Precision Agroecology

A Vision for a Sustainable Future

The concept of Precision Agroecology was coined eight years ago [1], but the concept has not caught on beyond that paper and the description given there is distinct from the vision I present today. While the paper discussed the possibility of using technology to improve ecological outcomes while producing food — it centered on an approach that still reflected the values and theory of traditional farming. Traditional approaches to farming in European history have approached it as an industry of inputs and outputs. The theoretical basis for agroecology, on the other hand, attempts to view agriculture environments as complex systems that are not simply a chain of events but more complex systems that involve cycles and often intense externalities.

By approaching agriculture using systems theory and viewing agriculture as a practice of control theory, we can start to model in more precise terms the often scientifically shaky field of agroecology. In doing so, we can gain insights that can further inform how we can optimize both economic and ecological outputs using novel methods. Rather than attempting to reduce all complexities, as agriculture research has historically done, the methods of signal processing and artificial intelligence could supplement farmers on the nuances of their practice while increasing genetic diversity within the field. The vision of Precision Agroecology should ultimately be to precisely monitor, define and regulate complex agricultural systems in ways that produce optimal yields, human health, and ecological outputs. It is a rejection of oversimplification that has plagued European traditional agriculture and a challenge to researchers to dive into the challenges of complexity.

This approach to farming centers on the use of new data approaches to understand complex systems that were not accessible to us before. For example, by analyzing maps of nitrogen within soils, legumes could be targeted at regions that are nitrogen deficient while other crops could be placed in regions that have an excess of available nitrogen within the soil. This example can also extend to water and soil content. By having a more complete picture of which areas receive more flooding, higher pHs, etc. complex maps can be built that pair desired crops with micro conditions that would return higher yields and better suit the environment. Data collected about past seasons can ensure a viable crop rotation is maintained that reduces the risk of disease and the need for pesticides.

Multispectral remote sensing from drones can scan fields on a daily basis to search for irregularities and diseases. New technology allows disease detection to be spotted much sooner than the visual spectrum alone. This can allow a more reactionary approach which eliminates diseased plants before the disease becomes widespread and eventually eliminate the need to do preventive pesticide sprays that can harm both farm workers and beneficial insects.

In order to realize this vision, more investment into key research areas is necessary. Currently, there are no Unmanned Aerial Vehicles (UAVs) that are truly autonomous, especially for landing and taking off or have a long enough flight time to collect data efficiently. Solar UAVs present options for using solar energy to extend flight time indefinitely, so massive fields can be observed on a daily basis. Improvements in autonomous flight will enable future UAVs to start, collect the necessary data and land again each day without any human input, which will slash the price of data collection. Sensors need to become cheaper and more accurate to allow for real-time data collection in ways that are accessible to farmers. Advancements in artificial intelligence and computer vision will allow agricultural robots to better navigate their environment and make intelligent conclusions.

A key component to making these advancements impactful is long-term and diverse datasets that illustrate the growing conditions of not just a few commercially dominating crops, but information on a wide genetically-diverse set of crops that these systems can call upon to infer decisions about new varieties that come their way. Unless information is available on more than just a few key currently economical crops, precision agriculture will just further enforce the domination of large industrial monocultures instead of enabling a sustainable, diverse future.

Sources

1. Brown, T. T., Huggins, D. R., Smith, J. L., Keller, C. K., & Kruger, C. (2011, December). Manipulating Crop Density to Optimize Nitrogen and Water Use: An Application of Precision Agroecology. In AGU Fall Meeting Abstracts.
2. McBratney, A., Whelan, B., Ancev, T., & Bouma, J. (2005). Future directions of precision agriculture. Precision agriculture, 6(1), 7–23.