The Courier & Advertiser (Fife Edition)
Dark Horse shines a light on crop yield
Anew crop monitoring service has been hailed a success after it predicted winter wheat yields within 0.5% accuracy.
The service from Dark Horse Technologies uses satellite imagery and drones to monitor crops, diagnose growth problems and predict yields.
It is being trialled on Niall Jeffrey’s mixed arable and beef unit at Bielgrange Farm in East Lothian.
The farm is part of the satellite farms programme at the Edinburgh-based Agri-EPI Centre – one of four centres of agricultural innovation across the UK – where new technologies and techniques are trialled in real farm settings.
Mr Jeffrey said the Dark Horse technology was trialled on a difficult field of winter wheat, where the crop was poorly established.
“Dark Horse remotely programmed a drone mission and all I had to do was go to the field and press start then, when it finished the mission, upload the memory card online to Dark Horse,” said Mr Jeffrey.
“We intially predicted a yield of around eight tonnes per hectare (t/ha) and revised this throughout the growing season as Dark Horse showed us how things were progressing.”
He said the final prediction, made three weeks before harvest, was 6.4 t/ ha, and the actual yield was 6.43t/ha.
“Going forward, the prediction could help me with the management of harvest storage and forward selling,” said Mr
Jeffrey.
“We were impressed with the accuracy of the results.”
Dark Horse Technologies founder, Jared Bainbridge, said the monitoring service combined multiple sources of data and imagery into a model capable of predicting yields and mapping croploss events with a high degree of accuracy.
He said: “We are also able to integrate with existing on-farm machines to ensure sprayers target the areas of the field which need the most help, saving time and money.”
Further trials of the technology are planned at Bielgrange including a field-scale seed treatment trial and the measurement of grass output from a paddock grazing system for cattle next summer.