Ongoing project
AI-driven tree nut yield forecasting (AS24006)
Delivery Partner: University of New England (UNE)
This project aims to develop advanced, predictive yield models for four key nut tree crops (walnut, pecan, macadamia, and almond) by integrating machine learning with remote sensing, weather data, and ground-based measurements. By partnering with Stahmann Webster, the project will leverage extensive historical and current orchard data, alongside agronomic expertise, to identify the most influential factors affecting yield variability. It will also explore temporal patterns such as biennial bearing and seasonal fluctuations, helping to build scalable models that forecast yield with greater accuracy.
The outcomes of this work will support improved decision-making and farm management practices, with clear pathways to commercial application. By enhancing yield forecasting capabilities, the project will contribute to greater productivity, profitability and resilience across the nut tree industry. Its modelling approaches are designed to be transferable, enabling broader adoption across other crops and regions through industry-wide collaboration, ultimately driving innovation and sustainability across the horticulture sector.
This project is funded through Hort Innovation's Frontiers program
© 2026 Horticulture Innovation Australia Limited.
