Articles https://doi.org/10.1038/s43016-022-00573-0 Global food insecurity and famine from reduced crop, marine fishery and livestock production due to climate disruption from nuclear war soot injection Lili Xia 1 ✉, Alan Robock 1, Kim Scherrer 2,3, Cheryl S. Harrison4, Benjamin Leon Bodirsky 5,6, Isabelle Weindl 5, Jonas Jägermeyr 5,7,8, Charles G. Bardeen9, Owen B. Toon10 and Ryan Heneghan11 Atmospheric soot loadings from nuclear weapon detonation would cause disruptions to the Earth’s climate, limiting terrestrial and aquatic food production. Here, we use climate, crop and fishery models to estimate the impacts arising from six scenarios of stratospheric soot injection, predicting the total food calories available in each nation post-war after stored food is consumed. In quantifying impacts away from target areas, we demonstrate that soot injections larger than 5 Tg would lead to mass food shortages, and livestock and aquatic food production would be unable to compensate for reduced crop output, in almost all countries. Adaptation measures such as food waste reduction would have limited impact on increasing available calories. We estimate more than 2 billion people could die from nuclear war between India and Pakistan, and more than 5 billion could die from a war between the United States and Russia—underlining the importance of global cooperation in preventing nuclear war. E xtraordinary events such as large volcanic eruptions or nuclear weapons would have smaller global impacts17, once a nuclear war war could cause sudden global climate disruptions and affect starts, it may be very difficult to limit escalation18. food security. Global volcanic cooling caused by sulfuric acid The scenarios we studied are listed in Table 1. Each scenario aerosols in the stratosphere has resulted in severe famines and polit- assumes a nuclear war lasting one week, resulting in the number and ical instability, for example, after the 1783 Laki eruption in Iceland1 yield of nuclear weapons shown in the table and producing different or the 1815 Tambora eruption in Indonesia2,3. For a nuclear war, amounts of soot in the stratosphere. There are many war scenar- the global cooling would depend on the yields of the weapons, the ios that could result in similar amounts of smoke and thus similar number of weapons and the targets, among other atmospheric and climate shocks, including wars involving the other nuclear-armed geographic factors. In a nuclear war, bombs targeted on cities and nations (China, France, United Kingdom, North Korea and Israel). industrial areas would start firestorms, injecting large amounts of Recent catastrophic forest fires in Canada in 201719 and Australia soot into the upper atmosphere, which would spread globally and in 2019 and 202020,21 produced 0.3–1 Tg of smoke (0.006–0.02 Tg rapidly cool the planet4–6. Such soot loadings would cause decadal soot), which was subsequently heated by sunlight and lofted high in disruptions in Earth’s climate7–9, which would impact food produc- the stratosphere. The smoke was transported around the world and tion systems on land and in the oceans. In the 1980s, there were lasted for many months. This adds confidence to our simulations investigations of nuclear winter impacts on global agricultural pro- that predict the same process would occur after nuclear war. duction10 and food availability11 for 15 nations, but new informa- Nuclear war would primarily contaminate soil and water close tion now allows us to update those estimates. Several studies have to where nuclear weapons were used22. Soot disperses globally once recently analysed changes of major grain crops12–14 and marine wild it reaches the upper atmosphere; thus, our results are globally rel- catch fisheries15 for different scenarios of regional nuclear war using evant regardless of the warring nations. Here, we focus on the cli- climate, crop and fishery models. A war between India and Pakistan, mate disruption from nuclear war, which would impact global food which recently are accumulating more nuclear weapons with higher production systems on land and in the oceans. So far, an integrated yield16, could produce a stratospheric loading of 5–47 Tg of soot. estimate of the impacts of the entire range of war scenarios on both A war between the United States, its allies and Russia—who pos- land- and ocean-based food production is missing. We examine the sess more than 90% of the global nuclear arsenal—could produce impacts of six war scenarios, generating 5 Tg to 150 Tg of soot, on more than 150 Tg of soot and a nuclear winter4–9. While amounts the food supply (Table 1). We use model simulations of major crops of soot injection into the stratosphere from the use of fewer nuclear and wild-caught marine fish together with estimated changes in Department of Environmental Sciences, Rutgers University, New Brunswick, NJ, USA. 2Institut de Ciència i Tecnologia Ambientals, Universitat Autònoma 1 de Barcelona, Cerdanyola del Vallès, Spain. 3Department of Biological Sciences, University of Bergen, Bergen, Norway. 4Department of Ocean and Coastal Science, Center for Computation and Technology, Louisiana State University, Baton Rouge, LA, USA. 5Potsdam Institute for Climate Impact Research, Potsdam, Germany. 6World Vegetable Center, Tainan, Taiwan. 7NASA Goddard Institute for Space Studies, New York, NY, USA. 8Center for Climate Systems Research, Columbia University, New York, NY, USA. 9National Center for Atmospheric Research, Boulder, CO, USA. 10Laboratory for Atmospheric and Space Physics, Department of Atmospheric and Ocean Sciences, University of Colorado, Boulder, CO, USA. 11School of Mathematical Sciences, Queensland University of Technology, Brisbane, Queensland, Australia. ✉e-mail: lilixia@envsci.rutgers.edu 586 Nature Food | VOL 3 | August 2022 | 586–596 | www.nature.com/natfood Nature Food Articles cold stress14. Calorie reduction from agriculture and marine fish- Table 1 | Number of weapons on urban targets, yields, direct eries shows regional differences (Supplementary Fig. 4), with the fatalities from the bomb blasts and resulting number of people strongest percentage reductions over high latitudes in the Northern in danger of death due to famine for the different scenarios we Hemisphere. Even for the India–Pakistan case, many regions studied become unsuitable for agriculture for multiple years. For example, Soot Number of Yield Number of direct Number of people in the 27 Tg case, mid- to high latitudes of the Northern Hemisphere (Tg) weapons (kt) fatalities without food at the show reductions in crop calorie production greater than 50%, along end of Year 2 with fish catch reductions of 20–30%. The nuclear-armed nations in mid- to high latitude regions (China, Russia, United States, France, 5 100 15 27,000,000 255,000,000 North Korea and United Kingdom) show calorie reductions from 16 250 15 52,000,000 926,000,000 30% to 86%, and in lower latitudes (India, Pakistan and Israel), the 27 250 50 97,000,000 1,426,000,000 reduction is less than 10% (Supplementary Tables 1 and 2). Impacts in warring nations are likely to be dominated by local problems, 37 250 100 127,000,000 2,081,000,000 such as infrastructure destruction, radioactive contamination and 47 500 100 164,000,000 2,512,000,000 supply chain disruptions, so the results here apply only to indirect 150 4,400 100 360,000,000 5,341,000,000 effects from soot injection in remote locations. 150 4,400 100 360,000,000 a 5,081,000,000 Impacts on total human calorie intake. To estimate the effect on The 5 Tg case scenario is from ref. 16 for an India–Pakistan war taking place in 2008; the 16–47 Tg cases are from ref. 18 for an India–Pakistan war taking place in 2025; and the 150 Tg case is from the total food calories available for human consumption, we con- ref. 51, which assumes attacks on France, Germany, Japan, United Kingdom, United States, Russia sider diet composition, calorie content of different food types, crop and China. The last column is the number of people who would starve by the end of Year 2 when usage and changes in food production that we did not directly the rest of the population is provided with the minimum amount of food needed to survive, assumed to be a calorie intake of 1,911 kcal per capita per day, and allowing for no international model (Methods). In 2010, FAO23 reported that 51% of global trade; from Supplemental Information, Supplementary Table 5, the Partial Livestock case, in which calorie availability came from cereals, 31% from vegetables, fruit, 50% of livestock grain feed is used for human consumption, and 50% of livestock grain feed is roots, tubers and nuts and 18% from animal and related products, used to raise livestock, using the latest complete data available for the year 2010. For 2010, the total population of the nations used in this study was 6,700,000,000. There are many other of which fish contributed 7%, with marine wild catch contributing scenarios in which these amounts of soot could be produced by a nuclear war, and the scenarios 3% (Fig. 3a). The crops and fish we simulated provide almost half of we use are only meant to be illustrative examples. The last column is the case with the fewest these calories and 40% of the protein. Further, only portions of the number of deaths without international trade, and other cases are available in the Supplementary simulated food production are available for human consumption. Information. aAssuming total household waste is added to food consumption. Many crops (for example, maize and soybean) are used mainly for non-food uses such as livestock feed (Fig. 3c). In addition, the total number of calories available as food is other food and livestock production to assess the impacts on global highly dependent on human reactions to nuclear wars. We assume calorie supply. that international trade in food is suspended as food-exporting nations halt exports in response to declining food production Results (Methods). Furthermore, we considered three societal responses, Impacts on crops and fish catch productivity. Using climate, crop Livestock, Partial Livestock and No Livestock (Supplementary Table and fishery models (Methods), we calculate calorie production for 3). For the Livestock response scenario, representing a minimal different food groups, for each year after a range of six different adaptation to the climate-driven reduction in food production, stratospheric soot injections. The climatic impacts would last for people continue to maintain livestock and fish as normal. Although about a decade but would peak in the first few years (Fig. 1). harvesting a larger share of crop residues for feed or adding new Global average calorie production from the crops we simulated feed such as insect-based supplements may increase the potential decreased 7% in years 1–5 after the war even under the smallest, 5 Tg livestock feed, we assume that no new feed supplements are added soot scenario (Fig. 2a; comparable to previous multi-model results14, and the ratio of agricultural grains, residues and grazed biomass to Supplementary Fig. 2) and up to 50% under the 47 Tg scenario. In livestock feed is the same. Calories from all crops are reduced by the the 150 Tg soot case, global average calorie production from crops average reduction in our four simulated crops, and calorie changes would decrease by around 90% 3–4 years after the nuclear war. The from marine wild-caught fish are calculated with business-as-usual changes would induce a catastrophic disruption of global food mar- fishing behaviour. The No Livestock response represents a scenario kets, as even a 7% global yield decline compared with the control where livestock (including dairy and eggs) and aquaculture produc- simulation would exceed the largest anomaly ever recorded since tion are not maintained after the first year, and the national frac- the beginning of Food and Agricultural Organization (FAO) obser- tions of crop production previously used as feed are now available vational records in 196114. to feed humans. In addition, fishing pressure intensifies, simulated Fish are another important food resource, especially in terms through a fivefold increase in fish price15. Similar responses took of protein supply. Nuclear war would reduce the wild fish catch15, place in New England in the ‘year without a summer’ after the 1815 but the reduction would be less than for land agriculture (Fig. 2b), Tambora volcanic eruption2. Even though the temperature changes because reduction in oceanic net primary productivity—the base were smaller than modelled in any of the nuclear war scenarios of the marine food web—is moderate (from 3% in 5 Tg to 37% here, crop failures forced farmers to sell their livestock because they in 150 Tg), and ocean temperature changes are less pronounced could not feed them3, and previously unpalatable fish were added (Fig. 1). Terrestrial crop production dominates the total calorie to their diet2,3,25. We test a full range (0–100%) of the fraction of change of crops and fisheries combined (Fig. 2c), because global food-competing feed26 that could be used by humans and select crop production is 24 times higher than wild fisheries in terms of 50% as an example in some plots and tables. Between the Livestock dry matter, and staple crops contain around five times more calo- and No Livestock cases, we also consider a Partial Livestock case, in ries than fish per unit retail mass23,24. In total, marine wild capture which the remaining portion of livestock grain feed after converting fisheries contribute 0.5% of total calories but 3.5% of global average to human consumption would be used for raising livestock. protein supply (Fig. 3 and Supplementary Fig. 3). Final biofuel products (biodiesel and ethanol) only count for Cooling from nuclear wars causes temperature limitations for 0.5% of the plant-based products27, which could be repurposed crops, leading to delayed physiological maturity and additional as food in the form of plant oil (~1.8% of total food calories) and Nature Food | VOL 3 | August 2022 | 586–596 | www.nature.com/natfood 587 Articles Nature Food a Cropland 2 m air temperature b Sea surface temperature 2 2 0 0 Sea surface temperature (°C) –2 2 m air temperature (°C) –2 –4 –4 –6 –6 5 Tg –8 16 Tg –8 –10 27 Tg –10 37 Tg –12 47 Tg –12 150 Tg –14 –14 –16 –16 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Year after nuclear war Year after nuclear war c Cropland solar radiation d Ocean solar radiation Change relative to control simulation (%) Change relative to control simulation (%) 0 0 0 0 –30 –16 –30 –17 Solar radiation (W m–2) Solar radiation (W m–2) –60 –32 –60 –34 –90 –48 –90 –51 –120 –64 –120 –69 –150 –80 –150 –86 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Year after nuclear war Year after nuclear war e Cropland precipitation f Ocean net primary productivity 1.0 20 Net primary productivity (ton C per month) 0.5 17 Change relative to control simulation (%) Change relative to control simulation (%) 0.5 10 0 0 Precipitation (mm d–1) 0 0 –0.5 –17 –0.5 –10 –1.0 –34 –1.0 –20 –1.5 –30 –1.5 –51 –2.0 –40 –2.0 –68 –2.5 –50 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Year after nuclear war Year after nuclear war Fig. 1 | Climatic impacts by year after different nuclear war soot injections. a–f, Changes in surface temperature (a), solar radiation (c) and precipitation (e) averaged over global crop regions of 2000 (Supplementary Fig. 1) and sea surface temperature (b), solar radiation (d) and net primary productivity (f) over the oceans following the six stratospheric soot-loading scenarios studied here for 15 years following a nuclear war, derived from simulations in ref. 18. These variables are the direct climate forcing for the crop and fishery models. The left y axes are the anomalies of monthly climate variables from simulated nuclear war minus the climatology of the control simulation, which is the average of 45 years of simulation. The right y axes are the percentage change relative to the control simulation. The wars take place on 15 May of Year 1, and the year labels are on 1 January of each year. For comparison, during the last Ice Age 20,000 years ago, global average surface temperatures were about 5 °C cooler than present. Ocean temperatures decline less than for crops because of the ocean’s large heat capacity. Ocean solar radiation loss is less than for crops because most ocean is in the Southern Hemisphere, where slightly less smoke is present. alcohol (3.4% of total food calories). Byproducts of biofuel have Fig. 4). Here, we focus on two calorie intake levels in nations: calo- been added to livestock feed and waste27. Therefore, we add only the rie intake to maintain normal physical activity and calorie intake calories from the final product of biofuel in our calculations. Global lower than the basal metabolic rate (also known as the resting averaged household waste is around 20% (ref. 28). If we assume that energy expenditure)29. The two levels vary in countries depending after a nuclear war there would be 50% less or 100% less household on the composition and physical activity of the population. Food waste, these extra calories would become available. consumption of less than the first level would not allow a person National consequences of calorie loss depend on the amount of to maintain their normal physical activity and keep their weight at fallow cropland, regional climate impacts, population levels and the same time, and less than the basal metabolic rate would cause assuming a complete halt of international food trade (Methods; fast weight loss even with only sedentary activity and thus would 588 Nature Food | VOL 3 | August 2022 | 586–596 | www.nature.com/natfood Nature Food Articles a Crop b Marine fish 0 0 Available caloric change (%) Available caloric change (%) –20 –20 –40 –40 5 Tg 16 Tg –60 –60 27 Tg 37 Tg –80 47 Tg –80 150 Tg ref. 14 –100 –100 1 2 3 4 5 6 7 8 9 10 11 12 13 14 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Year after nuclear war Year after nuclear war c Crop and marine fish d Grass 0 0 Annual leaf carbon change (%) Available caloric change (%) –20 –20 –40 –40 –60 –60 –80 –80 –100 –100 1 2 3 4 5 6 7 8 9 10 11 12 13 14 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Year after nuclear war Year after nuclear war Fig. 2 | Calorie production changes for crops and fish, and accumulated carbon change for grasses following different nuclear war soot injections. a–c, Global average annual crop calorie production changes (%; maize, wheat, rice and soybeans, weighted by their observed production (2010) and calorie content; a), marine fish production changes (%; b) and combined crop and fish calorie production changes (%; c) after nuclear war for the different soot-injection scenarios. d, Grass leaf carbon is a combination of C3 and C4 grasses, and the change is calculated as annual accumulated carbon. For context, the grey line (and shaded area) in a are the average (and standard deviation) of six crop models from the Global Gridded Crop Model Intercomparison (GGCMI, ref. 14) under the 5 Tg scenario. CLM5crop shows a conservative response to nuclear war compared with the multi-model GGCMI response. quickly lead to death29. With a 5 Tg injection, most nations show experience smaller impacts than other countries. But if this scenario decreasing calorie intake relative to the 2010 level but still suffi- should actually take place, Australia and New Zealand would prob- cient to maintain weight (Fig. 4 and Supplementary Fig. 5). With ably see an influx of refugees from Asia and other countries experi- larger soot-injection cases, severe starvation occurs in most of the encing food insecurity. mid–high latitude nations under the Livestock Case. When 50% of The global average calorie supply post-war (Fig. 5a) implies that food-competing feed is converted for human consumption in each extreme regional reductions (Fig. 4) could be overcome to some nation, some nations (such as the United States) would maintain extent through trade—but equal distribution of food globally would sufficient calorie intake under scenarios with smaller soot injec- probably be a major challenge. Under the Livestock case, if food were tions, but weight loss or even severe starvation would occur under evenly distributed over the world and household waste were 20% larger soot-injection cases (Fig. 4 and Supplementary Fig. 5). as in 201028, there would be enough food for everyone under the Under the 150 Tg scenario, most nations would have calorie 5 Tg scenario to support their normal physical activity; if household intake lower than resting energy expenditure29. One exception is waste were reduced from 20% to 10%, extra calories would support Australia. After we turn off international trade, wheat contributes everyone under the 16 Tg scenario; and if there were no household almost 50% of the calorie intake in Australia, and production of rice, waste, even under the 27 Tg case, everyone would consume sufficient maize and soybean in Australia are less than 1% that of wheat23,24. calories for survival. With the most optimistic case—100% livestock Therefore, the wheat response to simulated nuclear wars largely crop feed to humans, no household waste and equitable global food determines calorie intake in Australia. Because spring wheat is distribution—there would be enough food production for every- used to represent wheat, and simulated spring wheat there shows one under the 47 Tg case. Assuming international trade ceased and increasing or small reductions under nuclear war scenarios in food was distributed optimally within each country11, such that which more favourable temperatures occur for food production, the the maximum number of people were given the calorie intake to calorie intake in Australia is more than other nations. However, this maintain their weight and normal physical activity28, the percentage analysis is limited by the FAO data, which are collected at national of population that could be supported can be calculated (Fig. 5b levels. Within each nation, particularly large ones, there may be and Supplementary Fig. 6b). Under the 150 Tg case, most countries large regional inequities driven by infrastructure limitations, eco- would have less than 25% of the population survive by the end of nomic structures and government policies. New Zealand would also Year 2 (Supplementary Fig. 7). In 2020, 720 million–811 million Nature Food | VOL 3 | August 2022 | 586–596 | www.nature.com/natfood 589 Articles Nature Food a Global averaged diet composition b Global averaged protein composition (FAO, 2010) (FAO, 2010) e Rice Ric Soybean ize Ma Other cereals 42% Wh Wh e 51% iz e Ma e at at Soybean Fruit & veg. 2,855 Other fish 80 Other cereals kcal per capita 19% g per capita per day Other fish per day ds Marine fish See . g ve Liv 18% Marine fish e it & sto 31% Fru Roots 39% ds ck Cereals and soybean Alcohol ck Livesto See Vegetables, fruit, roots, tubers and seeds Alcohol Roots Meat, fish, dairy and eggs c d Usage of crop-based products Crop and marine fish production and usage (FAO, 2010) Food 42.8 Maiz e Food Feed_monogastrics 24.0 Feed Crop-based products Industry Feed_ruminants 8.6 Rice 81% Others 18% Loss 17.6 Marine fish Final products of biofuel 0.5 Maize % Seed 2.5 13 69% W at he Rice n ea yb So Soybean Others 4.0 Wheat 0 10 20 30 40 50 Marine fish % of dry matter crop-based production Fig. 3 | Global average human diet and protein composition and usage of crop-based products. a, Global average human diet composition23. Percentages are % of available calories. Veg. is vegetables. b, Global average human protein diet composition23. Marine wild capture contributes 75% of marine fish46. Percentages are % of dry matter production. c, Distribution of four major cereal crops and marine fish between human food and other uses24,47. Percentages are % of dry matter production. d, Usage of crop-based products in 2010 (% of dry matter crop-based production)26. The colour gradient legend in grey in c illustrates the usage of different crops and fish in colours. While humans consume most of the wheat and rice grown, most maize and soybeans are used for livestock feed. people suffered from undernutrition worldwide30, despite food pro- reduction in overall production. We find particularly severe crop duction being more than sufficient to nourish a larger world popu- declines in major exporting countries such as Russia and the United lation. Thus, it is likely that food distribution would be inequitable States, which could easily trigger export restrictions and cause both between and within countries. severe disruptions in import-dependent countries24. Our no-trade response illustrates this risk—showing that African and Middle Discussion Eastern countries would be severely affected. Using state-of-the-art climate, crop and fishery models, we calculate Our analysis of the potential impacts of nuclear war on the how the availability of food supplies could change globally under food system does not address some aspects of the problem, leaving various nuclear war scenarios. We combine crops and marine fish them for future research. In all the responses, we do not consider and also consider whether livestock and animal products continue reduced human populations due to direct or indirect mortality and to be an important food source. possible reduced birth rate. The total number and composition of For a regional nuclear war, large parts of the world may suffer population changes would affect available labour, calorie produc- famine—even given the compensating behaviours considered in tion and distribution. Also, we do not consider farm-management this paper. Using crops fed to livestock as human food could off- adaptations such as changes in cultivar selection, switching to more set food losses locally but would make limited impacts on the total cold-tolerating crops or greenhouses31 and alternative food sources amount of food available globally, especially with large atmospheric such as mushrooms, seaweed, methane single cell protein, insects32, soot injections when the growth of feed crops and pastures would be hydrogen single cell protein33 and cellulosic sugar34. Although severely impaired by the resulting climate perturbation. Reducing farmer adaptation35 and alternative food sources could reduce the household food waste could help in the small nuclear war cases negative impact from a simulated nuclear war, it would be challeng- but not in the larger nuclear wars due to the large climate-driven ing to make all the shifts in time to affect food availability in Year 2, 590 Nature Food | VOL 3 | August 2022 | 586–596 | www.nature.com/natfood Nature Food Articles Livestock case Partial Livestock case No Livestock case (no trade) (no trade) (no trade) 5 Tg 16 Tg 27 Tg 37 Tg 2010, no trade 47 Tg Starving Normal Losing weight 150 Tg 150 Tg + 50% food waste 150 Tg + 100% food waste Fig. 4 | Food intake (kcal per capita per day) in Year 2 after different nuclear war soot injections. The left map is the calorie intake status in 2010 with no international trade; the left column is the Livestock case; the middle column is the Partial Livestock case, with 50% of livestock feed used for human food and the other 50% still used to feed livestock; and the right column is the No Livestock case, with 50% of livestock feed used for human food. All maps assume no international trade and that the total calories are evenly distributed within each nation. Regions in green mean food consumption can support the current physical activity in that country; regions in yellow are calorie intake that would cause people to lose weight, and only sedentary physical activity would be supported; and regions in red indicate that daily calorie intake would be less than needed to maintain a basal metabolic rate (also called resting energy expenditure) and thus would lead to death after an individual exhausted their body energy reserves in stored fat and expendable muscle28. 150 Tg + 50% waste is half of the household waste added to food consumption, and 150 Tg + 100% waste is all household waste added to food consumption. and further work should be done on these interventions. Current micronutrients to survive the ensuing years of food deficiency (we food storage can alleviate the shortage in Year 1 (ref. 14) but would estimate the impact on protein supply in Supplementary Fig. 3). have less impact on Year 2 unless it were rationed by governments Large-scale use of alternative foods, requiring little-to-no light to or by the market. Expanding or shifting cropping land to favour- grow in a cold environment38, has not been considered but could be able climate regions would increase crop production. Further stud- a lifesaving source of emergency food if such production systems ies on adaptation and the impacts on short-term food availability were operational. are needed, but those topics are beyond the scope of this study. In conclusion, the reduced light, global cooling and likely trade Adaptation in fisheries is also not considered, such as changes in restrictions after nuclear wars would be a global catastrophe for the use of discarded bycatch and offal in fisheries. These include food security. The negative impact of climate perturbations on the reduced availability of fuel, fertilizer and infrastructure for food total crop production can generally not be offset by livestock and production after a war, the effects of elevated ultraviolet radiation36 aquatic food (Fig. 5a). More than 2 billion people could die from on food production and radioactive contamination37. While this a nuclear war between India and Pakistan, and more than 5 billion analysis focuses on calories, humans would also need proteins and could die from a war between the United States and Russia (Table 1). Nature Food | VOL 3 | August 2022 | 586–596 | www.nature.com/natfood 591 Articles Nature Food a Global averaged calorie intake in Year 2 after nuclear war b Global population (%) with 1,911 kcal per capita per day 100 0 94.8% (2010, no trade) Calorie intake (kcal per capita per day) Population with food support (%) 3,000 1 Population in danger (billion) 80 2,310 Calorie intake 2 in 2010 2,000 60 1,911 3 Losing weight 1,145 4 1,000 40 Less than resting energy 5 expenditure 20 0 5 Tg 16 Tg 27 Tg 37 Tg 47 Tg 150 Tg 5 Tg 16 Tg 27 Tg 37 Tg 47 Tg 150 Tg Nuclear war scenarios Calories from crops, vegetables, fruit and marine fishery Nuclear war scenarios (fishing effort is business as usual) Livestock case Livestock case Portion of calories from livestock and aquaculture Livestock production reduced from reduction of grass and crop production Calories from crops, vegetables, fruit and marine fishery (fishing effort is five times business as usual) Partial Livestock case Partial Livestock case Calories from livestock feed added to human consumption (remaining livestock grain feed is used for livestock) Ticks from bottom to top are 0%, 20%, 40%, 60%, 80% and 100% The remaining livestock crop feed is for raising livestock 100% of livestock grain feed to human food Food waste Calories from food waste; ticks are 50% and 100% 80% of livestock grain feed to human food 60% of livestock grain feed to human food 40% of livestock grain feed to human food 20% of livestock grain feed to human food 0% of livestock grain feed to human food Food waste 100% of livestock grain feed to human food 50% less food waste 100% of livestock grain feed to human food no food waste Fig. 5 | Overview of global calorie intake and sensitivity to livestock and food waste assumptions. a, Global average change in calorie intake per person per day in Year 2 post-war under the Livestock case (yellow bars) and for the Partial Livestock case (red bars), assuming that all food and waste is evenly distributed. For the Partial Livestock case, additional calories potentially available by human consumption of animal feed, mainly maize and soybeans, are plotted for various portions of converted animal feed (pink tick marks), and the remaining livestock crop feed is used for raising livestock. Critical food intake levels are marked in the right margin. b, Without international trade, the global population (%) that could be supported, although underweight, by domestic food production at the end of Year 2 after a nuclear war if they receive the calories supporting their regular physical activity29 and the rest of the population would receive no food, under the Livestock and Partial Livestock cases. The blue line in b shows the percentage of population that can be supported by current food production when food production does not change but international trade is stopped. National data are calculated first (Supplementary Tables 4 and 5 and Supplementary Fig. 5) and then aggregated to global data. The results here provide further support to the 1985 statement by reductions range from 1.5 °C (5 Tg soot injection) to 14.8 °C (150 Tg soot US President Ronald Reagan and Soviet General Secretary Mikhail injection), peaking within 1–2 years after the war, with temperature reduction lasting for more than 10 years. The cooling also reduces precipitation over Gorbachev and restated by US President Joe Biden and Russian summer monsoon regions. Similar but smaller reductions of solar radiation and President Vladimir Putin in 2021 that ‘a nuclear war cannot be won temperature are projected in marine regions (Fig. 1b,d), with resulting changes in and must never be fought’. lower trophic-level marine primary productivity. We applied local changes at every grid cell to the crop and fish models. Methods Climate model. All nuclear war scenarios9,18 are simulated using the Community We use a state-of-the-art global climate model to calculate the climatic and Earth System Model (CESM)39. This model includes interactive atmosphere, biogeochemical changes caused by a range of stratospheric soot injections, each land, ocean and sea ice. Both atmosphere and land have a horizontal resolution associated with a nuclear war scenario18 (Tables 1 and 2). Simulated changes in of 1.9° × 2.5°, and the ocean has a horizontal resolution of 1°. The atmospheric surface air temperature, precipitation and downward direct and diffuse solar model is the Whole Atmosphere Community Climate Model version 4 (ref. 40). The radiation are used to force a state-of-the-art crop model to estimate how the land model is the Community Land Model version 4 with the carbon–nitrogen productivity of the major crops (maize, rice, spring wheat and soybean) would be cycle. CESM output at 1- and 3-hour resolution, including 2 m air temperature, affected globally, and changes in oceanic net primary production and sea surface precipitation, specific humidity and downward longwave radiation and solar temperature are used to force a global marine fisheries model. We combine these radiation (separated into direct and diffuse radiation), is used to drive the offline results with assumptions about how other crop production, livestock production, crop model simulations. There are three ensemble members of the control fish production and food trade could change and calculate the amount of food that simulation, which repeats the climate forcing of 2000 for 15 years, three ensemble would be available for each country in the world after a nuclear war. members of the 5 Tg case and one simulation for each other nuclear war scenario. The simulated surface climate disruptions due to the nuclear war scenarios are In all the simulations, the soot is arbitrarily injected during the week starting summarized in Fig. 1. Averaged over the current crop regions, surface downwelling on May 15 of Year 1. Our scenarios assume that all stored food is consumed in solar radiation reduces by 10 W m−2 (5 Tg soot injection) to 130 W m−2 (150 Tg soot Year 1 and we present analysis of the remaining food in Year 2. If the war occurred injection). With less energy received, the maximum average 2 m air temperature at the end of a calendar year, there would still be food available in Year 2, so what 592 Nature Food | VOL 3 | August 2022 | 586–596 | www.nature.com/natfood Nature Food Articles Table 2 | Changes in food calorie availability (%) in Year 2 after a nuclear war for the nations with nuclear weapons and global average assuming no trade after simulated nuclear wars under the Livestock case, the Partial Livestock case and the No Livestock case with 50% livestock feed to human consumption Nations 5g 16 Tg 27 Tg 37 Tg 47 Tg 150 Tg 150 Tg + hw 150 Tg + tw Livestock case China −14.1 −47.4 −64.0 −77.7 −83.5 −99.5 −99.5 −99.4 France 1.6 −13.0 −17.2 −26.0 −29.4 −98.3 −97.9 −97.6 India −4.6 −4.7 −2.0 −9.4 −21.6 −66.8 −64.1 −61.4 Israel 25.4 −5.0 −1.9 −4.5 3.8 −63.3 −55.2 −46.9 North Korea −18.1 −88.7 −90.7 −98.3 −99.2 −99.9 −99.9 −99.9 Pakistan −0.9 15.9 18.5 −13.8 −19.9 −85.3 −84.1 −82.8 Russia 1.3 −62.3 −83.1 −89.6 −90.9 −99.0 −98.9 −98.7 United Kingdom 1.2 −29.7 −40.5 −56.9 −72.2 −89.0 −87.1 −85.2 United States −11.2 −45.8 −62.0 −75.7 −86.3 −98.8 −98.5 −98.3 Global −8.2 −23.8 −33.4 −42.1 −49.2 −81.3 −78.8 −76.5 Partial Livestock case (50% livestock feed to human consumption, 50% livestock feed to livestock) China −2.1 −40.6 −59.8 −75.2 −81.5 −99.6 −99.5 −99.5 France 58.8 33.2 28.2 16.3 10.8 −99.0 −98.8 −98.6 India −1.4 −0.2 2.7 −4.7 −17.8 −65.0 −62.1 −59.2 Israel 16.7 0.3 −1.1 −3.8 −8.1 −75.5 −70.0 −64.5 North Korea −13.3 −88.7 −90.9 −98.8 −99.5 −99.9 −99.9 −99.9 Pakistan −1.4 20.8 21.6 −11.9 −16.0 −83.6 −82.2 −80.8 Russia 25.3 −56.1 −80.6 −88.0 −88.7 −99.0 −98.9 −98.7 United Kingdom 30.7 −12.5 −27.0 −50.7 −72.1 −94.4 −93.4 −92.5 United States 70.8 −0.5 −32.7 −60.4 −78.9 −99.1 −98.9 −98.7 Global 13.5 −5.7 −17.4 −28.1 −36.7 −77.1 −74.0 −71.2 No Livestock case (50% livestock feed to human consumption) China −11.4 −46.6 −64.0 −77.9 −83.5 −99.7 −99.6 −99.5 France 45.2 20.5 16.6 6.6 1.4 −99.8 −99.7 −99.7 India −5.4 −3.1 −0.2 −7.0 −20.0 −65.8 −63.0 −60.1 Israel −24.7 −26.7 −30.6 −32.4 −44.3 −91.5 −89.5 −87.7 North Korea −16.0 −89.6 −91.8 −99.4 −99.8 −99.9 −99.9 −99.9 Pakistan −12.1 12.7 11.8 −19.4 −21.2 −83.7 −82.4 −81.0 Russia 16.4 −60.7 −82.6 −89.0 −89.2 −99.0 −98.9 −98.8 United Kingdom 15.9 −24.2 −37.3 −59.7 −79.7 −99.8 −99.8 −99.8 United States 60.3 −8.1 −38.6 −65.1 −81.8 −99.5 −99.4 −99.3 Global 5.9 −12.1 −22.9 −32.9 −40.8 −79.0 −76.3 −73.7 The total calorie reduction is referenced to the observed food calorie availability in 2010. China here includes Mainland China, Hong Kong and Macao. 150 Tg + hw is half of the household waste added to food consumption, and 150 Tg + tw is total household waste added to food consumption. Bold is used for headings and global averages. we label Year 2 should be relabelled Year 3. However, since the severe climate and rice, soybeans, spring wheat, sugar cane and cotton, and also simulates natural food impacts last for more than 5 years (Figs. 1 and 2), the same conclusions apply vegetation, such as grasses. In this study, we used the output of the cereals (maize, to a world after a nuclear war. rice, soybeans and spring wheat) and grasses. Although CLM5crop does not simulate winter wheat, we assume winter wheat production is changed by the Direct climate model output use. Because climate models have biases, it is typical same amount as spring wheat, which has been found in other studies14; however, to bias correct model output before using it as input for crop models. There are this may underestimate the winter wheat response, because winter wheat would various techniques that attempt to use past observational data to address changes experience colder temperatures during its growing period that would be more in the mean and variance, but none are perfect and all are limited by assumptions likely to cross critical thresholds14. Surface ozone and downward ultraviolet that future relations between model output and crop model input can be based on radiation would also be impacted by nuclear war36, but CLM5crop is not able to the recent past. A common method14 is the delta method in which an observational consider those impacts, which might exacerbate the losses. In addition, the crop reanalysis weather dataset is used and monthly means of temperature, precipitation model does not consider the availability of pollinators, killing frost and alternative and insolation are modified according to the climate model simulations. This seeds. The model simulates rainfed crops and irrigated crops separately, and all comes with the advantage of realistic internal variability important for crop results presented here refer to the total production of rainfed and irrigated crops. modelling12–14 but does not adjust changes in variance, which might be an Irrigated crops are simulated under the assumption that freshwater availability is unrealistic assumption under higher emission scenarios, such as the 150 Tg not limiting43. Although evaporation is reduced with cooling, it is possible that case. Here, because we are using a crop model that has already been calibrated our result may underestimate the negative impact from precipitation reduction, with the same climate model that we are using, we use raw climate model output especially for the large injection cases. (1.9° × 2.5°) to force the crop model, and this allows variance to change too. CLM5crop was evaluated41 using FAO observations (average of 1991–2010), and it does a reasonable job of reproducing observed spatial pattern of maize, rice, soybean Crop model. Crop simulation uses the Community Land Model version 5 crop and spring wheat yield. Also, time series of crop yields simulated by CLM5crop (CLM5crop)41–43 in the Community Earth System Model version 2 (CESM2). compare with FAO data from 2006 to 2018, and CLM5crop reasonably represents Dynamic vegetation is not turned on. CLM5crop has six active crops, maize, global total production and average yields of maize, rice, soybean and spring wheat42. Nature Food | VOL 3 | August 2022 | 586–596 | www.nature.com/natfood 593 Articles Nature Food CLM5crop is spun up for 1,060 years by repeating the past 10 years of the products are soybean and byproducts soybean oil and soybean cake, rice products CESM control to reach the equilibrium of four soil carbon pools. The crop are rice and byproduct rice bran oil and wheat product is wheat. Products for food simulations are at the same resolution as CESM simulations (1.9° × 2.5°). The crop purposes are the sum of food supply in each category and the processing product planting date is determined by growing degree days, and the location of cropland is minus the total byproducts (the difference includes processing for the purpose of fixed for all crops. alcohol or sugar). Fishery model. Fish and fisheries responses are simulated with the BiOeconomic Calorie calculations. For the Livestock case, national-level available calories are mArine Trophic Size-spectrum (BOATS) model15,44,45. BOATS was used to calculate calculated by the size-structured biomass of commercially targeted fish based on gridded (1° CL = Cplantbased × 1 − Rcy + Clivestock−ruminant × 1 − Rgrass ( ) ( ) horizontal resolution) inputs of sea surface temperature and oceanic net primary production from CESM. The model also interactively simulates fishing effort and +Clivestock−monogastric × 1 − Rcy + Clivestock−monogastric ( ) fish catch through a bioeconomic component that depends on fish price, cost of fishing, catchability and fisheries regulation15. Details are found in ref. 15 and F ×Rgrass × 1 − Rcy × F ruminant−cropfeed ( ) references therein. monogastric−cropfeed (4) +Caquaculture × 1 − Rcy + Cmarine−catch × 1 − Rmarine−catch−y ( ) ( ) Combining crop and marine fish data. Supplementary Table 1 shows the total calorie reductions for each of the nine nuclear states from just the simulated crops + 1 − Rcy × Cplantbased × ( ) ffinal−product−biofuel ffood and marine fish. Data for countries can be found in Supplementary Table 2. To calculate nation-level calories available from simulated crops and fish, we where CL is calories available in each nation L (kcal per capita per day) under the weight the production by the calorie content of each type of food. We use data Livestock case, Cplantbased, Clivestock-ruminant and Clivestock-monogastric are calories available from FAO23,24,46,47. Nation-level calorie reduction (%) from total production of from plant-based products, ruminants and monogastrics27 and Caquaculture and maize, rice, soybean, wheat and marine fish is thus calculated as: Cmarine-catch are calculated by calorie availability from fish27 multiplied by the ratio Pi ci Riy of aquaculture and catch46. Rgrass is grass production change, and Rmarine-catch-y is wiy = ∑5 (1) marine capture change. Fruminant-cropfeed is the fraction of crop feed for ruminant, and i=1 Pi ci Riy Fmonogastric-cropfeed is the fraction of crop feed for monogastrics26. Rcy is crop production and change calculated as: 5 Pi ci Riy wiy = ∑4 , (5) Ry = Riy wiy , (2) ∑ i=1 Pi ciRiy i=1 and where index i is maize, rice, soybean, wheat or marine fish wild catch, wiy is the calorie weight of each commodity per country each year, Pi is the national 4 Rcy = Riy wiy (6) ∑ production of item i in FAO-Food Balance Sheet (FBS)23,24, ci is calories per 100 g dry mass for each item23, Riy is national production reduction (%) of each item in i=1 year y after the nuclear wars and Ry is nation-averaged calorie reduction (%) of the five items in year y after the nuclear wars. where index i is maize, rice, soybean or wheat, wiy is the calorie weight of each commodity per country each year, Pi is the national production of item Effects on other food types. Other crops. National averaged calorie reduction (%) i in FAO-CBS47, ci is calories per 100 g retail weight for each item23 and Riy is of the four simulated crops is applied to the total calories of all crops in 2010 to national production change (%) of each item in year y after the nuclear wars. estimate simulated nuclear war impacts on this category. ffinal−product−biofuel is the fraction of final product of biofuel in plant-based product, and ffood is the fraction of food in plant-based product. For the No Livestock case, national-level available calories are calculated by Livestock and aquaculture. We assume these two types of food share a similar response to simulated nuclear war as they involve feeding animals in a relatively CNL = Cplantbased × 1 − Rcy + Cmarine−catch × 1 − Rmarine−catch−y ( ) ( ) controlled environment. For global calculations for livestock, we assume that +Cplantbased × ffeed−to−food × 1 − Rcy × pfeed−for−human ( ) 46% are fed by pasture and 54% are fed by crops and processed products48 and (7) use national-level data26 to calculate reduction of livestock feed from pasture and f + 1 − Rcy × Cplantbased × final−product −biofuel ( ) crop-based products. We assume that livestock production is linearly correlated ffood with the feed. Annual leaf carbon of grasses (both C3 and C4) is used to estimate pasture changes, and reduction of the four simulated crops is used for crop feed CNL is national-level available calories in the No Livestock case. ffeed-to-food is the changes. For aquaculture, the feed is only from crops and processed products, fraction of food crops that are used as feed relative to their usage as food, calculated and the production is also correlated with the amount of feed fish receive. Direct based on their calorie content26. pfeed-for-human is the percentage of livestock grain feed climate change impacts on livestock and fish are not considered. used for human consumption. We tested 0%, 20%, 40%, 50%, 60%, 80% and 100% Inland fish capture is not considered in this study. Because inland fish and used 50% for Table 2 and Fig. 4. contribute only 7% of total fish production46, adding inland fisheries would not For the Partial Livestock case, national-level available calories are calculated by change the main conclusions of this study. CPL = Cplantbased × 1 − Rcy + Cmarine−catch × 1 − Rmarine−catch−y ( ) ( ) International trade. All food commodity trade calculations are based on the 2010 +Cplantbased × ffeed−to−food × 1 − Rcy × pfeed−for−human ( ) FAO Commodity Balance Sheet (FAO-CBS), FAO-FBS and processed data from +(1 − pfeed−for−human ) × Clivestock−ruminant × 1 − Rgrass ( ( ) previous studies24,27,28,47. This dataset provides the production and usage of each food and non-food agricultural product for each country and imports and exports (8) +Clivestock−monogastric × 1 − Rcy ( ) and thus allows the calculation on a national basis of food usage and calorie availability. F ) +Clivestock−monogastric × Rgrass × 1 − Rcy × F ruminant−cropfeed ( ) Domestic availability of a food in each country comes from domestic monogastric−cropfeed production and reserves, reduced by exports and increased by imports. We ffinal−product−biofuel + 1 − Rcy × Cplantbased × ( ) calculate no international trade by applying the ratio of domestic production and ffood domestic supply to each food category and the food production in different usages: CPL is national-level available calorie in Partial Livestock case. On the basis of Pdp the assumed percentage of livestock crop feed to convert to human consumption, Cfood−notrade = Cfood × (3) Pds instead of wasting the remaining portion of livestock crop feed as in No Livestock case, here we use the remaining livestock crop feed to raise livestock. where Cfood is national-level calorie supply from different food types26,46, Cfood-notrade The percentage of national household waste is calculated by is national-level calorie supply from different food types with the assumption of no international trade, Pdp is national-level domestic production for each type of Cavailable − Cintake Pwaste = 100% × (9) food in FAO-CBS and Pds is national-level domestic supply for each type of food in Cavailable FAO-CBS. Domestic supply is the available food on the market, including domestic production, export and import. Pwaste is the percentage of national household waste of food calorie availability in Food usage of maize, soybean, rice and wheat is calculated from FAO-CBS. 2010, Cavailable is the food calorie availability per day per person in each country and In FAO-CBS, maize products are maize and byproduct maize germ oil, soybean Cintake is the national calorie intake per day per person27. 594 Nature Food | VOL 3 | August 2022 | 586–596 | www.nature.com/natfood Nature Food Articles Calorie requirements. The population percentage supported by available calories 5. Aleksandrov, V. V. & Stenchikov, G. L. On the modeling of the climatic calculated for the Livestock, Partial Livestock and No Livestock responses indicates consequences of the nuclear war. Proc. Applied Math (Computing Centre of the macro-level consequences for food security (Fig. 4). The current average the USSR Academy of Sciences, 1983). human available calorie supply is 2,855 kcal per capita per day, including food 6. Robock, A. Snow and ice feedbacks prolong effects of nuclear winter. Nature intake and food waste (Fig. 3). Calorie requirements vary significantly with age, 310, 667–670 (1984). gender, size, climate, level of activity and underlying medical conditions. Ref. 27 7. Robock, A. et al. Climatic consequences of regional nuclear conflicts. estimated the national-level calorie availability, calorie intake, calorie from Atm. Chem. Phys. 7, 2003–2012 (2007). plant-based product, livestock and fish and also calculated the calorie intake 8. Robock, A., Oman, L. & Stenchikov, G. L. Nuclear winter revisited with a of an underweight population with current physical activity of an underweight modern climate model and current nuclear arsenals: still catastrophic population with sedentary physical activity and calorie intake lower than the basal consequences. J. Geophys. Res. 112, D13107 (2007). metabolic rate. We assume that the calorie intake of an underweight population 9. Coupe, J., Bardeen, C. G., Robock, A. & Toon, O. B. Nuclear winter responses with current physical activity is needed to support life and regular labour activity. to global nuclear war in the Whole Atmosphere Community Climate Model Version 4 and the Goddard Institute for Space Studies ModelE. J. Geophys. Uncertainties. This work was done with one Earth system model, with only one Res. Atmos. 124, 8522–8543 (2019). ensemble member for all the cases with soot injections >5 Tg, only one crop model 10. Harwell, M. A. & Cropper Jr., W. P. in Environmental Consequences of Nuclear and only one fishery model. For the 5 Tg case and the control, there are three War, SCOPE 28, Volume II Ecological and Agricultural Effects, 2nd Edn (eds ensemble members, but only the ensemble averages are used. The three ensemble Harwell, M. A. & Hutchinson, T. C.) Ch. 4 (Wiley, 1989). members for the 5 Tg case are very similar (Supplementary Fig. 8), so climate 11. Cropper Jr., W. P. & Harwell, M. A. in Environmental Consequences of Nuclear variability for the larger forcings would be much smaller than the signal. War, SCOPE 28, Volume II Ecological and Agricultural Effects, 2nd Edn (eds CESM is a state-of-the-art climate model, and its simulations of the impacts of Harwell, M. A. & Hutchinson, T. C.) Ch. 5 (Wiley, 1989). nuclear war have been almost identical to simulations with other models for the 12. Xia, L. & Robock, A. Impacts of a nuclear war in South Asia on rice 5 Tg (refs. 49,50) and 150 Tg (ref. 9) cases. However, further developments in climate production in mainland China. Climatic Change 116, 357–372 (2013). models, such as including organic carbon in fire emissions, and better simulating 13. Özdoğan, M., Robock, A. & Kucharik, C. Impacts of a nuclear war in South aerosol growth and interactions with the surrounding environment, may improve Asia on soybean and maize production in the Midwest United States. climate prediction after a nuclear war. Climatic Change 116, 373–387 (2013). CLM5crop and BOATS are also state-of-the-art models, but future simulations 14. Jägermeyr, J. et al. A regional nuclear conflict would compromise global food with different models would certainly be useful. CLM5crop compares well with security. Proc. Natl Acad. Sci. USA 117, 7071–7081 (2020). other crop models in response to nuclear war forcing14 (Supplementary Fig. 2). If 15. Scherrer, K. J. N. et al. Marine wild-capture fisheries after nuclear war. Proc. anything, CLM5crop underestimates the crop response to nuclear war (Fig. 2 and Natl Acad. Sci. USA 117, 29748–29758 (2020). Supplementary Fig. 2). Because most crop models were developed for the current 16. Toon, O. B. et al. Atmospheric effects and societal consequences of regional or warmer climates, further research is needed to understand how crops react to a scale nuclear conflicts and acts of individual nuclear terrorism. Atm. Chem. suddenly cold environment. Our study is the first step to reveal national food security Phys. 7, 1973–2002 (2007). after nuclear wars, but crops may not respond uniformly to the same forcing in 17. Robock, A. & Zambri, B. Did smoke from city fires in World War II cause each nation, given different farming practices. In addition, multi-model assessment global cooling. J. Geophys. Res. Atmos. 123, 10314–10325 (2018). will be essential to fully investigate this problem, and crop model developments 18. Toon, O. B. et al. Rapid expansion of nuclear arsenals by Pakistan and India are important to understand impacts from surface ozone, ultraviolet radiation and portends regional and global catastrophe. Sci. Adv. 5, eaay5478 (2019). freshwater availability. Furthermore, local radioactive contamination and climate 19. Yu, P. et al. Black carbon lofts wildfire smoke high into the stratosphere to change from nuclear war would impact the insect community. The influence on form a persistent plume. Science 365, 587–590 (2019). pests, pollinators and other insects is unclear, and hence further studies are needed. 20. Yu, P. et al. Persistent stratospheric warming due to 2019–2020 Australian Some assumptions in this study could be examined in future work. For wildfire smoke. Geophys. Res. Lett. 48, e2021GL092609 (2021). example, to turn off international trade, the ratio of local production to domestic 21. Peterson, D. A. et al. Australia’s Black Summer pyrocumulonimbus super supply is applied on a national level. Also, to calculate national calorie intake after outbreak reveals potential for increasingly extreme stratospheric smoke nuclear wars, we assume that food is evenly distributed in each country. Economic events. npj Clim. Atmos. Sci. 4, 38 (2021). models will be necessary to further understand the contributions of trade and local 22. Ambio Advisory Group, Reference scenarios: How a nuclear war might be food distribution systems to human calorie intake after nuclear wars. fought, in Nuclear War: The Aftermath (ed Peterson, J.) Ch. 3 (Pergamon This study uses calorie intake from ref. 27, and food loss from harvesting is not Press, 1983). considered. If human behaviour and the food industry would change substantially, 23. Food Composition Tables (FAO, 2022); http://www.fao.org/3/X9892E/ this would affect our conclusions. X9892e05.htm 24. Food Balances (FAO, 2022); http://www.fao.org/faostat/en/#data/FBSH Reporting summary. Further information on research design is available in the 25. Falkendal, T. et al. Grain export restrictions during COVID-19 risk Nature Research Reporting Summary linked to this article. food insecurity in many low- and middle-income countries. Nat. Food 2, 11–14 (2021). 26. Weindl, I. et al. Livestock and human use of land: productivity trends and Data availability dietary choices as drivers of future land and carbon dynamics. Glob. Environ. Data of crop yield, grass production, national livestock feed, national calorie and Change 47, 121–132 (2017). national plant product usage are available at https://osf.io/YRBSE/. Additional data 27. Bodirsky, B. et al. mrcommons: MadRat Commons Input Data Library. R that support the findings of this study are available from the corresponding author version 1.9.3 https://github.com/pik-piam/mrcommons (2022). upon request. 28. Bodirsky, B. L. et al. The ongoing nutrition transition thwarts long-term targets for food security, public health and environmental protection. Sci. Rep. Code availability 10, 19778 (2020). The source code for the CESM(WACCM) model used in this study is freely 29. National Research Council, Energy, in Recommended Dietary Allowances, available at https://www.cesm.ucar.edu/working_groups/Whole-Atmosphere/ 10th Edn Ch. 3 (US National Academies Press, 1989). code-release.html, and the code for CLM5 is available at https://www.cesm.ucar. 30. FAO, IFAD, UNICEF, WFP and WHO The State of Food Security and edu/models/cesm2/land/. Nutrition in the World 2021 (FAO, 2021). 31. Alvarado, K. A., Mill, A., Pearce, J. M., Vocaet, A. & Denkenberger, D. Received: 24 August 2021; Accepted: 14 July 2022; Scaling of greenhouse crop production in low sunlight scenarios. Sci. Total Environ. https://doi.org/10.1016/j.scitotenv.2019.136012 (2019). Published online: 15 August 2022 32. Denkenberger, D., Pearce, J. M. Feeding Everyone No Matter What: Managing Food Security After Global Catastrophe (Academic Press, 2014). References 33. Martínez, J. B. G. et al. Potential of microbial protein from hydrogen for 1. Oman, L., Robock, A., Stenchikov, G. L. & Thordarson, T. High-latitude preventing mass starvation in catastrophic scenarios. Sustain Prod. Consum. eruptions cast shadow over the African monsoon and the flow of the Nile. 25, 234–247 (2021). Geophys. Res. Lett. 33, L18711 (2006). 34. Throup, J. et al. Rapid repurposing of pulp and paper mills, biorefineries, and 2. Wood, G. D. Tambora: The Eruption That Changed The World (Princeton breweries for lignocellulosic sugar production in global food catastrophes. Univ. Press, 2014). Food Bioprod. Process. 131, 22–39 (2022). 3. Stommel, H. & Stommel, E. Volcano Weather; The Story of 1816, The Year 35. Hochman, G. et al. Economic incentives modify agricultural impacts of Without a Summer (Seven Seas Press, 1983). nuclear war. Env. Res. Lett. 17, 054003 (2022). 4. Turco, R. P., Toon, O. B., Ackerman, T. P., Pollack, J. B. & Sagan, C. Nuclear 36. Bardeen, C. G. et al. Extreme ozone loss following nuclear war resulting in winter: global consequences of multiple nuclear explosions. Science 222, enhanced surface ultraviolet radiation. J. Geophys. Res. Atmos. 126, 1283–1292 (1983). e2021JD035079 (2021). Nature Food | VOL 3 | August 2022 | 586–596 | www.nature.com/natfood 595 Articles Nature Food 37. Grover, H. D. & Harwell, M. A. Biological effects of nuclear war II: impact on and the Open Philanthropy Project. B.L.B. has received funding from the European the biosphere. Bioscience 35, 576–583 (1985). Union’s Horizon 2020 research and innovation programme under grant agreement 38. Denkenberger, D. C. & Pearce, J. M. Feeding everyone: solving the food crisis number 776479 (COACCH) and 821010 (CASCADES). I.W. and B.L.B. have received in event of global catastrophes that kill crops or obscure the sun. Futures 72, funding from the German Federal Ministry of Education and Research (BMBF) in the 57–68 (2015). context of the project ‘FOCUS–Food security and sustained coastal livelihoods through 39. Hurrell, J. W. et al. The Community Earth System Model: a framework for linking land and ocean’ (031B0787B). R.H. was supported by the European Research collaborative research. Bull. Am. Meteorol. Soc. 94, 1339–1360 (2013). Council under the European Union’s Horizon 2020 Research and Innovation Programme 40. Marsh, D. R. et al. Climate change from 1850 to 2005 simulated in under grant agreement 682602. We thank I. Helfand for valuable suggestions on the work CESM1(WACCM). J. Clim. 26, 7372–7390 (2013). and D. Lombardozzi for supporting CLM5crop simulations. 41. Lombardozzi, D. L. et al. Simulating agriculture in the Community Land Model Version 5. J. Geophys. Res. Biogeosci. 125, e2019JG005529 (2020). Author contributions 42. Fan, Y. et al. Solar geoengineering can alleviate climate change pressure on L.X., A.R., K.S. and C.S.H. designed the study. C.G.B. conducted the climate model crop yields. Nat. Food 2, 373–381 (2021). simulations, L.X. conducted the crop simulations and K.S. and R.H. conducted the 43. Lawrence, D. M. et al. The Community Land Model version 5: description of fishery simulations. J.J. provided GGCMI crop model simulations. B.L.B. and I.W. new features, benchmarking, and impact of forcing uncertainty. J. Adv. provided national livestock feed and national calorie intake data. L.X. analysed the Modeling Earth Systems 11, 4245–4287 (2019). data with contributions from all the authors. A.R. and L.X. wrote the first draft, and all 44. Carozza, D. A., Bianchi, D. & Galbraith, E. D. Formulation, general features authors contributed to editing and revising the manuscript. and global calibration of a bioenergetically-constrained fishery model. PLoS ONE 12, e0169763 (2017). 45. Carozza, D. A., Bianchi, D. & Galbraith, E. D. The ecological module of Competing interests BOATS-1.0: a bioenergetically constrained model of marine upper trophic The authors declare no competing interests. levels suitable for studies of fisheries and ocean biogeochemistry. Geosci. Model Dev. 9, 1545–1565 (2016). Additional information 46. The State of World Fisheries and Aquaculture 2020. Sustainability in Action Supplementary information The online version contains supplementary material (FAO, 2020); https://doi.org/10.4060/ca9229en available at https://doi.org/10.1038/s43016-022-00573-0. 47. Commodity Balances (FAO, 2022); https://www.fao.org/faostat/en/#data/CB Correspondence and requests for materials should be addressed to Lili Xia. 48. Mottet, A. et al. Livestock: on our plates or eating at our table? A new analysis of the feed/food debate. Glob. Food Sec.14, 1–8 (2017). Peer review information Nature Food thanks Deepak Ray, Ertharin Cousin, 49. Mills, M. J., Toon, O. B., Lee-Taylor, J. & Robock, A. Multi-decadal global Michal Smetana and the other, anonymous, reviewer(s) for their contribution to cooling and unprecedented ozone loss following a regional nuclear conflict. the peer review of this work. Earth’s Future 2, 161–176 (2014). Reprints and permissions information is available at www.nature.com/reprints. 50. Stenke, A. et al. Climate and chemistry effects of a regional scale nuclear Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in conflict. Atmos. Chem. Phys. 13, 9713–9729 (2013). published maps and institutional affiliations. 51. Toon, O. B., Robock, A. & Turco, R. P. Environmental consequences of nuclear war. Phys. Today 61, 37–42 (2008). Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adap- tation, distribution and reproduction in any medium or format, as long Acknowledgements as you give appropriate credit to the original author(s) and the source, provide a link to This study was supported by the Open Philanthropy Project, with partial support from the Creative Commons license, and indicate if changes were made. The images or other the European Research Council under the European Union’s Horizon 2020 Research and third party material in this article are included in the article’s Creative Commons license, Innovation Programme under grant agreement 682602. A.R. and L.X. were supported unless indicated otherwise in a credit line to the material. If material is not included in by National Science Foundation grants AGS-2017113 and ENG-2028541. K.S. was the article’s Creative Commons license and your intended use is not permitted by statu- supported by the European Research Council under the European Union’s Horizon tory regulation or exceeds the permitted use, you will need to obtain permission directly 2020 Research and Innovation Programme under grant agreement 682602 and Research from the copyright holder. To view a copy of this license, visit http://creativecommons. Council of Norway project 326896. C.S.H., C.G.B. and O.B.T. were supported by the org/licenses/by/4.0/. Open Philanthropy Project. J.J. was supported by the NASA GISS Climate Impacts Group © The Author(s) 2022 596 Nature Food | VOL 3 | August 2022 | 586–596 | www.nature.com/natfood nature portfolio | reporting summary Corresponding author(s): Lili Xia Last updated by author(s): Jul 6, 2022 Reporting Summary Nature Portfolio wishes to improve the reproducibility of the work that we publish. This form provides structure for consistency and transparency in reporting. For further information on Nature Portfolio policies, see our Editorial Policies and the Editorial Policy Checklist. Statistics For all statistical analyses, confirm that the following items are present in the figure legend, table legend, main text, or Methods section. n/a Confirmed The exact sample size (n) for each experimental group/condition, given as a discrete number and unit of measurement A statement on whether measurements were taken from distinct samples or whether the same sample was measured repeatedly The statistical test(s) used AND whether they are one- or two-sided Only common tests should be described solely by name; describe more complex techniques in the Methods section. A description of all covariates tested A description of any assumptions or corrections, such as tests of normality and adjustment for multiple comparisons A full description of the statistical parameters including central tendency (e.g. means) or other basic estimates (e.g. regression coefficient) AND variation (e.g. standard deviation) or associated estimates of uncertainty (e.g. confidence intervals) For null hypothesis testing, the test statistic (e.g. F, t, r) with confidence intervals, effect sizes, degrees of freedom and P value noted Give P values as exact values whenever suitable. For Bayesian analysis, information on the choice of priors and Markov chain Monte Carlo settings For hierarchical and complex designs, identification of the appropriate level for tests and full reporting of outcomes Estimates of effect sizes (e.g. Cohen's d, Pearson's r), indicating how they were calculated Our web collection on statistics for biologists contains articles on many of the points above. Software and code Policy information about availability of computer code Data collection Data were computer output. Data analysis The source code for the CESM(WACCM) model used in this study is freely available at https://www.cesm.ucar.edu/working_groups/Whole- Atmosphere/code-release.html, and the code for CLM5 is available at https://www.cesm.ucar.edu/models/cesm2/land/. For manuscripts utilizing custom algorithms or software that are central to the research but not yet described in published literature, software must be made available to editors and reviewers. We strongly encourage code deposition in a community repository (e.g. GitHub). See the Nature Portfolio guidelines for submitting code & software for further information. Data Policy information about availability of data March 2021 All manuscripts must include a data availability statement. This statement should provide the following information, where applicable: - Accession codes, unique identifiers, or web links for publicly available datasets - A description of any restrictions on data availability - For clinical datasets or third party data, please ensure that the statement adheres to our policy Data of crop yield, grass production, national livestock feed, national calorie, and national plant product usage are available at https://osf.io/YRBSE/. Additional data that support the findings of this study are available from the corresponding author upon request. 1 nature portfolio | reporting summary Human research participants Policy information about studies involving human research participants and Sex and Gender in Research. Reporting on sex and gender N/A Population characteristics N/A Recruitment N/A Ethics oversight N/A Note that full information on the approval of the study protocol must also be provided in the manuscript. Field-specific reporting Please select the one below that is the best fit for your research. If you are not sure, read the appropriate sections before making your selection. Life sciences Behavioural & social sciences Ecological, evolutionary & environmental sciences For a reference copy of the document with all sections, see nature.com/documents/nr-reporting-summary-flat.pdf Ecological, evolutionary & environmental sciences study design All studies must disclose on these points even when the disclosure is negative. Study description We used computer models to simulate the impacts of smoke from fires generated by nuclear war on climate and crops. We evaluated the results based on several potential adaptation responses. Research sample We used one climate model, forced by six different scenarios, and one crop model. Sampling strategy We used all the data we had. Data collection N/A Timing and spatial scale N/A Data exclusions N/A Reproducibility N/A Randomization N/A Blinding N/A Did the study involve field work? Yes No Reporting for specific materials, systems and methods We require information from authors about some types of materials, experimental systems and methods used in many studies. Here, indicate whether each material, system or method listed is relevant to your study. If you are not sure if a list item applies to your research, read the appropriate section before selecting a response. March 2021 2 Materials & experimental systems Methods nature portfolio | reporting summary n/a Involved in the study n/a Involved in the study Antibodies ChIP-seq Eukaryotic cell lines Flow cytometry Palaeontology and archaeology MRI-based neuroimaging Animals and other organisms Clinical data Dual use research of concern March 2021 3