Global Famine after Nuclear War


 In a nuclear war, bombs targeted on cities and industrial areas would start firestorms, injecting large amounts of soot into the upper atmosphere, which would spread globally and rapidly cool the planet. The soot loadings would cause decadal disruptions in Earth’s climate, which would impact food production systems on land and in the oceans. In 1980s, investigations of nuclear winter impacts on global agricultural production and food availability for 15 nations, but new information now allows us to update those estimates. Recently, several studies analyzed changes of major grain crops and marine wild-catch fisheries for different scenarios of regional nuclear war using sophisticated models. However, the impact on the total food supply available to humans is more complex. Here we show that considering all food sources and potential adaptation measures, such as using animal feed directly for humans, famine would result for most of Earth even from a war between India and Pakistan using less than 3% of the global nuclear arsenal. We look at the climate impacts from a range of scales from regional to global nuclear war, and estimate the total amount of food calories available in each nation, including crops, livestock, and fisheries, for each year following a nuclear holocaust. Our findings quantify the global indirect impacts of nuclear war away from target areas, and demonstrate the need to prevent any scale of nuclear war.


2
In a nuclear war, bombs targeted on cities and industrial areas would start firestorms, 32 injecting large amounts of soot into the upper atmosphere, which would spread globally and 33 rapidly cool the planet 1 , 2 , 3 . The soot loadings would cause decadal disruptions in Earth's 34 climate 4,5,6 , which would impact food production systems on land and in the oceans. In 1980s, 35 investigations of nuclear winter impacts on global agricultural production 7 and food availability 8 36 for 15 nations, but new information now allows us to update those estimates. Recently, several 37 studies analyzed changes of major grain crops 9,10,11 and marine wild-catch fisheries 12 for different 38 scenarios of regional nuclear war using sophisticated models. However, the impact on the total 39 food supply available to humans is more complex. Here we show that considering all food sources 40 and potential adaptation measures, such as using animal feed directly for humans, famine would 41 result for most of Earth even from a war between India and Pakistan using less than 3% of the 42 global nuclear arsenal. We look at the climate impacts from a range of scales from regional to 43 global nuclear war, and estimate the total amount of food calories available in each nation, 44 including crops, livestock, and fisheries, for each year following a nuclear holocaust. Our findings 45 quantify the global indirect impacts of nuclear war away from target areas, and demonstrate the 46 need to prevent any scale of nuclear war. 47 5 We underline that impacts in warring nations are likely to be dominated by local problems, like 101 infrastructure destruction and radioactive contamination, so the results here apply only to indirect 102 effects from soot injection in remote locations. 103

Impacts on total human calorie intake 104
To estimate the effect on the total food energy available for human consumption, we 105 consider diet composition, energy content of different food types, crop usage, and changes in foods 106 that we did not directly model (Methods). In 2013, the Food and Agriculture Organization 21,22 107 reported that 51% of global calorie intake was from cereal, 31% from vegetables, fruit, roots, 108 tubers, and nuts, and 18% from meat and related products, of which fish contributed 7%, with 109 marine wild catch contributing 3% (Figure 3a). The crops and fish we simulated provide almost 110 half of these calories and 40% of the protein. Further, only portions of the simulated foods are 111 consumed by humans. Many crops (e.g., maize and soybean) are used mainly for livestock feed 112 and biofuel (Figure 3c). 113 In addition, the total number of calories available as food is highly dependent on human 114 reactions to nuclear conflicts. We assume that international trade in food is suspended as food 115 exporting nations halt exports in response to declining food production (Methods). Furthermore, 116 we considered two societal responses, Livestock and No Livestock, two contrasting extreme 117 scenarios (see Table S3), in between which the complex societal reactions would be likely to fall. 118 For the Livestock response scenario, representing a minimal adaptation to the climate-driven 119 reduction in food supply, people continue to maintain livestock and fish as normal. Calories from 120 all cereals, vegetables, fruit and nuts are reduced by the average reduction in our four simulated 121 crops, and caloric changes from marine wild-caught fish are calculated with business-as-usual 122 fishing behavior. Grass leaf carbon (Figure 2d) is used to estimate pasture change, and simulated 6 crop production change is used to estimate animal feed from grains. Average animal feed has ratio 124 of 46% grass to 54% crops 23 . The No Livestock response represents a scenario where livestock 125 (including dairy and eggs) and aquaculture production are not maintained after the first year, and 126 the national fractions of crop production previously used as feed are now available to feed humans. 127 In addition, fishing pressure intensifies through a five-fold increase in fish price 12 . Similar 128 responses took place in New England in the "Year Without a Summer" after the 1815 Tambora 129 volcanic eruption. Even though the temperature changes were smaller than modeled in any of the 130 nuclear war scenarios here, crop failures forced farmers to sell their livestock because they could 131 not feed them 15 and previously unpalatable fish were added to their diet 14,15,24 . Since all livestock 132 feed from crops is not easily adaptable for human consumption, we test a full range (0%-100%) of 133 the fraction of animal feed that could be used by humans, and use 50% as an example in some 134 plots and tables. In all responses, we do not consider reduced human populations due to direct or 135 indirect mortality or farmer adaptations such as changes in planting dates, cultivar selection, or 136 switching to more cold-tolerating crops. 137 National consequences of calorie loss depend on fallow cropland, regional climate impacts, 138 population levels, and assuming a complete halt of international food trade (Methods; Figure 4). 139 Here, we focus on two calorie intake levels: 2200 kcal/capita/day and 1600 kcal/capita/day. Food 140 consumption of less than 2200 kcal/capita/day would not allow a person to maintain their weight, 141 and less than 1600 kcal/capita/day would be less than needed to maintain a basal metabolic rate 142 (also known as the resting energy expenditure), and thus would quickly lead to death 24 . With a 5 143 Tg injection, most nations show decreasing calorie intake relative to the 2013 level (Table S4), but 144 still sufficient to maintain weight ( Figure 4). With larger soot injection cases, severe starvation 145 occurs in most of the mid-high latitude nations under the Livestock Case. When 50% of livestock 7 feed is converted for human consumption in each nation, some nations (such as U.S.) would 147 maintain sufficient calorie intake under scenarios with smaller soot injections, but weight loss or 148 even severe starvation would occur under larger soot injection cases ( Figure 4, Table S5). Under 149 the 150 Tg scenario, most nations would have calorie intake lower than resting energy 150 expenditure 24 except for Australia (see Figure 4 caption). However, this analysis is limited by the 151 Food and Agriculture Organization data, which are collected at national levels. Within each 152 nation, particularly large ones, there may be large regional inequities driven by infrastructure 153 limitations, economic structures, and government policies. 154 The global average caloric supply post-war ( Figure 5a) implies that extreme regional 155 reductions ( Figure 4) could be overcome to some extent through trade, but equal distribution of 156 food is likely to be a major challenge. One could make the assumption of optimal food distribution 157 within each country 8 , in which the maximum number of people are given the 2200 kcal/capita/day 158 needed to maintain their weight and level of activity and calculate the percentage of population 159 that could be supported this way (Figure 5b). Under the 150 Tg case, most countries will have less 160 than 25% of the population survive by the end of Year 2 ( Figure S4). However, people and 161 surviving governments would react in more complex ways, and that is a subject for future research. 162 For example, if some favored people get more than the minimum, then more people would die. 163

Discussion and conclusions 164
Using state-of-the-art climate, crop, and fishery models, we calculate how the availability 165 of food would change in the world under various nuclear war scenarios. We combined crops and 166 marine fish, and also consider whether livestock, including dairy and eggs, continues to be an 167 important food source. 168 8 Even for a regional nuclear war, large parts of the world would have famine. Using 169 livestock feed as human food could offset food losses locally, but does not make much difference 170 in the total amount of food available globally, especially at large soot injections when the growth 171 of feed crops and pastures is severely impaired by the climate perturbation. We find particularly 172 severe crop declines in major exporting countries like Russia, U.S., and China, which could easily 173 trigger export restrictions and then cause severe disruptions in import dependent countries 25 . Our 174 no-trade response illustrates this risk, and shows that African and Middle Eastern countries would 175 be severely affected. 176 Our analysis of the potential impacts of nuclear war on the food system does not address 177 some aspects of the problem leaving them for future research. These include reduced availability 178 of fuel and infrastructure for food production after a war, the effect of elevated UV on food 179 production, and radioactive contamination 26 . We also underline that while this analysis focuses 180 on calories, humans would also need proteins and micronutrients to survive the ensuing years of 181 food deficiency, and we estimate the impact on protein supply in Figure S2. Large-scale use of 182 alternative foods requiring little-to-no light to grow in a cold environment 27 , if possible, has not 183 been considered. 184 In conclusion, the reduced light, global cooling and likely trade restrictions after nuclear 185 wars would be a global catastrophe for food security. The negative impact of climate perturbations 186 on the total crop production can generally not be offset by livestock and aquatic food (Figure 5a). 187 The results here provide further support to the 1985 statement by U.S. President Ronald Reagan 188 and Soviet General Secretary Mikhail Gorbachev, and restated by Presidents Biden and Putin in 189 2021, that "a nuclear war cannot be won and must never be fought." 190 9

Methods 192
We use a state-of-the-art global climate model to calculate the climatic and biogeochemical 193 changes caused by a range of stratospheric soot injections, each associated with a nuclear war 194 scenario (Table 1,  14.8°C (150 Tg soot injection) peaking within 1-2 years after the war with temperature reduction 207 lasting for more than 10 years. The cooling also reduces precipitation over summer monsoon 208 regions. Similar but smaller reductions of solar radiation and temperature are projected in marine 209 regions (Figures 1b and 1d), with resulting changes in lower trophic level marine primary 210 productivity. We applied local changes at every grid cell to the crop and fish models. is not turned on. CLM5crop has six active crops: maize, rice, soybeans, spring wheat, sugar cane, 239 and cotton, and also simulates natural vegetation, such as grasses. In this study, we used the output 240 of the cereals, maize, rice, soybeans, and spring wheat, and of grasses. Although CLM5crop does 241 not simulate winter wheat, we assume winter wheat production is changed by the same amount as 242 spring wheat, which has been found in other studies 11 , however this may underestimate the winter 243 wheat response, because winter wheat would experience colder temperatures during its growing 244 period that would be more likely to cross critical thresholds 11 . Surface ozone and downward 245 ultraviolet radiation would also be impacted by nuclear war 32 , but CLM5crop is not able to 246 consider those impacts, which might exacerbate the losses. In addition, the crop model does not 247 consider the availability of pollinators, killing frost, and alternative seeds. The model simulates 248 rainfed crops and irrigated crops separately, and all results presented here refer to the total 249 production of rainfed and irrigated crops. Irrigated crops are simulated under the assumption that 250 fresh water availability is not limiting. Irrigation water is from the nearby runoff or the ocean 33 . 251 Although evaporation is reduced with cooling, it is possible that our result may underestimate the 252 negative impact from precipitation reduction, especially for the large injection cases.

Combining crop and marine fish data 266
Table S1 shows the total calorie reductions for each of the nine nuclear states from just the 267 simulated crops and marine fish. Data for all the countries in the world can be found in Table S2. 268 To calculate nation-level calories available from simulated crops and fish, we weight the 269 production by the caloric content of each type of food. We use data from the Food and Agricultural 270 Organization 21,22 . Nation-level calorie reduction (%) from total production of maize, rice, 271 soybean, wheat and marine fish is thus calculated as: 272 Where index i is maize, rice, soybean, wheat, or marine fish wild catch, wiy is the caloric weight 275 of each commodity per country each year, Pi is the national production of item i in FAO-Food 276 Balance Sheet (FBS) 21,22 , ci is calories per 100 g dry mass for each item 21 , Riy is national production 277 reduction (%) of each item in year y after the nuclear conflicts, and Ry is nation-averaged calorie 278 reduction (%) of the five items in year y after the nuclear conflicts. 279

Effects on other food types 280
Other cereal, vegetables, fruit, roots, tubers and nuts. National averaged calorie reduction 281 (%) of the four simulated crops is applied to the total calories of other cereals, vegetables, fruit, 282 roots, tubers and nuts in 2013 to estimate simulated nuclear war impacts on this category. 283 Livestock and aquaculture. We assume these two types of food share a similar response 284 to simulated nuclear war as they involve feeding animals in a relatively controlled environment. 285 For livestock we assume that 46% are fed by pasture, and 54% are fed by crops and processed 286 products 23 . Livestock production is linearly correlated with the feed. Annual leaf carbon of grass 287 (both C3 and C4) is used to estimate pasture changes, and reduction of the four simulated crops is 288 used for crop feed changes. For aquaculture, the feed is only from crops and processed products, 289 and the production is also linearly correlated with the amount of feed they receive. Direct climate 290 change impacts on livestock and fish are not considered. 291 Inland fish capture is not considered in this study. Since inland fish only contributes to 7% 292 of total fish production 37 , adding inland fishery will not significantly change the main conclusions 293 of this study.  Food usage of maize, soybean, rice and wheat is calculated from FAO-CBS. In FAO-CBS, 319 maize products are maize and by-product maize germ oil, soybean products are soybean, and by-320 products soybean oil and soybean cake, rice products are rice and by-product rice bran oil, and 321 wheat product is wheat. Products for food purposes is the sum of food supply in each category, 322 and also the processing product minus the total by-products (the difference includes processing for 323 the purpose of alcohol or sugar). 324

Caloric requirements 325
The population percentage supported by available calories calculated for the Livestock and 326 No Livestock responses indicate the macro-level consequences for food security (Figure 4). The 327 current average human caloric intake is 2844 kcal/capita/day (Figure 3). Caloric requirements 328 vary significantly with age, gender, size, climate, level of activity, and underlying medical 329 conditions. The consumption of less than 2200 kcal/capita/day would not allow an average person 330 to maintain their weight 24 . 1600 kcal/day is the resting energy expenditure for an average person 331 and a sustained diet less than that would be life threatening in someone who did not have 332 substantial stores of body fat 24 . We assume 2200 kcal/capita/day is needed to support life and 333 regular labor activity, which means that in Year 2 after the six nuclear scenarios, there are billions 334 of people threatened by food insufficiency (Figure 4). However, further developments in climate models, such as including organic carbon in fire 344 emissions, and better simulating aerosol growth and interactions with the surrounding 345 environment, may improve climate prediction after a nuclear war. 346 CLM5crop and BOATS are also state-of-the-art models, but future simulations with 347 different models would certainly be useful. CLM5crop compares well with other crop models in 348 response to nuclear war forcing 11 ( Figure S1). If anything, CLM5crop underestimates the crop 349 response to nuclear war (Figures 2, S1). Because most crop models were developed for the current 350 or warmer climates, further research is needed to understand how crops react to a suddenly cold 351 environment. Our study is the first step to reveal national food security after nuclear conflicts, but 352 crops may not respond uniformly the same forcing in each nation, given different farming 353 practices. In addition, multi-model assessment will be essential to fully investigate this problem, 354 and crop model developments are important to understand impacts from surface ozone, UV and 355 freshwater availability. 356 Some assumptions in this study could be examined in future work. For example, to turn 357 off international trade, the ratio of local production to domestic supply is applied on a national 358 level. Also, to calculate national calorie intake after nuclear wars, we assume that food is evenly 359 distributed in each country. Economic models will be necessary to further understand the 360 contributions of trade and local food distribution systems to human calorie intake after nuclear 361

conflicts. 362
This study uses calorie intake from FAO data, and food loss and waste are not considered,   The left two maps are the caloric intake status in 2013 with international trade on and off, the 429 middle column is the Livestock Case, and the right column is the No Livestock Case with 50% of 430 livestock feed used for human food. All assume no international trade and that the total calories 23 are evenly distributed within each nation. Food consumption of less than 2200 Kcal/Capita/Day 432 would not allow an average seized adult to maintain their weight, and less than 1600 433 Kcal/Capita/Day would be less than needed to maintain a basal metabolic rate (also called resting 434 energy expenditure) 24 , and thus would lead to death after an individual exhausted their body energy 435 reserves in stored fat and expendable muscle. Australia is the only nation with enough calorie 436 intake under the 150 Tg scenario, but it may be greatly overestimated. After we turn off 437 international trade, wheat contributes almost 50% of the calorie intake in Australia, and production 438 of rice, maize, and soybean in Australia are less than 1% that of wheat 22 . Therefore, the wheat 439 response to simulated nuclear wars largely determines calorie intake in Australia. Since we use 440 spring wheat to represent wheat, and simulated spring wheat in Australia shows increasing or small 441 reductions under nuclear war scenarios in which more favorable temperatures occur for food 442 production, the calorie intake in Australia is more than other nations.