> ## Content Index > Fetch the complete content index at: https://www.fieldcircuit.co.za/llms.txt > Use this file to discover other available public pages before exploring further. # When irrigation remote control is worth the complexity: a decision framework for South African growers - URL: https://www.fieldcircuit.co.za/when-irrigation-remote-control-is-worth-the-complexity/ - Published: 2026-09-08T05:23:01.000Z - Updated: 2026-09-08T05:23:01.000Z - Description: Farm SCADA is not one magic box. It is a chain of local controllers, valve actuators, telemetry and phone alerts, and it earns its keep only where control points, night labour and the cost of a missed event justify it. - Author: Louwrens Erasmus - Tags: Agriculture, Irrigation, SCADA, Operations & Resilience, Technology The vendor cases behind this article all sell remote irrigation control on the same three promises: save water, cut diesel and end the 2am rounds. The same cases, read carefully, say something more useful. At their own reporting the systems deliver on those promises only under specific conditions, and the conditions decide whether the remote layer is an asset or a liability on a particular farm. The question is not whether SCADA is good. It is when a remote control layer stops being worth the complexity. In the documented cases behind this article that judgement rests on four conditions: the number of control points and the distance between them, the cost of a missed action, the value of night labour, and whether the water and power situation rewards precision. One limitation should be stated before the framework. No vendor case in the material behind this article publishes a farm-level installation price in rand, and no independent South African cost figure appears in the sources cited here. The framework below therefore sorts the conditions that make a remote layer worth considering, and it leaves the rand-cost question open for the grower to price against a written integrator quote. ### What a real install is A commercial irrigation control system is a chain of small devices, not a single product. Local controllers sit at the point of control and run the schedule even when nobody is watching, as with the panel a Valley or Zimmatic pivot carries. Field decoders sit near the valves. At Costa Group's Australian avocado project, [Mottech's case study](https://mottech.com/project/costa-group-avocado-substrate-australia-2017-to-current/?ref=fieldcircuit.co.za) describes each valve and filter assembly controlled through decoder boxes in the field. Pressure switches confirm the valve opened hydraulically, and a separate switch watches filter flushing. As the Mottech case illustrates, confirming that the valve opened is what separates a useful remote system from one that only sends commands. Actuators move the water. Rubicon's case studies count them precisely: five in-channel and 49 bay actuators on one 806-hectare property, and 69 bay actuators on a 240-hectare farm. A radio or cellular network ties the controllers together, and master software plus phone alerts do the alarming. The Costa project, according to the same Mottech case study, carries a 12 volt backup battery on the controller to ride through power interruptions. The documented installs that pay off share the same shape: many control points, confirmed feedback on each valve movement, and a grower who is not physically present for every event. What a grower takes on is not an acronym. It is the maintenance, communications, power and people needed to keep that chain alive. ### The documented cases One vendor case study describes a Washington State University test orchard running WiseConn drip control. It is not an independent publication of the university's raw data, and no second source confirms the figures. The case study reports a 52.4 percent cut in water against its comparison, a 22 percent gross profit improvement and a gain of about 4,201 US dollars per acre. Treat those as vendor-reported; the underlying trial is a university one, but the comparison and the payback framing are the vendor's. Rubicon's Australian cases are vendor-documented but name real farmers and real outcomes. In its case study, [Russell Pell's 806-hectare](https://rubiconwater.com/project/russell-pell-farmconnect/?ref=fieldcircuit.co.za) dairy and cropping property moved 22 megalitres a day, twice its previous capacity. It cut bay irrigation time from six hours to between 90 minutes and two, and lifted maize to around four tonnes per megalitre. In the same vendor case study, Ray Thornton's 240-hectare farm closed a jammed gate remotely from 30 kilometres away when the system texted him that it would not close. His line is the cleanest statement of the failure mode vendors sell against: some systems do not have fault reporting, so you do not know whether an outlet has opened or shut. South Africa has its own documented case. [Farmer's Weekly](https://www.farmersweekly.co.za/agri-technology/machinery-and-equipment/more-possibilities?ref=fieldcircuit.co.za) describes Agrico Web Control letting a farmer near Hopetown run pivots on variable-speed pumps remotely from Bloemfontein, starting pumps three or four kilometres away from his phone and adjusting pressure across contours. It is a working example of the scattered-operator pattern, not a theoretical one. ### What fails The failure that worries people most is also the least documented in the cases behind this article. No published farm-level incident database records a communications dropout leaving a pump running until damage, and the vendor case studies that describe control hardware do not publish a standard statement of what the controller does when the radio or cellular link drops mid-run. The nearest documented equivalents are deliberate attacks, and they are analogues only, not evidence of the comms-drop failure mode. In April 2023, hackers hit irrigation controllers at farms in Israel's Jordan Valley and disabled automated irrigation, forcing growers back to manual operation. The Oldsmar water treatment incident in Florida in 2021, where an attacker raised a chemical setpoint through remote access and was only stopped by an operator watching the screen, shows what a remotely reachable control surface can do when it is abused. CISA's advisory record on internet-exposed controllers with default credentials, [AA23-335A](https://www.cisa.gov/news-events/cybersecurity-advisories/aa23-335a?ref=fieldcircuit.co.za), describes the class of cheap field equipment that increasingly sits on farm networks. For a farm, the alarm system is not a control room with shifts. It is the grower's phone, or nobody. In the author's judgement, an alarm that wakes someone on false positives every night is worse than no alarm, because the operator will learn to discount alerts that keep firing without a real event. ### The South African layer Load shedding is the difference between the South African decision and the Australian or American one. Remote control does not help if there is no power. Even a telemetry-only connection has value in that environment: it tells you the pump stopped and when it comes back, which is itself the first step of a response. The cost of silent outage is documented: in [Agri SA's 2023 survey](https://www.africanfarming.com/2023/06/28/load-shedding-slashes-yields-and-shrinks-herds/?ref=fieldcircuit.co.za), around 39 percent of farmers reported lower crop yields under load shedding and 56 percent lost income. Where load shedding is documented to cut yields, a controller with battery backup of the kind Mottech describes at the Costa project merits an explicit place in the specification, not an afterthought. There is a second documented reality. If the cheap win on a farm is scheduling discipline and soil moisture sensing rather than full automation hardware, the decision framework has to say so. Where these cases report water cuts, the cuts come from sensor-driven scheduling and control rather than from the remote layer alone. The largest reported figure, 52.4 percent at the WSU orchard, is vendor-reported drip-control data. ### The decision framework Remote irrigation control earns its complexity when the sum of three things clears the cost of hardware, communications subscription, maintenance and security work. First, the number of control points and the distance between them. Every good case study is a many-points, scattered-operation story: 50 or 60 actuators across a farm, pumps kilometres apart, an owner who does not live on the place. Second, the cost of a missed action, which is highest on high-value, water-sensitive crops and lowest on cheap-water field crops. Third, the value of the night round and the field drive it removes, the single most repeated benefit in every documented case. A farm with a single pivot visible from the kitchen, cheap water and no night shift can defend a local controller with scheduled checks; for that operation the remote layer may never pay for itself. A farm with scattered control points, a real missed-event cost and a night round that somebody currently does is where the case gets strong. In either case, before signing, the grower should get the integrator to state in writing what the controller does when the link drops, whether outputs are hydraulically confirmed, and who fixes a radio network at 3am. The missing piece is the price. Neither the vendor cases behind this article nor the independent sources cited here publish a farm-level installation cost in rand, so the grower cannot complete the framework from this article alone. Before the money moves, put two questions to the integrator or scheme operator in writing: what does the controller do when the radio or cellular link drops mid-run, and what does a farm-level installation cost in rand today? Both answers determine whether the remote layer is an asset or a liability, and both have to come from a written quote rather than a brochure.