Featured Post

Holobionts: a new Paradigm to Understand the Role of Humankind in the Ecosystem

You are a holobiont, I am a holobiont, we are all holobionts. "Holobiont" means, literally, "whole living creature." It ...

Tuesday, November 8, 2022

A limit to growth in food production – thoughts about the soil holobiont

What if we are not able to further increase our food production? What will we eat tomorrow?

By Thorsten Daubenfeld

As we are celebrating the 50
th anniversary of the “Limits to Growth” study I recently came across the question “how do we feed the world in upcoming years?”. Some people may argue that we already have sufficient food supply but only need a more effective and efficient system of distribution of the existing food. However, already the late Roman empire stumbled across this challenge and wasn’t able to solve it. Others argue that we have sufficient knowledge at our hands to further increase the yield of our crops (fertilizers, agrochemicals, genetically modified crops) and “technology will solve the problem”.

As a physical chemist, I love data. And the FAO (Food and Agricultural Organization of the United Nations) provides plenty of data on this topic. Together with some of my students, we decided to investigate this topic a little bit more in detail. Our key hypothesis is: There is a clear limit to growth in food production – and it already becomes visible.

We first had a look at the top 40 crops (by global production quantity) and plotted the yield (in t/ha) for each country and each of the last 60 years. Most of them showed a pattern like the one we observed for wheat (Fig. 1):

Fig. 1: Evolution of wheat yield in t/ha, 1961 – 2020. Gray dots represent wheat yield per country for the respective year, the orange line represents the global average yield (weighted by production area).

Globally, we have increased the average yield per hectare more than threefold in the last six decades. So by taking a look at the orange line in Fig. 1, we may argue that there is no indication that the growth in food production may slow down. But what could be more interesting is that there seems to be an absolute maximum in how many tonnes of wheat you are able to produce per hectare. This number has been hovering around 10 t/ha for more than 20 years. No single country, whatever they did to maximize their yield, by whatever technology that was at their hands, was able to cross this limit.

The same pattern can be observed for tomatoes, it is even more impressive in our view (Fig. 2).


Fig. 2: Evolution of tomato yield in t/ha, 1961 – 2020. Gray dots represent tomato yield per country for the respective year, the orange line represents the global average yield (weighted by production area).

The Netherlands was able to massively increase the yield of tomato production by growing tomatoes in greenhouses. But again, whatever they (and others) were able to do by means of technology: the biophysical limit for tomato production seems to be around 500 tonnes per hectare. No single county was able to sustainably surpass this limit in the last 30 years. Despite our celebrated technological advances in genetics and agrochemicals.

In all of the top 40 crops we examined, there is not a single example that shows any signs of (exponential) growth – rather a sigmoidal curve as for wheat and tomatoes that seems to approach a maximum value. Or no growth at all in yield.

You now may argue that we just have to learn from the “top yield countries” and copy their recipe for success to other countries. However, this has not been done, neither for wheat nor for tomatoes – nor for any of the other top 40 crops. Otherwise, we would have seen a much larger growth in recent years. But why?

Looking at the data again, we plotted the yield per country against the production area of the respective county – and obtained the picture shown in Fig. 3.


Fig. 3: Wheat yield per county plotted against production area. Each dot represents the yield in t/ha for one county and one year (1961-2020).

In Fig. 3, you see all countries and all yields for wheat for the years 1961-2020. Of course, this means that the same country is shown multiple times. But you see a pattern that emerges: the larger your production area, the lower your yield. And the “top yield countries” are the ones with the lowest production area. This pattern is similar for other crops as well and so far, my key takeaway would be: we cannot simply “copy” the recipe of the top-yield countries to the top-area countries. To put it simply: greenhouses for tomatoes might work for a small country like the Netherlands (910,000 tonnes of production in 2017). But copying this for China (about 60,000,000 tonnes of production in 2017) would mean a lot (!) of greenhouses.

There is another part of the story that may be subjective, but is part of me as a holobiont: when I think of tomatoes, I always remember some days spent at a friend’s family house somewhere west of Pescara (Italy) at the hillsides of the Abruzzi mountains. They grew their own fruits and vegetables in their garden and, in the summer evenings, we had dinner together outside the house. Part of the dinner was the home-grown tomatoes that were much larger than anything I ever saw before as one tomato slice was as big as my two hands. Coupled with olive oil and sea salt, this was one of the most delicious food I ever came across in my life. This was a tomato from a year when Italy “only” harvested around 52 t/ha. In the same year, the Netherlands was able to produce more than 450 t/ha of tomatoes. I have also eaten a lot of tomatoes from the Netherlands. But not a single one of them was able to evoke such strong (holobiont?) feelings in me like the big Italian tomatoes in my story. So thinking about yield and numbers from the perspective of a holobiont, there is definitely more to food than just “yield optimization”.

But let's come back to our numbers. Another question that we would like to investigate is how the countries with high yields managed to obtain that growth. My guess is that most of them increased use of fertilizers, agrochemicals, or genetically modified crops – which are not sustainable (e.g., we are running out of high concentrated phosphate mines to have sufficient phosphate fertilizers) and whether GM crops are really a “progress” still remains to be seen. After having lived on a farm myself for more than 20 years, I would cast some doubt on this. And whether technology is able to produce the wonderful “tomatoes from the Abruzzes” may also be questioned.

So what do you think? Are we running towards a limit to growth in food? Or am I too skeptical? What is the “price for growth” we are paying or going to pay? As for the latter question, I just would like to point toward the challenge of uranium accumulation in groundwater due to long-term phosphate fertilizer use.

In their 1972 study “Limits to Growth”, Meadows et al. were mainly looking at the accessibility of arable land when thinking about the limits of food production. While this is another major challenge, I think that we should have a look at what we are really doing when “optimizing” yield. All crops have to be grown in soil. And soil is a very complex system, maybe also a holobiont in our understanding. Putting the soil holobiont under permanent and rising stress due to the maximization of one output variable (tonnes of crop per hectare) may not be the wisest way to take care of this system.


Acknowledgments: Thorsten wants to thank his students Diana Carrasco and Mirijam Uhland for their contribution to this work.


  1. I am astonished that you do not mention the matter of the limitations of depletion of water from all sources!!!!!! Water is the limiter.C.

  2. This research article effectively describes the notion of "photosynthetic ceiling", the maximum amount of biomass which can be obtained from the sunlight falling on a square meter in a year, for instance. This will vary for different crops, but grasses are pretty efficient, including corn. There are problems approaching photosynthetic ceiling, as the article discusses. A lot of inputs in fertilizer, irrigation, soil-depletion, pesticides, mechanical plowing, harvesting and transport, etc. are needed to optimize this one measurable goal.

    If we look to small, diversified farms, which were common in most of the world prior to the second world war, we see optimization for best usable output, with available inputs in that location. Farmers are very practical people, and always looking for ways that work-a-little-better.
    My friend Rodger calls that "farm smarts", the ability to adapt quickly with whatever is available. Growing vegetable gardens is heling me learn that. Each garden in each location, each year is somewhat different. That's heling me get some perspective.
    I feel that we need a lot of small, local adaptive gardens and farms to help work out this transition to less cheap energy.

  3. https://www.researchgate.net/publication/367821038_A_Case_for_Limiting_Total_Food_Production_for_Human_Consumption