A sensor-based decision support system for variable rate irrigation systems

The sustainability of irrigated agriculture is threatened by increasing water demands from a growing population, diminishing aquifers, recurring droughts and the effects of a changing climate. The commercial availability of variable rate irrigation technology for self-propelled sprinkler irrigation systems gives farmers access to unprecedented control of the irrigation water applied to their fields. Unlike conventional irrigation systems, which are designed with the objective of achieving uniform irrigation, VRI systems can apply variable amounts of water to a field. In combination with sensing systems that can estimate the individual water requirements of different portions of a field, VRI systems can improve water management because water can be applied in the right amount at the right location.

The commercial availability of variable rate irrigation technology for self-propelled sprinkler irrigation systems gives farmers access to unprecedented control of the irrigation water applied to their fields.

Decision support systems

Although VRI systems have been commercially available since 2004, their adoption has been slower than expected in part because there is a lack of decision support systems capable of achieving an accurate estimation of plant water requirements in different portions of a field. To achieve this goal, such a decision support system must be supported by a network of sensing systems with a type, number and spatial arrangement of nodes to ensure an accurate estimation of site-specific water requirements. An irrigation scheduling supervisory control and data acquisition system (ISSCADAS) was developed by scientists with the USDA-Agricultural Research Service in Bushland, Texas, to fill this gap. The ISSCADAS incorporates site-specific irrigation scheduling methods informed by weather, plant and soil water sensing systems (see fig. 1).

A software named ARSPivot was developed to allow the seamless integration of the ISSCADAS into the operation of VRI center pivots. The software uses weather data and canopy temperature data collected throughout a field using a network of wireless canopy temperature sensors to generate site-specific irrigation prescription maps based on the estimation of plant stress. A network of soil water sensors can be used in combination with canopy temperature sensors to generate site-specific irrigation prescription maps based on the estimation of plant stress and soil water status (see fig. 2).

The ISSCADAS and its ARSPivot software have been used in experiments involving corn, soybean, cotton, grain sorghum and potatoes that have been conducted in semi-arid, semi-humid and humid regions in the United States. The overall results have consistently shown no significant differences in the yield obtained per unit of water between the ISSCADAS and the neutron probe soil moisture sensor, which is considered the gold standard for irrigation scheduling but requires expensive and slow manual labor and specialized training. Thus, no significant difference with the neutron probe indicates success with the ISSCADAS.

One major limitation of ARSPivot is that it only supports VRI center pivots from a single manufacturer that is operated with an outdated control panel. Due to this limitation, the ISSCADAS cannot be used on VRI linear move irrigation systems nor on VRI center pivot systems from other manufacturers. Although center pivots are far more common than linear move systems, supporting linear move systems is particularly important for the community of agricultural researchers and extensionists, given that these systems can irrigate the rectangular shapes of agricultural fields that are common in experiment stations.

Improving the system

A three-year research project was started in the summer of 2023 with the objective of improving the ISSCADAS and its ARSPivot software. The project is funded by the USDA-National Institute of Food and Agriculture’s Engineering for Precision Crop and Water Management program. The project, titled “Partnership: Making the most of a limited water supply by improving a site-specific irrigation decision support system,” aims to make ARSPivot compatible with any type or brand of center pivot or linear move irrigation system with VRI capabilities. The new version of the software will also incorporate an interface to a crop water use and yield model, with the objective of allowing the software to analyze many potential irrigation management decisions (i.e., when to irrigate and how much to irrigate in different parts of a field). Data collected from the network of weather, plant and soil water sensing systems will be automatically fed into the crop water use and yield model to improve its estimations.

The project also involves alfalfa, corn and cotton experiments in Reno, Nevada; Davis, California; and Bushland, Texas, respectively, using VRI linear move and center pivot systems. These experiments will be conducted to determine if site-specific deficit irrigation scheduling methods incorporated in the new version of ARSPivot can improve the crop water productivity of these crops, in comparison to deficit irrigation scheduling following local best management practices. Deficit irrigation can reduce demands on aquifers and other water supplies but must be carefully managed to reduce yield losses that can lead to unprofitable operations.

The new version of ARSPivot will be released as an open-source software that can be modified by anyone interested in using the software for different crops or regions or interested in incorporating additional functions to the software. The resulting sensor-based decision support system for self-propelled sprinkler VRI systems will provide the community of agricultural researchers, extensionists and consultants with a much-needed open platform upon which novel site-specific irrigation management strategies based on sound scientific principles can be evaluated and implemented.

Manuel Andrade, PhD, is an assistant professor of water and irrigation management for the department of agriculture, veterinary and rangeland sciences at the University of Nevada, Reno, Nevada.
Susan O’Shaughnessy, PhD, is a research agricultural engineer for the USDA-ARS Conservation and Production Research Laboratory in Bushland, Texas.
Steven Evett, PhD, is a research soil scientist for the USDA-ARS Conservation and Production Research Laboratory in Bushland, Texas.
Isaya Kisekka, PhD, is a professor of agricultural water management and irrigation in the department of land, air and water resources for the University of California, Davis, California.