Joe D. Luck 2016-10-25 00:25:39
Current trends and future opportunities We’ve arrived at an exciting time in the crop production industry—what many are calling the Digital Age of Agriculture. While many may define “digital agriculture” in different ways, there’s no question that producers and other professionals in our industry are using data at an ever-increasing rate. In my opinion, using data to drive decisions in crop production, instead of experience alone, is a sure sign that digital agricultural is a real trend—a trend that began with precision agriculture over 25 years ago. What started with georeferenced soil sampling and yield monitors has evolved into sensors of all kinds available from a variety of platforms. The number of sensors generating data in our fields today is remarkable. Modern tractors and implements have become mobile sensor suites that can wirelessly transmit machine or agronomic data to cloud-based storage platforms. Unfortunately, raw data alone holds little value, and that’s one of the biggest challenges we face today. Many in our industry have underestimated the challenges associated with agricultural data collection, processing, and analysis. In fact, I have spoken to many producers who have given up on trying to figure out how to get more out of their data. Several factors may have contributed to this conundrum, including data quality, security, transferability, and ease of analysis. Data quality For most users, as-applied and yield data are the most sought after datasets generated during the year, all of which rely on sensors. While most users want 100% accuracy in these datasets, we must temper their expectations because of the challenges associated with recording this valuable information. No matter the application, sensors always have some inherent error embedded in their output, and while we do our best to minimize this error, it is unavoidable. To use sensor data in agricultural applications, we must georeference the data, which requires additional information from global navigation satellite systems (GNSS). To achieve the highest accuracy possible, we often recommend real-time kinematic (RTK) GNSS correction, which can be costly and often limits adoption by potential users. The need for highly accurate GNSS data becomes even more apparent as as-applied datasets from field operations are layered with imagery collected from aerial platforms, including airplanes or unmanned aerial vehicles (UAVs). I have experienced such negative effects when combining as-applied and yield data with aerial images for on-farm research trials. The future of sensors in agriculture is wide open, ranging from multi-spectral and hyper-spectral sensing of plants or soils during the growing season to grain quality during or after harvest. Development of algorithms or models that incorporate sensor data to provide decision support to end users in near real-time continues to be a critical need in crop production. Changing cultural practices in which growers make decisions based on experience is challenging; however, demonstrating the potential of data-based decisions can pay off if it encourages producers to adopt a proven technology. In Nebraska, for example, sensor-based, real-time control of nitrogen and irrigation applications during the growing season is showing producers the potential for optimizing input use efficiency in their fields. Data security Data security has been a hot topic in recent years. In agriculture, producers see the data they generate as a valuable product, and they want to control and protect that product. Most producers focus on data generated from their field equipment, which can flow into either machine or agronomic data streams. Many companies have also noticed the value of producer-generated data. Data agreements are evolving to help protect producers’ agronomic information while providing industry access to machine-related data, with the goal of improving equipment performance. I have been fortunate to be involved with the Ag Data Coalition (ADC), a diverse group focused on developing a data management platform that will put farmers in control of their data. This farmer-centric effort that the ADC has committed to will continue to drive the conversation about data ownership in a positive direction. Data transferability Data transferability has improved in recent years, but it can still pose a challenge in some cases. There has been a great deal of activity in this area, and good work has been accomplished by entities like Ag Gateway, the Open Ag Data Alliance (OADA), and the Agricultural Industry Electronics Foundation (AEF). Data standardization almost seemed like an unachievable task, but members of Ag Gateway have made great progress in the past few years toward this goal, and data exchange will be easier in the future because of it. Along those lines, interoperability of agricultural equipment has been a major challenge for many producers who have to deal with multi-manufacturer tractor-to-implement communication. AEF’s work on ISO Bus standards has helped to alleviate many of these issues via the virtual terminal. Annual Plugfest events sponsored by AEF allow industry professionals from different companies to get together and test their latest electronic components. These efforts will have positive returns for end users and producers. Standardized datasets along with improved communication protocols will help accelerate innovations in machine automation and improve user confidence in the decision support systems that use the data. Data analysis Data analysis has always been challenging because we have to learn proper analysis techniques as well as how to use specific software to achieve our desired results. Geographic information system (GIS) software packages for agricultural data applications have matured over time, along with the rest of the industry. In the early days of precision agriculture, GIS software was not intuitive for most users. Farm management information systems (FMIS) software quickly evolved from basic GIS packages to improve the user experience and organize data in a structured format (e.g., grower, farm, field, year, operation, etc.). Over the past two decades, many FMIS software options have been developed with impressive analysis capacities. Users can now easily build their own nutrient prescription algorithms or conduct multi-layer queries, which can be applied at the field or farm level. While we still have to scrutinize the data and ensure proper analysis techniques, there’s no question that these tools are becoming easier to use. Each year, we offer training to hundreds of producers, crop consultants, and retailers through the Nebraska Extension Precision Ag Data Management Workshops, and it’s clear that agricultural professionals want to learn how to gather actionable information from their data. Many producers and other agricultural professionals continue to push forward with exciting applications for the data they’re generating. Producers are growing more crops with fewer inputs, and they’re using essential inputs like water and nitrogen more efficiently than ever, which is better for the planet. Through many different avenues, agricultural data has been helping to drive these improvements. The trend toward digital agriculture will continue to grow in the future, and I’m excited to see how this new piece will fit into the puzzle of feeding our growing global population. ASABE member Joe D. Luck, P.E., Assistant Professor and Precision Agriculture Engineer, Biological Systems Engineering, University of Nebraska-Lincoln, USA, firstname.lastname@example.org.
Published by ASABE. View All Articles.
This page can be found at http://bt.e-ditionsbyfry.com/article/Precision+Ag+Data+Usage/2621115/350793/article.html.