Economic Returns From <scp>AI</scp> ‐Driven Precision Agriculture in Degraded Ecosystems: Productivity Effects Measured Using <scp>UAV</scp> Remote Sensing

ABSTRACT Land degradation has significantly reduced agricultural productivity worldwide, with over half of the world's agricultural land classified as degraded, leading to substantial annual losses. Precision agriculture powered by artificial intelligence (AI) offers a promising solution to rehabilitate degraded ecosystems by optimizing resource use and improving yields sustainably. This study evaluates the economic benefits of AI‐driven precision agriculture, focusing on AI‐based irrigation scheduling and productivity monitoring using unmanned aerial vehicle (UAV) remote sensing with LiDAR. This study proposed a deep neural network–based machine learning framework that integrates high‐frequency UAV campaigns acquiring multispectral and LiDAR data over degraded agricultural plots. These datasets are processed by AI algorithms to estimate crop requirements and optimize irrigation schedules. The model, tested in a conceptual mixed‐farming scenario, employs machine learning techniques, including Random Forest (RF), eXtreme Gradient Boosting (XGBoost), and Support Vector Machine (SVM). Compared to the conventional method, these models show accuracy improvements of 3.2%, 2.7%, and 4.3%, respectively, with Kappa coefficients improving by 0.064, 0.044, and 0.087. The results demonstrate significant productivity gains, with crop yields increasing by 15 to 25, along with notable water savings, leading to improved economic returns. Remote sensing measurements show enhanced vegetation cover and biomass on rehabilitated plots. The study concludes that investment in AI and UAV technology can yield a positive return on investment (ROI) through higher yields and reduced input costs over several growing seasons, based on observations in Shaanxi and Hebei provinces, China.