Moving towards the system perspective

The third article in a series on space food, an introduction to the systems thinking that will define how we feed humans beyond Earth.

In the previous articles in this series, I introduced you to the idea that food is far more than a source of nutrients, and why the psychological and sensorial dimensions of food matter enormously in the context of deep space missions. But now I want to take a step back, zoom out, and look at the bigger picture. Because as important as all of those things are, they represent only pieces of a much larger puzzle.

Here is the problem. When most people think about feeding astronauts on the Moon or Mars, they think about technology. Plant growth chambers, hydroponics, insects, algae, 3D-printed food. And yes, all of these are important. But here is the thing: a food production technology is not the same as a food system. And that distinction matters more than almost anything else when it comes to getting this right.

Let me explain what I mean.

What is a system, really?

The word system gets thrown around a lot, and it can mean almost anything. A set of pipes. A supply chain. A smartphone. But when we talk about a food system for the Moon or Mars, we are talking about something far more complex: a dynamic network of technologies, biological processes, human behaviors, and resource flows, all operating within a tightly constrained and closed environment where nothing can simply be discarded and every failure has consequences.

To avoid scattered or overly speculative thinking when approaching this kind of challenge, a systems perspective becomes not just useful, but essential. What that means in practice is that we need to think about the food system not as a collection of individual parts, but as an integrated whole where production, post-harvest management, waste management, preparation, and the social and cultural dimensions of eating are all connected through feedback loops and resource flows.

No single component of this system can be evaluated in isolation. A production technology may be entirely useless without the proper post-harvest management capabilities, if the crew's workload makes it impractical to operate, or if the food it produces is simply not palatable enough to eat. And as we know from the Mercury and Apollo missions, astronauts who will not eat their food will become malnourished, regardless of how nutritious that food theoretically is.

Scales: thinking in time, space, and governance

When designing a food system for deep space, one of the first things to get straight is the concept of scale. Scale, broadly speaking, refers to the level of organization in time, space, and governance.

At the temporal scale, a food system needs to function reliably across timeframes that range from the day-to-day, ensuring the crew is fed today and tomorrow, to the long-term, managing propagation materials, recycling cycles, and gradual changes in crew composition over months or years. It also needs to account for decadal changes in the technology landscape and the gradual evolution of mission objectives.

Spatially, things become equally layered. At the individual level, a crew member's consumption patterns and available working time directly shape how resources and labor are organized at the habitat level. That, in turn, influences how the elements of the food system are configured and how different subsystems interact with one another. And zoom out further still, and you are dealing with Earth-level decisions: international research funding, regulatory frameworks, and the priorities of national space agencies all shape what gets built and how.

Understanding how all of these scales overlap and influence one another is essential when setting a strategic objective for any space food production system. A decision made at the policy level on Earth today will have consequences in a lunar habitat two decades from now.

Cross-scale interactions: when levels talk to each other

Understanding scales is one thing. Understanding how they interact is another.

A shift in the power budget at the habitat level, for example, does not just affect the lights in a plant growth module. It cascades. It limits which crops can be grown, altering nutritional output. It may force a shift toward organisms with lower energy requirements, like insects or mushrooms, which in turn changes the gas balance inside the habitat, affecting CO2 sequestration and oxygen production. Cause and effect are not always proportional or straightforward in complex systems.

These cross-scale dynamics are why requirements from the individual crew member can feed back all the way up to international regulatory standards, and why global research trends can accelerate or constrain what a small crew on the Martian surface is actually able to eat. Everything is connected, even when the connections are not immediately obvious.

Elements: the building blocks of a space food system

So what does a space food system actually consist of? In its most fundamental form, it can be broken down into five primary elements: production technologies, post-harvest management (storage and processing), waste management, preparation, and the socio-cultural element of consumption.

Each of these elements can itself be examined at multiple levels. Production technology, for instance, can be studied at the level of a single LED light inside a growth chamber, or at the level of an entire habitat's integrated life support system. This nested structure means that the food system is not just complex, it is multilayered, with emergent properties at each level that cannot be predicted simply by looking at the components below.

It is also worth noting that most research and development efforts today focus almost exclusively on production technologies. This is understandable, because growing food in space is a fascinating and technically demanding challenge. But in doing so, critical elements such as post-harvest management, waste valorization, preparation, and the social dimension of eating are frequently overlooked. And a food system with excellent production but no meaningful waste recovery or preparation capability is not a food system at all. It is just a farm with nowhere to go.

Cross-element interactions: where the real complexity lives

Here is where things get genuinely interesting, and genuinely difficult.

The elements of a food system do not just coexist, they are interdependent in ways that can produce completely unexpected outcomes. A production technology that yields highly nutritious food may, at the same time, produce food with limited sensory appeal. If the crew finds it unpalatable, consumption drops. If consumption drops, caloric and nutritional intake drops. And the nutritional content of food has no significance whatsoever if the food is not consumed.

This type of feedback means that you cannot optimize one element without accounting for the others. A seemingly minor design decision in post-harvest processing can affect palatability, which affects crew morale, which affects cognitive performance, which affects the crew's ability to operate the very production systems that feed them. These ripple effects are the reason why unsystematic approaches fall short. The production, post-harvest management, waste management, preparation, and socio-cultural elements of a food system must be developed in concert, with robust pathways for the exchange of both resources and knowledge.

Drivers: the forces shaping everything

Now, why are we building this system at all, and what forces are shaping how it develops? These are not trivial questions.

Drivers are the major forces that determine how a food system operates, evolves, and responds to change. For a space food system, these include nutritional and health needs, psychological and socio-cultural well-being, environmental and resource constraints, technological innovation, regulatory frameworks, and economic considerations. And crucially, these drivers do not operate independently. They interact with each other, and they do so across all of the scales I described above.

And then there is a driver that often gets overlooked in the space context: the sustainability challenges we face right here on Earth. Feeding a growing global population with fewer resources, less arable land, and a more unstable climate is one of the defining challenges of our time. This matters for space food development, because the potential to spin technologies off into commercial terrestrial applications directly influences funding priorities, research directions, and the pace at which innovation moves. In other words, what gets developed for the Moon is not only shaped by what astronauts need. It is also shaped by what the market on Earth might one day find useful. That is a powerful driver, and it means that the technologies we choose to develop, and the way we design them, will be influenced by their potential beyond the habitat.

Environmental drivers are a good example. Reduced gravity affects heat and mass transfer, fluid behavior, and the physicochemical properties of food. This is not a minor inconvenience. It means that food processing methods designed on Earth will not necessarily work the same way on the Moon or Mars, and new approaches will need to be developed. Add elevated radiation levels, which can increase the pathogenicity of certain bacteria and degrade food shelf-life stability, and the engineering challenge becomes considerably more demanding.

Understanding the drivers, and how they interact, is not just an academic exercise. It is what allows us to make strategic decisions about where to focus research and development efforts, and why. Without this understanding, we risk spending time and resources solving the wrong problems.

Robustness versus resilience: two different ways to survive

These two words get used interchangeably, but in systems thinking, they mean very different things. And understanding the difference matters a great deal when designing food systems for environments where failure is not an option.

Robustness is about withstanding known threats. It is the fail-safe approach: you define the risks, you engineer against them, and the system resists. This is the language of traditional engineering, and it is absolutely necessary. Systems should be designed with redundancy and failsafes to withstand technical and programmatic challenges.

But resilience is something different. Resilience is about the capacity to adapt to unforeseen disturbances, the threats you did not see coming and therefore could not engineer against. Where robustness gives you a system that resists, resilience gives you a system that absorbs, adapts, and recovers.

Attributes that enhance resilience include things like modularity, which allows the crew to physically reconfigure hardware to bypass local malfunctions; diversity, which means that a failure in one production technology does not collapse the entire food supply; and transparency, which enables quicker detection and resolution of vulnerabilities before they cascade. Crucially, resilience in a food system extends beyond the technical. Since these are socio-ecological constructs, the crew's capacity for self-organization, their ability to adapt strategies, experiment with preparation techniques, and respond dynamically to changing conditions, is itself an attribute of resilience.

The goal, in short, is not just a system that does not break. It is a system that keeps feeding the crew even when things go wrong in ways nobody anticipated.

System-level requirements: a new set of rules

Developing components for a space food system requires revisiting the frameworks we use to evaluate them. NASA's Technology Readiness Level, or TRL, is the standard metric for assessing how ready a technology is for deployment. But TRL was designed to evaluate individual components in isolation. In a complex, multi-element food system, that is no longer sufficient.

A crop production system, for instance, must not only operate efficiently at the module level. It must also integrate seamlessly with post-harvest management, habitat-wide life support systems, and broader mission architecture. The system-level requirements that flow from this include things that might seem obvious but are frequently underestimated in practice: integration, meaning that all technologies must be compatible with one another and with the habitat infrastructure; adaptability, meaning that systems must flex to accommodate changing crew needs or unexpected disruptions; and evolvability, meaning that the infrastructure must be capable of integrating new discoveries and upgrades as missions progress.

Food safety adds another layer. As missions transition from pre-packed provisions to producing food directly on site, entirely new regulatory challenges emerge. Current safety regulations for equipment in space are robust for the operational scenarios they were designed for. But they have yet to fully adapt to the requirements of in situ food production. This gap will need to be addressed long before anyone is harvesting crops on the Martian surface.

Ground test demonstrators: building confidence before we go

All of the thinking above amounts to nothing if it cannot be validated. And this is where ground test demonstrators come in.

Before physical systems are integrated in space, virtual modeling and simulation through digital twins offer a valuable first step. These tools allow us to explore complex dependencies, run sensitivity analyses, and simulate mission-scale dynamics under varying conditions without the cost and risk of building hardware. But virtual models only go so far. At some point, the integrated system needs to be physically assembled and tested under conditions that actually resemble what the crew will experience.

Ground test demonstrators, configured to emulate real mission scenarios, are where technology selections get validated in an operational context. Critically, these are not just tests of individual components. They are tests of how the whole system behaves when all elements are running simultaneously: production systems, post-harvest processing, waste management, resource recovery, and the preparation and consumption experience of the crew.

This kind of joint testing exposes things that no amount of modelling will ever reveal. Dependencies and integration challenges that only emerge when technologies are operating together. Resource imbalances that appear when the system is running under realistic conditions. Unexpected feedback loops between the biological and the human elements of the system. The lessons learned feed back into improved designs, which are then tested again. Iterative, grounded, and fundamentally humble about the limits of what we can predict in advance.

Ultimately, the development of space food systems must follow a pathway from conceptual design and virtual assessment to physical validation, moving from simple modules to fully integrated systems that encompass everything from production to consumption and back again.

Closing remarks

What I have tried to convey in this article is that building a food system for the Moon or Mars is not an engineering problem in the conventional sense. It is a systems challenge of considerable complexity, one that demands we think beyond individual technologies and consider how everything interacts, across elements, across scales, and across the many drivers that shape what we build and why.

The good news is that this way of thinking, with all of its complexity, is also what makes this field genuinely exciting to work in. Every decision connects to everything else. Every improvement in one area creates new possibilities, and new constraints, in another. And the process of working through all of that, methodically and collaboratively, across disciplines that rarely sit at the same table, is exactly the kind of challenge that will ultimately move humanity forward.