Protein-rich crops: growing soybean in vertical farms

Not all crops are created equally: some are simply more nutritious than others. What is the best way to grow these crops and how can we make them even more nutritious? The BU Greenhouse Horticulture of Wageningen University & Research investigated the production of protein-rich crops in their Vertical Farm.

Why should we focus on protein-rich crops?
We are starting to shift towards a more sustainable model for protein production and consumption, also known as the “protein transition”. There is a growing concern for the environmental footprint and ethics of excessive animal protein production. Hence, there is a growing interest in finding alternative, sustainable ways to produce proteins and exploring viable business models. The production of protein-rich crops in controlled environment agriculture is currently still in its infancy and viable business models have not yet been established. Therefore BU Greenhouse Horticulture initiated this experiment as a pilot study.

How do you select which crops to grow?
Protein-rich crops would require high productivity and nutritious content to be viable across markets and production systems. Candidate crops were identified for their potential protein quality and quantity. These crops included for example Leguminosae and other new and ‘forgotten’ vegetables. Crops were finally selected for a high harvest index and yield, as well as a compact crop height and growing cycle. Furthermore a high Digestible Indispensable Amino-Acid Score (DIAAS) was taken into account an used as protein quality indicator. It provides an accurate measure of the amounts of amino acids absorbed and the protein’s contribution to human amino acid and nitrogen requirements of the human body.

Soybean (Glycine max) was considered the most promising candidate: it features the highest protein content and quality, offers a broad spectrum of interesting products, and has the most available reference data.

Why should we grow soybean in a controlled environment?
Traditionally soybean is cultivated in the field, mainly in Brazil and the United States of America. Controlled environment agriculture (CEA) can improve the production of essential amino acids in the cultivation of protein-rich crops. Greenhouses and vertical farms are already being used to produce highly nutritious crops with improved functional ingredients (e.g., minerals, vitamins). The production in CEA offers the opportunity to optimize production, improve crop protection, decrease land area used, and potentially steer crop physiological mechanisms, provided there is sufficient understanding of crop response to the environment.

How do you steer a protein-rich crop?
There is currently little knowledge on the effect of plant growth conditions on protein quantity and quality. The compounds of crops are typically steered using light intensity, light spectrum, day length, temperature treatment, and nutrition. In order to see whether it is possible to steer protein quantity and quality in soybean, a light treatment (low Red: Blue ratio) and a temperature treatment (low temperature) were compared to a reference treatment. The design of the treatments for two cultivars (Viola and Obelix) was based on literature research. To check the effect of the treatments, the cultivation cycle of the soybean cultivars was closely monitored. The protein content of the beans was measured on four instances, in order to determine the rate of protein content accumulation during the early reproductive stage of the pods. The protein quality was also tracked by measuring the amino acid composition of mature seeds.

How could we steer more swiftly and precisely?
Our ability to steer the nutritional quality of the crop is restricted by our ability to measure the plant physiology and specifically the protein content in vivo. The chemical analysis of protein content takes to produce, time, money, and energy. In this project, we worked on a spectral measurement method that can accurately predict the protein content in soybean, based on the spectral characteristics of the pod. Not only does this method circumvent long-term reliance on slow chemical analysis methods, it also circumvents any form of destructive sample preparation by measuring directly on the pod. These measurements are rapid (1-10 seconds) and do not require technical experience. This would make the technique perfect for implementation directly in the production cell without the need for harvesting.

Temperature, light and cultivar all had a statistically significant effect on total yield of beans (g m-2).

  • Temperature: Higher yields were achieved at higher temperatures in both cultivars. This result is related to the number of seeds, as average seed weight was hardly affected by temperature. Seed protein concentration was increased at lower temperatures, which is consistent with information found in the literature. Light treatments may hold the key to further enhancing protein-rich crop production.
  • Light: Light treatment had a clear effect on plant elongation and yield. Plants were grown under light with a low Red: Blue ratio had a 20% shorter stem compared to plants exposed to reference light treatment. Additionally, the yield was reduced. Light treatments may hold the key to further enhancing protein-rich crop production.
  • Cultivar: Total yield was also dependent on the cultivar. The yield of Viola was 16% lower on average than the yield of Obelix. The yield in controlled environment cultivation may be increased by selecting a more ‘suitable’ cultivar.

For spectral imaging, models were trained on random sets of spectral images containing samples from both cultivars. Two additional models were trained only on the spectral images of one cultivar, while the validation was done with the spectral images of the other cultivar.

  • Effectiveness: No published research was found where predictions on bean protein content were made based on the spectral characteristics of the pod. The reported models are based either on reflection spectra collected on dried and/or ground bean samples. The protein content of soybean can be steered more effectively by measuring the pod in vivo.
  • Potential: The results strongly imply that there is potential to translate the developed spectral imaging method into a protocol that can operate within the vertical farm (or a different growing environment) without the need for destructive harvests.

This experiment was a first, but important step into a new field for the controlled environment agriculture industry. In the future, we may be able to actively measure and steer the protein content of crops grown in greenhouses and vertical farms.

For more information:
Wageningen University & Research 


Publication date:

Receive the daily newsletter in your email for free | Click here

Other news in this sector:

Sign up for our daily Newsletter and stay up to date with all the latest news!

Subscribe I am already a subscriber