Alternative Foods as a Solution to Global Food Supply Catastrophes
Table Of Contents
Thesis Abstract
<p> <b>ABSTRACT </b></p><p>Analysis of future food security typically focuses on managing gradual trends such as population
growth, natural resource depletion, and environmental degradation. However, several risks
threaten to cause large and abrupt declines in food security. For example, nuclear war, volcanic
eruptions, and asteroid impact events can block sunlight, causing abrupt global cooling. In
extreme but entirely possible cases, these events could make agriculture infeasible worldwide for
several years, creating a food supply catastrophe of historic proportions. This paper describes
alternative foods that use non-solar energy inputs as a solution for these catastrophes. For
example, trees can be used to grow mushrooms; natural gas can feed certain edible bacteria.
Alternative foods are already in production today, but would need to be dramatically scaled up to
become the primary food source during a global food supply catastrophe. Scale-up would require
extensive depletion of natural resources and difficult social coordination. For these reasons,
large-scale use of alternative foods should be considered only for desperate circumstances of
food supply catastrophes. During a catastrophe, alternative foods may be the only solution
capable of preventing massive famine and maintaining human civilization. Furthermore,
elements of alternative foods may be applicable to non-catastrophe times, such growing
mushrooms on logging residues. Society should include alternative foods as part of its
contingency planning for food supply catastrophes and possibly during normal times as well.
<br></p>
Thesis Overview
<p><br></p><p>Despite technical advancements and an abundance of food globally, food security is a
major ongoing challenge. 870 million people do not have enough to eat and under-nutrition
contributes to the premature deaths of over six million children annually.1
Land degradation,
fresh water scarcity, overfishing, and global warming all threaten to diminish food supplies.
Food demand is meanwhile increasing due to population growth and a rising middle class in the
developing world that can purchase more and different foods. Improved technologies have
helped farmers grow more, but extreme wealth inequality still leaves the world’s poorest
struggling to afford enough food. These and other trends virtually guarantee that feeding
humanity will require major dedicated effort into the future.
But it gets worse. These trends show gradual shifts in food security under otherwise
“normal” circumstances. However, a range of extreme events could cause large abrupt declines
in global food production from conventional agriculture.2
If one of these events occurs, humanity
could face global famine of historic proportion. The collapse of human civilization or even the
extinction of the human species are distinct possibilities. Other species around the world would
likely also go extinct, including species that would otherwise survive the extinction event already
underway.
One major threat comes from events that block sunlight by sending large quantities of dust,
smoke, or ash into the atmosphere. This could happen if Earth collides with a large asteroid or
comet, such as the one believed to have caused the dinosaurs’ extinction. It could also happen
from a supervolcanic eruption, such as the Toba eruption 75,000 years ago that some scientists
propose almost killed off our early human ancestors.3
And it could happen from a nuclear war,
with the atmosphere coated by the ashes of incinerated cities.
Some abrupt food supply threats come from rapid environmental change or direct threats to
crops. These food catastrophes may not be as severe as the sun-blocking catastrophes, but they
can still cause large and abrupt declines in food production. Global warming could cross
thresholds in the Earth system.4
For example, a rapid change in ocean circulation could bring
dramatic shifts in global weather patterns. Specific crops can be threatened by natural pests, as in
the Irish Potato Famine. Biotechnology could bring even more devastating engineered crop
pathogens. For comparison, biosecurity experts are actively debating the potential for certain
“gain-of-function” experiments to cause deadly human pandemics.5
Similar research may be able
to bring crop pandemics. And these are just some known scenarios. Additional threats may lurk
beyond the current horizon of science.
As an illustration of how bad things could get, consider one scenario, a nuclear war between
India and Pakistan. Smoke from the burning cities would block sunlight and reduce global
temperatures by about 1ºC for about a decade.6
Crop simulations project food production
declines around 20% to 50%.7
Combining that with existing poverty and malnourishment gives
an estimated two billion people at risk of starvation.8
And that is for a war with “only” 100
nuclear weapons. A war using more of the world’s 15,800 nuclear weapons would bring even
worse consequences.
There are several solutions for abrupt global food catastrophes. If agriculture is still possible,
it can be diverted from livestock and biofuels production to direct human consumption, though
larger catastrophes would leave less food to divert. Additional food could also come from
oceans, though this a limited option and it could further threaten marine biodiversity. Another
solution is to stockpile food prior to the catastrophe, though this is expensive and it can worsen
pre-catastrophe food security.
In light of the enormous threat of global food supply catastrophe and the shortcomings of
other solutions, we propose a new solution. The essence of it is to produce food with energy
from sources other than the Sun. Ultimately, crops do not need sunlight per se. They just need
energy. We call our solution “alternative food” because it uses alternatives to sunlight, just like
“alternative energy” uses alternatives to fossil fuels. Alternative foods are already in limited
production and could be scaled up following a major catastrophe.9
The simplest type of alternative food is plants grown from artificial light. Today, indoor
agriculture powered by light-emitting diodes is being explored as a solution to land scarcity and
resource-intensive outdoor agriculture.10 These indoor farms could produce any of the crops
currently grown around the world. However, a lot of energy is lost converting the initial energy
source into electricity, then into light, then into plants. As a result, all of the world’s electricity
could feed only a small portion of the world’s population. To feed everyone, other solutions are
needed.
<br></p><p>
A better solution comes from foods powered by fossil fuels. Today, the company Unibio
grows bacteria from natural gas and sells it as livestock feed.11 The same livestock could feed
some people after a catastrophe. More people could be fed by adapting the Unibio process for
direct human consumption. Our prior research on this has prompted charming headlines like
“Bacterial slime: It’s what’s for dinner”.12 But as odd as it might seem, getting food from
bacteria could keep many people alive in a catastrophe. So too for other techniques using natural
gas or other fossil fuels. And thanks to progress in food science, the resulting foods may even
taste good.
A different energy source does not need so much infrastructure: biomass. After a catastrophe,
biomass would be available from the trees and other plants that are still around. Biomass could
be harvested by foraging or lumbering. Collecting a lot of biomass could damage ecosystems,
creating another tradeoff. However, this tradeoff would only be faced in the event of a food
catastrophe. Furthermore, if the sun is blocked, then some or all of the trees would die anyway,
depending on how much of the sun is blocked. This makes alternative foods from biomass an
especially attractive solution.
Biomass can be fed into the food supply in several ways, as illustrated in the food web.
Wood can be fed to beetles, which can in turn be fed directly to humans or to a more appetizing
intermediate species. Using an intermediate species would greatly reduce the amount of food
available to humans. Horses, cows, goats, and sheep can be fed leaves and non-woody plants.
Mushrooms can grow on all of the above types of biomass. Finally, if woody biomass is partially
consumed by mushrooms or bacteria, this could be fed to rats or even chickens.
<br></p><p>
Some plants and plant parts can be fed directly to humans. Familiar foods include nuts and
edible leaves. Less familiar options could also help during a food catastrophe. Some leaves (such
as pine needles) can be boiled to make tea. Some biofuels turn cornstalks and other residues into
sugar with enzymes. Then the sugar is fed to a fungus to make ethanol. But if people are short on
food, they could just eat the sugar. Biomass foods cannot provide everything found in a grocery
store, but they can keep people from starving to death. Some of these techniques can even
improve food security during “normal” times, such as by feeding sawmill wood waste to
mushrooms. As the food web illustrates, the waste from one organism can become the food for
another organism
The best solutions for abrupt food catastrophes will vary from place to place.13 Local social
and environmental factors are important. Some places have more energy for indoor agriculture,
or more biomass, or more fossil fuels. Some places have technical and political capacity that is
better suited for certain solutions. Some places have cultural preferences for certain types of
foods. For these and other reasons, food catastrophe solutions should be developed locally, to
ensure that each community has a solution that works for itself.
<br></p><p>
There is another reason to develop these solutions locally. In the aftermath of a major global
catastrophe, regions could become isolated from each other. Travel, trade, and communications
all depend on complex systems of infrastructure. A catastrophe big enough to damage global
agriculture could also disrupt these systems, though agriculture will usually be more sensitive to
environmental catastrophes than most built physical infrastructure. Self-sufficient communities
will be best positioned to weather out the storm.14
<br></p><p>
Finally, the solutions presented here could also be used to protect biodiversity. Plant
biodiversity is relatively easy to preserve by storing seeds, such as in the Svalbard “doomsday”
seed vault. Protecting animal biodiversity is harder. For that, alternative foods can help. If no
food is available in the wild, humans could divert some alternative foods to preserving nonhuman animal species. It would be impossible to keep every animal alive, but it should be
possible to keep each species from going extinct. As few as 100 individuals can be enough to
prevent a species from going extinct. 100 individuals per species could be fed without any
significant loss to the human food supply. Therefore, in addition to keeping many or even all
humans alive, alternate foods could save most of the animal biodiversity that would have been
lost in a catastrophe.
Everyone should hope that no abrupt global food supply catastrophe ever occurs. But while
people should hope for the best, they should prepare for the worst. Alternative foods are a
solution that could keep millions or billions of people alive during even the most severe food
catastrophes. They require only modest advance preparation and no diverting of food into
stockpiles. Indeed, alternative foods can even strengthen food security now by opening up new
means for food production and using resources more efficiently. For these reasons, and given the
extremely high stakes with abrupt global food supply catastrophes, we believe alternative foods
are a solution well worth pursuing.
<br></p>