Choosing the right fertigation decision support system for your operation

Increasing margins, managing farm labor challenges and meeting the increasing demands for excellent environmental stewardship are problems that all farms face. These issues are particularly challenging on irrigated farms requiring high intensity management. Properly managing nutrients is an important aspect of all these challenges. For farms with irrigation, fertigation is often part of a good nutrient management plan. According to the U.S. Department of Agriculture, nearly 12 million acres were fertigated in the U.S. in 2018. Of these acres, nearly 3.5 million were in corn production. Orchards and vineyards and vegetable production represented 2.6 and 1.2 million acres, respectively. Other heavily fertigated crops included alfalfa, cotton, wheat, hay and potatoes.

While management practices vary between crops, three primary fertigation management strategies have been the status quo. For many farmers and agronomists, experience and intuition built through years of observation and trials are the primary factors behind fertigation management plans and adjustments. For others, guidance from university extension services is paramount to their management practices. In recent years, plant tissue analysis has also emerged to track plant nutrition throughout the season whether to adjust the plan during the season or prepare a better plan for the next season. While each of these management strategies has merit, they also present challenges.

Particularly, they tend to lack field-specific, year-specific and crop-specific calibration. They also tend to be low-resolution (one data point or decision for the whole field). Coupled with the effects of temporal (season-to-season and within season) and spatial variability in crop nutrient demand, these approaches often fall short of optimal. Either more nitrogen is applied than the crop demanded leaving risk of environmental nitrogen contamination or too little nitrogen is applied to meet crop demand for maximum productivity. Several decision support approaches are available to support farmers and agronomists in making optimal fertigation management decisions. In general, these decision support systems fall into three categories: advanced plant sap analysis, crop and soil models, and sensor-based algorithms.

Plant sap analysis is a step up from standard plant tissue analysis. It involves high-frequency (about weekly) plant-tissue sampling from old and new leaves on the plant. Samples are sent to the lab, analyzed, returned and mapped to the field or sampling location. From there, the analysis must be translated to a fertigation decision. Success with PSA requires proper sample handling, quick turnaround times, consistent timing (time of day, specifically) and identical lab selection. Some of the biggest benefits are that it provides multinutrient information, is not dependent on weather conditions and requires little to no data to get started. Some limitations of PSA are that it is labor intensive, is not calibrated to hybrid/variety or soil type, and generally provides only one data point for a large area.

Crop and soil models integrate weather, crop hybrid/variety, soil property and production practice information to predict total crop nutrient usage and remaining crop nutrient demand. Using a crop and soil model requires input of significant data including field location, application history, soil sample data, crop planting or growth data, and other production practice information. These systems also often require the user to enter a yield or production goal. Once this information is entered, the user can run the model to determine crop nutrient use up to the current date. Some of the benefits of crop and soil models for fertigation decision support are that these models are not impeded by environmental conditions or connectivity, users can generate recommendations on demand, insights for multiple nutrients may be provided, and results may be used for planning and scenario analysis. Some of the limitations of crop and soil models for fertigation decision support include no in-season calibration; extensive recalibration by region, crop and soil; and significant user data input requirements.

Sensor-based algorithms use real-time and accumulated data from images, soil sensors and other instruments to quantify crop nutrient demand and changes in soil nutrient availability. In some cases, sensor-based algorithms combine real-time and accumulated data from sensors with basic information associated with a field such as soil type zones, yield history and other production practices. With these systems, users receive an updated recommendation for action each time that new data is available. Some advantages of sensor-based algorithms include ample in-season calibration opportunities, full-field data coverage, frequent (daily) data collection and limited user data input to get started. Limitations of sensor-based solutions include impedance from environmental conditions (clouds, haze, smoke), limited predictivity and generally single-nutrient insights.

When it comes to selecting the best fertigation DSS to use, it is important to evaluate several factors about those systems and how they align with the needs of your operation. Those factors are accuracy/precision, nutrients analyzed, frequency of insights, spatial resolution, data input required, labor and turnaround time. For example, if I were managing a potato operation with intensive management requirements, high production standards and labor limitations, important factors for my operation would be accuracy/precision, frequency of insights, labor, and turnaround time. For specialty crop operations, the number of nutrients might be important. For farmers and agronomists managing highly variable fields, spatial resolution and labor might be important. Whatever those criteria are for your operation, they are important to use to evaluate the appropriate fertigation DSS.

In general, commercially available crop and soil models are currently best suited for fertigation of well-modeled crops in geographies with frequent cloud cover. In geographies with regularly clear skies, sensor-based solutions are well suited to provide fertigation recommendations across a range of crops. Both these DSS require some user data input, comfortability with software and familiarity with spatial data concepts. If any of those make you nervous, starting with SAP analysis and working with a professional to interpret them is probably the right approach to get you started. As you evaluate fertigation DSS, on-farm trial data is one of the best resources to use. Performance of a fertigation DSS should be validated through published on-farm research trials like those provided by the Nebraska On-Farm Research Network.

Each DSS has its own distinct set of limitations. I’m excited about the frontiers in fertigation DSS, including fusion of model and sensor-based solutions to mitigate environmental limitations to sensor-based solutions and enable simultaneous calibration and predictivity. The future is bright and fertigation DSS of the present are advancing fertigation management today.

Jackson Stansell is founder and CEO of Sentinel Fertigation. He received his master’s in agricultural and biological systems engineering from the University of Nebraska-Lincoln in 2021.