Optimization of Furrow Irrigation in Arid and Semi-Arid Regions through Advanced Simulation Models: A Systematic Review of the Literature

Abstract

Abstract Furrow irrigation is widely practiced in arid and semi-arid regions but suffers from low efficiency due to infiltration variability and water losses. This systematic review analysed 92 peer-reviewed studies (2017–2024) on short-cycle crops, identified through a reproducible search of five bibliographic databases (Scopus, Web of Science, IEEE Xplore, SpringerLink, ScienceDirect) supplemented by a manually directed search of publisher platforms, following Kitchenham & Charters’ protocol. Study quality was assessed narratively, based on validation against field data and comparison with established simulation software (SIRMOD, WinSRFR), rather than through a formal quality-appraisal tool. The narrative synthesis indicates that optimized irrigation techniques achieve efficiency gains ranging from 20% to over 50% relative to conventional practice, depending on crop, soil, and technology, addressing the two operational questions that determine furrow-irrigation performance: when to irrigate and how much water to apply. Advances were identified in coupled surface–subsurface modelling, soil moisture prediction, sensor-driven scheduling, and water–plant–atmosphere interaction. Persistent limitations include scarce field validation beyond single-furrow, single-season conditions, limited generalization across agroclimatic contexts, and interpretability challenges in machine learning (ML) models. Future research should prioritize embedding physical constraints into AI architectures, routine uncertainty quantification, multi-plot and multi-season validation, and pairing technological development with affordable hardware and farmer training to strengthen reproducibility and practical adoption in irrigation science.