Hyperlocal weather data supports more responsive greenhouse decisions

Speaking at the CEA talk at NAMPO Cape 2026, Emile Jordaan, General Manager of Metos South Africa, focused on how local weather measurement and short-term forecasting can support more responsive farm and greenhouse management.

“There are at least three things on a farm that you cannot change: you cannot change the location, you cannot change the weather, and you cannot change the fact that farming coincides with variability.”

He argues that while these factors cannot be controlled, their magnitude and effects can be measured. Accurate measurements, combined with short-term forecasting, can then support changes to operational decisions.

“If we can measure that accurately, and we have a fairly good opportunity to forecast the next seven days, we can do whatever is humanly possible to change decisions on the farm.”

From weather data to operational decisions
For protected cultivation, variability in temperature, humidity, wind, and other environmental conditions directly affects irrigation, energy use, crop protection, and ventilation decisions. Emile highlights spray timing as one example, noting the relevance of delta T, which uses temperature and humidity to assess droplet persistence on foliage. “You cannot do the calculation if you haven’t measured it.”

He also emphasises that measurement has value only when it informs action. “Data that does not change decisions is meaningless, useless.”

For greenhouses and other protected environments, he recommends combining broader mesoclimate measurements with microclimate monitoring within specific production areas. Conditions can vary over relatively short distances, making local measurements relevant to operational decisions.

© MetosHyperlocal forecasting
A central theme of the talk was the use of hyperlocalized weather forecasts supported by a feedback loop between forecast data and measurements from an on-farm weather station.

“A hyperlocalized forecast is possible if there’s a feedback loop between the weather station and the weather forecast.” The measured data can be used to assess forecast accuracy and continuously improve the forecast for the specific production location.

“If you have a weather forecast that feeds off the actual weather data that’s measured on the farm, you have a far better chance of getting the weather forecast right.”

Supporting proactive crop management
Emile describes a range of daily decisions that can be informed by local measurements and forecasts, including irrigation timing and duration, ventilation, heat and frost risk, and crop-protection scheduling.

Disease management is another application. “Temperature, humidity, rainfall and leaf wetness can be incorporated into models that estimate disease risk and, where forecasting is available, the likely development of that risk.”

According to him, the objective is to move from reacting to conditions after they occur towards planning around expected conditions. “We move from reactive to proactive, to proper plan, because you can trust your weather forecast.”

Technology and human decision-making
He stresses that technology does not remove the need for human involvement. Sensors, forecasts, crop models and automated alerts provide information, but operational decisions remain part of the management process.

“Human involvement remains critical. But how you use, for example, AI tools, how you use good technology that is available, will either increase or decrease the speed, and how much or how quickly you can adapt.”

He also notes that weather and climate information can be integrated into existing systems through application programming interfaces (APIs), allowing data to be transferred into other farm-management or greenhouse-control ecosystems.

For more information:
Metos South Africa
Emile Jordaan, General Manager
[email protected]
www.metos.co.za

Trending Products