Many who work in the AgTech space were raised on family-run farms and feel they had no choice but to gravitate to agricultural innovation. Dr. Alex Melnitchouck, who manages Olds College’s Advancing Agronomy Through Hyperlayer Data Collection and Analytics project, is not the son of a producer. Nevertheless, he, too, believes his agriculture career was preordained. That’s because his late father, Professor Dmitry Melnitchouck, was a legendary teacher of Agronomy whose 55-year career the Belarusian State Agricultural Academy recently recognized by naming a laboratory in his honour. As a youngster, Alex spent countless hours in that lab and others like it, setting him on a path that would see him supplement a PhD in Agronomy with an additional MSc in Soils and Environmental Pollution from the University of Reading in England. In short, his education, coupled with 30 years of diverse experience in private-sector and academic agricultural research and innovation, make him an ideal choice to head one of Olds College’s CAAIN-supported initiatives.
“One of the main goals of the Hyperlayer project,” Alex explains, “Is to streamline analytics to be able to accurately predict soil nutrient levels. There are a couple of pragmatic outcomes that will result from our success. First, current methods of sampling and analyzing soil are time-consuming and labour-intensive…and sometimes even dangerous.” He pauses to laugh before explaining that he has, when engaged in manual soil sampling, had guns pointed at him a couple of times by farmers who mistook his intentions. “But beyond that, there’s the issue of fertilizer cost, which can be significant. If predictive analytics can be trusted to provide reliable soil information, then that’s a major step forward in precision ag and allowing producers to reduce inputs while increasing yields.”
Alex is just as enthusiastic when extolling the project’s other virtues, which include collecting many layers of geospatial data for various existing and potential applications; validating emergent technologies for would-be innovators; teaching students in a real-world setting how to collect and analyze data; and disseminating project results and communicating the underlying science. Alex points to his time spent in industry as being particularly valuable when it comes to validating new AgTech. “First, I have a good feel for the kind of data the manufacturers need to assess the quality of their product. But there’s a second element that’s almost as important. Olds College receives many requests for assistance with evaluating the effectiveness and readiness of new agricultural techs. We have an excellent reputation that keeps getting stronger. But sometimes we have to say ‘no,’ and I have the background to be the bad cop.”
The project itself has yielded many interesting, and sometimes unexpected, results. For example, one request was made of the team to cross reference crop diseases against every layer of data they had accumulated. This research was conducted collaboratively with BASF/Xarvio and Agriculture and Agri-Food Canada. The expectation was that the results would show prevalence and severity to be a function of biomass density. But as it turned out, they revealed that another important contributing factor is field topography. In other words where there was higher ground there was less humidity and therefore less disease, and conversely, more in low-lying areas. This is just one example of being able to provide unanticipated value, and more are sure to emerge over time. This kind of analysis is possible only because of the massive amount of data the project is generating. Currently the researchers are building a detailed one-acre grid database of more than 20 different layers or characteristics, including soil organic matter, carbon, nitrogen, potassium, and micronutrients search as boron, zinc, and iron. And that’s just basic soil information. In addition, they are gathering data on aspects such as gamma ray soil spectrometry, soil electrical conductivity, multi- and hyperspectral-satellite imagery, and yield-monitoring data. They have also analyzed 60 layers of compaction, one for each centimetre measured by a digital penetrometer. With all this in mind, it’s easy to understand why Alex and his team feel this is the most analyzed piece of agricultural land in the world.
Having all this data is great but what can it do to improve the agricultural sector? What is its practical value? Alex explains it this way, “When you use GPS, the more satellites you’re connected to, the more accurate the reading of your location. We’re taking the same approach. If a piece of land generates a certain number of bushels per acre, and you have data on the field’s electrical connectivity, compaction, and nutrient levels, you can correlate that information with the yield. That, in turn, allows you to develop predictive algorithms telling farmers what kind of results they can expect from land with specific characteristics.” He pauses to consider his next words. “We know it works in our field, but as I described in my GPS example, the more sources of data you can draw on, the greater your accuracy. That’s why we are developing Hyperlayers of characteristics from different locations in Alberta. At a certain point, we will have collected a critical mass of data that will allow us to create truly predictive algorithms.”
When asked how CAAIN support has benefited the project, he notes that without it, Olds College could not have collected and analyzed the soil samples needed to develop the training data that forms the basis for the machine learning models that will generate the predictive algorithms. The funding also allowed Alex to hire the software developers and data scientists who built a custom digital tool capable of integrating the various types of data. This was a critical because there was no consistency to the different formats, and nothing of the kind existed. He is emphatic when stating that CAAIN’s involvement has permitted the project to proceed at a scale that would otherwise have been impossible.
Ultimately, Alex’s work is not about replacing agronomy. Farming is farming, and much of it will remain the same over the next decade. Crops will still have to be grown using a combination of inputs. But what can—must—change are the tools available to producers and agronomists. Advances in agricultural technology such as those that will result from the Hyperlayer project can make agriculture more efficient, viable, and sustainable, which will be necessary if we are to continue feeding a growing global population.
Total Project Value
Joy Agnew, PhD