• drought resilience
• irrigation management
• models
• monitoring
• sensors
• water-use efficiency.

The thesis explores innovative strategies to improve irrigation management and drought resilience in perennial tree crops by integrating advanced sensors, physical models, and geospatial analysis tools. A High-Throughput Screening (HTS) system was designed to model root water uptake and identify critical thresholds of water stress by analysing the correlation between actual evapotranspiration and root absorption under controlled conditions.
In parallel, a commercial data acquisition system was compared with an open-source solution based on Raspberry Pi for collecting soil electrical conductivity (ECb) data in the field, demonstrating how the open-source approach can offer a flexible and accessible alternative for precision agriculture.
Another key aspect of the research involved defining homogeneous management zones in an orchard using geostatistical analyses and fuzzy c-means clustering algorithms to integrate agronomic, pedological, and topographic variables. To improve the monitoring of reference evapotranspiration (ET0), an innovative atmometer equipped with pressure sensors was developed, enabling continuous and accurate hourly measurements with the aim of making advanced technologies more accessible to farmers.
Irrigation management was further optimized through the development of SWATMO-Kcb, an expert system that uses a dual-control approach (feedforward and feedback), integrating various sensors to estimate the water stress index in real-time and adapt the irrigation strategy to the actual plant conditions.
Finally, the thesis concludes with a study conducted in an agrivoltaic orchard in Australia, where LiDAR technology was used to analyse the three-dimensional structure of the canopy, light interception, and biomass distribution, providing valuable data to optimize productivity and water-use efficiency in complex agro-photovoltaic systems.