Spatial Data Engineering for Digital Agriculture

Summary

This thesis shows that spatial data engineering—the careful way we collect, clean, link, and move location-based data—is the foundation of good data science and a must for digital agriculture. It focuses on four connections: linking diverse data (sensors, satellites, records), linking models (weather, soil, crop growth), linking systems (so results arrive on time), and linking people (through shared standards and platforms). Together these turn raw numbers into useful, trusted advice.

The work explains how this base supports artificial intelligence: machine learning is powerful but can be hard to explain or fragile. Blending AI with science-based models, and keeping humans in the loop, makes results both accurate and understandable.

Three synthesis themes guide practice: practical FAIR data sharing, blending models with AI, and human-centred design. The thesis also stresses social and ethical choices that decide who benefits. Strong spatial data foundations are essential to make digital agriculture faster, fairer, and more trustworthy.