By Shane Thomas, Global Digital Ag Lead
The digital evolution within the ag industry has been constant over the last decade and continually evolving; however, it doesn’t come without its challenges. Digital ag and ag tech aren’t always easy to adopt. As of today, there is manual entry, there is incremental effort to ground truth models, assess data and a need to create the habit to be constantly on top of everything so that it is captured, correct and in the right place. But that’s the case with anything worth doing, it takes effort.
Digital ag today is akin to losing 20lbs. If you want to lose weight it comes with committing effort to a regimented exercise or physical activity routine, a plan to consume healthy food in a specific proportion and at specific times and this commitment typically takes months to achieve and then a continued effort to maintain it. This is exactly like planning and inputting a crop plan in your platform. Continued efforts to collect data, monitor scouting reports in-season, and a continued pursuit of understanding the why behind your yields, profitability or whatever metrics you value, is required. Just like losing 20lbs is an accomplishment, the financial implications driven by digital agriculture will be exceptional for farming in North America and beyond.
The good news is that this constant progress will eventually shift things to more informed modelling and more passive data collection, it’s simply a journey to evolve to that point. But it’s coming, and here is a framework in which to think of some of that evolution.
Hindsight, Insight and Foresight
These are the 3 types of “visions” we have to help make decisions, inform decisions and assess decisions. They are uniquely different but are all contributing to what we are constantly striving for: better outcomes on the farm and within the ag industry all the way to the consumer. Within the digital space these types of “sights” are influenced by various tools that I’ll give some examples of.
Hindsight
Using hindsight is the act of taking data and information from the previous year or years prior and utilizing it to make decisions in the upcoming year. It is the most common way to utilize digital tools today.
An example might simply be comparing how one variety fared versus another the previous year, looking at how fertilizer application amounts or methods compared to one another or it might be looking at how combine brands compared to one another when making a combine purchase decision. This is where a unique benchmarking tool comes into good use. Historically, we didn’t always have enough relevant data to look back on to make an informed decision for moving ahead. Now we have access to a much more robust array of data, being captured by unique sensors that enable us to reconvene after a long season and identify a better path forward in the upcoming season.
Real-Time Insights
Insight is having information that enables you to make an informed decision that can impact the outcome of the season in real time. Today we have numerous tools that can provide insight into how to proceed forward based on what’s happening in a field RIGHT NOW!
Again, being enabled by sensors I can point to two specific examples where this is happening today.
1. Soil Moisture Probes
These probes can tell you what the moisture level of the soil profile is right now to up to 4 feet. For irrigation farmers, this is powerful to plan irrigation events, but even a dryland farmer can understand how much moisture is in the soil and whether at depth there is enough moisture to justify an additional nitrogen application for example. This can be taken a step further to having a real time understanding of moisture + nitrogen levels based on sensor and modelling that can give you a data driven insight around how to proceed based on what’s currently in your soil.
2. Imagery-Derived Maps
NDVI imagery has been around for a while, but the resolution was not strong and the frequency was too sporadic. Now, with almost daily imagery and the ability to identify the variation across a field, you can see where a Group 1 wild oat escape may have occurred or where there may be cutworms moving into a crop without having to actually stumble upon it in the field. Real-time insights that can influence the bottom line and help get away from a “we’ll get ‘em next year” scenario to a “Let’s tackle it this year” approach.
Foresight
Foresight is the ability to see around the corner and “anticipate” problems or opportunities before they happen, insuring you have a strategy and plan of attack in place to mitigate any downside risk, or move on an opportunity that may soon arise. When it comes to digital ag, the capabilities to deliver on this type of sight are ever-growing. This is the anticipation of issues happening on farm and is deliverable through modelling and machine learning.
Monocyclic diseases like fusarium headblight in cereals and sclerotinia in canola can decimate yields and quality. These diseases are challenging because they typically infect at specific times and once they do, there is no way to cure them. This leaves the decision window for action to be very small. The good news? Sensors and modelling can enable us to have at least a somewhat clearer view of the future now. With the ability to monitor weather at the field level and forecast/model weather out ~10 days, plus sensors that can assess real time spore populations in the area, you can gain a clearer picture of your fields risk level of these diseases at key times. This empowers to make more informed decisions about what COULD happen and begins to give us a probability of risk (and yield loss) that allows us to make informed decisions.
Systematic, Integrated & Automated Farming Solutions
The convergence of all this technology enables a more systematic approach to farming that ultimately will help to manage costs and make more profitable decisions. To illustrate this, I’ll use a simple irrigation example.
If you use a historical soil test or map of a field you can understand the soil texture, organic matter, pH and more to begin to understand what the water holding capacity of the soil is. Looking at this in combination with yield potential maps based on yield maps, soil and NDVI maps we can begin to understand what each area of the field is capable of producing plus know the fertilizer amounts put down in the soil, enlightening the entire system with what is happening where. Now, you can put in a soil moisture sensor or two in the field to truly understand what the moisture levels are like throughout the field, giving you an idea of when the field is at capacity, or getting close to a yield limiting amount. This sensor can be connected specifically to the irrigation pivot as well as to your phone. Now, we can know the exact amount of moisture in the field, have a prediction of yield, and a soil water holding capacity map that works together in unison to understand when an irrigation event is necessary by zone within the field. This would then send you a notification that your pivot will be turning on to ensure optimal yield (not max yield because your costs will be understood and it will be about profit maximization, not yield maximization) and it will notify you of any plugged nozzles, pivot breakdowns or that the watering event has successfully completed. This systematic, integrated and automated approach is what digital ag is capable of today. And this is just the tip of the ice berg. Exciting right? This saves time, this saves money and it increases profits.
With the continued progression of hindsight to insight to foresight all leveraging big data, we will begin to see a speed up of uptake and use cases across all different types of farms in North America.
While the adoption hasn’t been rapid as of today and the full ability to execute is just starting to be delivered now, I think we can concede that the implications of big data are here, is influencing farming now and will continue to well into the future.
