Digital Agriculture


Farmers have a long track record of adopting new and innovative technology to make their farming business more successful. This includes things like seed technologies, crop protection products, agricultural machinery, water management strategies, among many others. In the past couple of decades we have seen an increasing use of digital tools across all parts of farm businesses. This is in line with the trend towards digitization in the wider Canadian economy, and we expect the digitization of farming and the agri-food system to continue, or even accelerate, in the coming years.

A  recent study of European agriculture  predicted that farms may be capable of producing in excess of 2 million data points per day by 2030, up from an estimated 250,000 data points in 2015. A similar estimate is cited in a recent  World Bank report , which suggests that data generated by precision agriculture equipment and related technologies is expected to reach 4.1 million data points per farm, per day, by 2050. 

At the same time, we know that for farmers it isn’t just about generating or producing more data. It is critically important to consider how agriculture data can be used to create new value for your farm and to strengthen the wider Canadian agri-food value chain. We will use the term digital agriculture to refer to exactly that - putting data and digital tools to work on-farm to create new value, strengthen farm businesses and manage risks. 

That said, you will also hear a number of other terms like smart farming and precision agriculture. Some  researchers have noted  that the term digital agriculture is more common in Australia and New Zealand, while smart farming is the more widely-used term in Europe. Based on our experience, you are likely to find both terms used in Canada (e.g.  Pan Canadian Smart Farm Network ,  Olds College Bachelor of Digital Agriculture ).  

And where does precision agriculture fit in? As we will see later in the module when we look at all the various types of agriculture data, precision agriculture is most closely associated with the combination of location data (e.g. GPS or global positioning system) along with advanced agricultural machinery and implements which are capable of controlling how seed, nutrients and crop protection are applied across the field. This concept extends to  precision livestock farming (PLF)  as well, where technologies are used to optimize livestock production, welfare and management. So, precision agriculture is an important part of digital agriculture, but ultimately the scope of digital agriculture goes beyond precision agriculture to include the numerous other parts of your farm business.


How to Organize Agriculture Data

We’ve already mentioned how agriculture data is a broad term that includes a range of different data types. We expect that maps well onto your experience given the wide range of activities that go into running your farm and farm business. At the same time, it can be helpful to have a structure to help organize the range of data your farm generates and manages. In fact, as we will see, there are a number of different ways that you can organize your agriculture data depending on what you want to do with it. This is a recurring theme that will come up throughout both this module, and the rest of the data literacy course.  

In their  Ag Data Glossary,  OpenTEAM, a community of agriculture stakeholders working on open source agriculture technology solutions, defines agriculture data as follows:

“A broad category of data types related to agricultural activities, including data about the land (e.g., soil and fertility data, geospatial data, etc.), data about on-farm crops and animals (e.g., seed type, yield, feed and health information about animals, etc.), data about or generated by farm equipment (e.g., model, fuel consumption, yield maps, etc.), and other information about farm management (e.g., commodity price, farm revenue, employment, etc.).”

Taking a wider perspective that includes the agri-food value chain and food system, a  European study  defined agri-food data as:

“… the representation of facts, events, phenomena or situations in the agri-food sector and the food supply chain, starting from the representation of facts concerning inputs and raw materials, through the representation of primary production, processing, manufacturing, up to distribution, retail, and consumption.”

When we expand our view to include the value chain, from producer to consumer, the scope of what constitutes agri-food data grows considerably to include data related to food processing, food waste, nutritional information, food safety, and more. For this module, we are going to focus our attention on agriculture data that is directly connected to farming and farm businesses. And for the most part, we will be focused on data that is generated and managed by the private sector - farms, agriculture input dealers, machinery companies, etc. 


With that in mind, a useful way to organize the wide range of agriculture data is the following broad categories:

  • Ear of Rice Agronomic Data
  • Cow Livestock Data
  • Desert Island Land Data
  • Notebook Farm Management Data
  • Tractor Machine and Equipment Data
  • Sun Behind Rain Cloud Climate and Weather Data

These categories are taken from the  template agriculture data use agreement  developed by the  Agriculture Data Transparent  (ADT) organization. They are consistent with other similar ways to organize agriculture data (e.g.  OECD ). At the same time, it’s not reasonable to expect these broad categories to perfectly capture all types of agriculture data. Instead, it is expected that individual farms or farming communities may be working with data that doesn’t fit neatly into these categories. For now, we can take them as a useful way to start organizing our thinking about agriculture data.

Taking Another Perspective: Type of Data

In a  recent blog post ,  Leaf Agriculture ’s CTO Luiz Santana shows us a different way to think about agriculture data from the perspective of a computer scientist.  

In this approach, agricultural data is organized based on the underlying properties of the data itself, rather than on its role in farming operations or the farm businesses.  

You can see a visual representation of this approach below. It certainly reinforces the importance of geolocated data (e.g. everything under GIS) in agriculture, where a large amount of the data generated from sensors or precision agriculture machinery is linked to a GPS coordinates, or a specific geolocated region (e.g. a field).  

For more on Luiz’s thinking check out his full blog post on  LinkedIn .


How is Agriculture Data Collected?

Agriculture data can be collected in many different ways:

  • Mobile Phone Walking through a field on foot and noting the state of the crop and presence of disease or pests using a smartphone application that tracks the GPS coordinates of the observations.

  • Thermometer An in-field weather station measures temperature, relative humidity, precipitation and other weather-related data on a continuous basis. This data is stored locally in the weather station and transmitted to the weather station company’s cloud-based server.  

  • Tractor Precision agriculture equipment (e.g. seeder, sprayer, combine) captures data about the machine and the operations it is performing as it passes over the field.  

  • Round PushpinDrone aircraft capture hyperspectral imagery of a field that can be analyzed to produce maps that track the development of crops and the presence of pests or disease. 

  • Test Tube Soil data is measured by gathering physical soil samples that are sent away for laboratory testing.  

The key message here is that any activity that gathers information about your farm can be thought of as data collection. This can include entering data manually into a smartphone app, or Excel spreadsheet. It could also be taking a photo with your mobile phone of a water drainage issue. It could even include creating a paper record, or maintaining a notebook. Although paper-based data can be harder to reuse, it is still an example of data collection. As we learned in the first module, data is any type of information but digital data offers a range of new opportunities compared to traditional paper record keeping.

Of course, there are certain advantages to automated data collection. Given that minimal human intervention is required, data that is logged automatically is more likely to be collected on a regular and consistent basis. Automated data collection can also make it possible to collect a greater volume of data. Remember that volume is one of the 5Vs of big data ( see 🏟️⁠⁠Aggregate Data & Big Data⁠ ). At the same time, it is critical to monitor the automated data collection process to ensure it is properly calibrated and configured to produce accurate, reliable, high quality data that is available when you need it. 



Next:  📁⁠⁠Catalogue of Agriculture Data⁠