reduced feed grain and meat production results in the direct benefit of lower non-CO2 emissions from U.S. crop and livestock systems. This direct mitigation is supported by additional indirect sources of mitigation from land use and management that occur in response to changing diets and associated market shifts. These land use changes, including reforestation of retired pastureland and cropland, increases carbon sequestration and reduces net emissions relative to the baseline. It is important to distinguish between rangeland that historically was grazed and pastureland that used to be forest. Grazing on well-managed rangeland can promote habitat values directly by limiting woody encroachment and maintaining plant diversity and indirectly by providing economic returns that prevent the conversion of the land to other uses such as urban development. Biodiversity conservation benefits of pasture abandonment can be achieved by targeting lands for reforestation. A recent analysis identified 32 Mha of current pasture that used to be forest and thus would be candidate areas for reforestation . Additional sensitivity analysis with the US FABLE Calculator shows that livestock sector intensification and improved livestock productivity could significantly support further mitigation. Production intensification and higher animal productivity frees up additional pastureland . In the Calculator sensitivity results, some of this land is converted to forests, providing additional indirect mitigation benefits that outweigh the direct non-CO2 emissions reduction from dietary shifts. In isolation, healthier diets reduce the demand for meat products, which theoretically reduces prices and production intensity on a per head of livestock basis . However, this relationship between demand and productivity may not hold in industries experiencing reduced production .
Thus, measures that maintain or boost livestock intensity and productivity could counter these impacts and spare land for conservation or climate mitigation purposes. However, constant or declining ruminant density sensitivity results, which reduce beef land use efficiency by 25% compared to 2020, provide an indication of the possible attenuated pastureland reductions if demand for beef is both reduced and shifts toward pasture-fnished products. Pasture-fed beef systems have 30% lower land use efficiency but have lower total GHG emissions if assuming increased soil organic carbon sequestration . Conversely, our results indicate that increasing U.S. crop productivity in the future may have a sizable land sparing effect globally, but only modestly in the U.S. In our simulations, higher crop productivity alone results in increased production and additional land used for crop production in the U.S. between 2040 and 2050. In GLOBIOM, higher crop productivity expands the U.S. comparative advantage in cropping systems, leading to increased exports of staple crops such as corn and soybeans, regardless of dietary preferences. Thus, regardless of whether healthier diets shift regional demands or productivity growth improves the U.S. comparative advantage in agricultural systems, interactions between the U.S. land use system and global markets should be accounted for in sustainability assessments. Ignoring these market interactions could result in biased environmental impact projections of policies,stackable planters technological improvements, or demand-side changes . Our results are consistent with studies that have explored healthier diet transitions in the U.S., as summarized in SM Table S9. We project a general decline in U.S. agricultural land demand under dietary transitions that reduce the domestic demand for meats and feed grains, consistent with 45% and 19% reduction in Behrens et al. and Birney et al. , respectively . Our GHG estimates are also in line with Behrens et al. , but are greater than Birney et al. , Hitaj et al. and Tom et al. , which report the same or greater carbon emissions from adopting the DGA . The LCA approaches used in these studies do not capture industry adjustments of healthier diets and use static levels of input use intensity. However, we note that of the four studies reporting GHG emissions of healthier U.S. diets, the two that used an environmentally extended input–output modeling approach, which are more comprehensive than processed-based approaches, found lower carbon emissions associated with healthier diets . We distinguish our results from previous LCA studies as our use of partial equilibrium and land use/GHG accounting modeling illustrates how healthier, but still omnivorous, diet transitions can alter markets and thus land management decisions at the intensive and extensive margins. Similar to EAT-Lancet , we show that healthier diet transitions could have other environmental co-benefits by reducing total nitrogen and phosphorus application under most healthier diet scenarios , though further analysis is needed to better understand the spatial distribution of nutrient application changes. While diets are a personal choice, government policy has more influence on diets than is widely appreciated. The federal government provides dietary guidelines and the government directly pays for meals for 30 million school children through the free and reduced lunch program, 2 million incarcerated people, 2 million active military members, and 2 million people in nursing homes or assisted living.
Guidelines are used by non-federal organizations to provide information on diet and health to the general public, influencing the dietary preferences of millions of additional Americans. Government food and agriculture policy has historically helped drive large shifts in diets and can do so again .There are important limitations of this study that require future research and model development. First, it is important to note that our dietary scenarios are not isocaloric. We designed the healthy U.S. diets to reflect not only changes in food group composition , but also to reduce caloric intake which reduces diseases associated with obesity . In our results, total U.S. per capita caloric intake declines by 7% in 2050 under the healthier U.S. diet scenarios, which better matches the EAT-Lancet recommendations . While this is not strictly a limitation, as we intentionally defined ‘healthy’ more holistically , it does reduce the comparability of our study results with those that isolate the effects of diet composition alone. Second, our approach for diet modeling may differ from other studies. To use the Calculator and GLOBIOM modeling approaches, we converted USDA Dietary Guidelines for Americans volumetric and weight-based recommendations for aggregate food groups into caloric equivalents using representative foods within those groups. The existing baseline diet in both models are based on the Food and Agriculture Organization reporting of total U.S. human demand for crop commodities. Thus, differences between the modeled healthier U.S. diet and the baseline U.S. diet may not be exactly the same as related U.S.- specific diet studies . Finally, each model is limited in its representation of U.S. food systems and the variety of food options available to consumers. For example, both the FABLE Calculator and this version of GLOBIOM do not account for calories from fish, and GLOBIOM does not offer a detailed breakdown of fruit and vegetable production systems. Healthier diets demand more alternative proteins and fruits and vegetables, which could drive the allocation of cropland toward a higher proportion of specialty crops. Importantly, some of these crops create other resource challenges, such as the high water demand of almonds. Representing specialty crops spatially in partial equilibrium frameworks is an important research gap that warrants future work and model development. Furthermore, each model is also limited in its representation of the U.S. forestry system, with minimal differentiation between planted/managed forest systems and unmanaged systems and their respective carbon dynamics. Responses to policy signals that influence the allocation of land between agriculture and forestry could also alter the distribution of passively and actively managed forest resource systems with different associated carbon profiles. Future work will be necessary to improve the representation of fruit and vegetable production systems and forest management dynamics in the U.S. for a more comprehensive sustainability policy assessment. Notably, we do not quantify a full suite of benefits and costs associated with healthier diet transitions, including improved health outcomes and distributional implications of agricultural transitions on rural communities . Future analysis should carefully consider a wider-range of policy implications to offer a more holistic perspective on healthier diet transitions. Future research on healthier diet transitions in the U.S. can build on this analysis by quantifying a broader range of economic costs and benefits of healthier diet transitions. Furthermore, this analysis assumes dietary change occurs through general preference shifts and does not consider the costs or associated trade offs of incentive structures needed to support dietary change. We also do not consider potential research and development costs needed to increase crop and livestock yields and to facilitate technology transfer. While this omission does not limit the scope of our current analyses, the magnitude of these costs remains an important policy design question. Nor do we quantify the sustainability and climate benefits of productivity improvement through novel technologies that also reduce input use intensity.