Minimizing pitfalls with irrigation decision technology

Improved skill to sense and compile critical indicators in the soils, plants and atmosphere has resulted in large-scale improvement in our capacity to manage agricultural water resources. There are a wide variety of science-based information, resources and decision tools available to a typical irrigator today. The irrigation and precision sensing industry offers a multitude of tools that can help an irrigator make optimal irrigation decisions to not only improve profitability but also produce societal benefits. While adopting and relying on technological solutions for irrigation decision-making, one should ensure that their usability and value addition are maximized, common errors are avoided and best practices are followed. To maximize the return on investment in irrigation decision tools, irrigators can consider these approaches to using technology effectively.

Soil moisture sensor siting considerations

Ensure that soil variability, drainage patterns and historical crop performance within the field are given due weight while identifying soil moisture sensor installation locations. Simple but effective approaches include siting sensors in the soil with lowest water-holding capacity or predominant soil type. If finances allow for multiple sensor locations, it is logical to capture within-field extremes of soil water dynamics to manage uncertainty. If you’re reading sensor output manually, it may be advisable to focus on accessibility to achieve a balance in data retrieval feasibility and representativeness. Avoid areas with possible edge effects, such as potential damage from machinery, poor application uniformity (near an end gun or pivot wheel tracks), near dripline connections, or field head and bottom (for surface irrigation). Multiple sensor installations also help shield against sensor malfunction, failure or unforeseen situations. Remote telemetry capabilities eliminate the need to manually read sensors, making it practically feasible to rely on multiple strategically located sensors across a field. Sensor depths (three to four sensors across the soil profile) should also be strategically selected for local field conditions (soil types, crop rooting depth and irrigation system characteristics) and individual sensor outputs should be appropriately weighted, rather than simply averaging sensor outputs.

Reliable soil water retention properties and irrigation triggers

To ensure success with soil moisture sensor-based irrigation scheduling, it is equally important to use high fidelity estimates of field-specific soil water retention properties such as field capacity and permanent wilting point. A robust soil-based irrigation strategy requires strong coupling of soil water content data with soil water retention properties. Where possible, sufficient verification of large-scale public data such as SSURGO using lab-determined properties is encouraged. It is common to encounter heterogenous soil layers like clay pans, and soil water retention properties used for irrigation scheduling should account for such heterogeneity. Soil moisture sensors are subject to errors and uncertainties, varying by sensor technology, soil texture and structure, and are challenging to predict and correct. Despite these, one can identify appropriate sensor-specific refill points to implicitly account for sensor errors. Careful retrospection of sensor output from previous years can help identify “apparent” saturated conditions, field capacity and stressed conditions, to estimate refill points that are suited to a sensor model and soil. When visualizing and comparing datasets across multiple soil profiles, it is often better to use relative indicators such as soil water depletion rather than absolute water content, to account for any within-field spatial heterogeneity in field capacity.

A robust soil-based irrigation strategy requires strong coupling of soil water content data with soil water retention properties.

Optical sensor considerations

Users of optical technology such as visible light, near-infrared and infrared temperature sensors for irrigation scheduling should verify that the crop footprint is free of nonwater stresses such as fertility and nutrition stress. Electromagnetic radiation can respond similarly to biotic and abiotic stresses, so the monitored area should be visually evaluated through the growing season for any other stressors to ensure that the stress signal in the sensor output is entirely because of water stress.

Checkbook method considerations

When relying on a checkbook method of irrigation scheduling, ensure that soil, weather and plant factors are accurate and reflect field conditions. Crop water use (evapotranspiration) is the largest, most complex and uncertain component of the soil water balance, and its robust determination is central to the accuracy of the checkbook method. In absence of a personal weather station, proximity to a publicly available automated reference weather site (e.g., ET networks) is critical to accurately capture site-specific evaporative demand and precipitation. Additionally, it should be ensured that reference evapotranspiration and crop coefficients used represent the same reference surface (grass or alfalfa) and computation methodology. Where possible, local research-based crop coefficients should be used.

Relying on multiple strategies

Wherever possible, more than one irrigation tool should be used for decision-making. If using soil moisture sensors to track soil moisture depletion, it is also advisable to complement them with a regionally relevant version of the many free-to-use irrigation apps based on the checkbook method. If using optical sensors to schedule irrigation, one can couple these with concurrent soil moisture data to verify if stressed conditions are reported via both sources. This can help in added scrutiny over a single irrigation scheduling method, local evaluation of free-to-use irrigation resources for further refinement and adjustments, while also improving trust in technology-assisted decisions through multiple sources. It is also useful to regularly interface with large-scale publicly available current and forecasted information on soil moisture and drought conditions for early warnings and improved preparedness.

Meetpal Kukal, PhD, is an assistant research professor of agricultural and biological engineering at the Pennsylvania State University.