Digital Water Quality Monitoring for Pre-Harvest Safety: A Critical Review of IoT, AI and Rapid Sensors in Irrigation Management
Year: 2026
Type: preprint
Abstract: Irrigation water is a recognized pathway for pre-harvest contamination of fresh produce, yet conventional laboratory testing limits monitoring frequency and timeliness of management decisions. This critical review examines how digital technologies - Internet of Things (IoT)-enabled sensors, rapid biosensors and molecular methods, and artificial intelligence (AI) and machine learning - may support irrigation water monitoring and pre-harvest food safety. Literature was purposively selected from Scopus, Web of Science, ScienceDirect, SpringerLink and IEEE Xplore, with an initial scoping from January 2018 to October 2026; the final 29 peer-reviewed studies that met critical relevance criteria clustered between 2020 and 2026, including two 2026 early-online articles, reflecting the recent emergence of this field. The review was complemented by guidance from the Food and Agriculture Organization (FAO), Codex Alimentarius Commission (2023), World Health Organization (WHO) and United States Food and Drug Administration (FDA 2024 Final Rule on Pre-Harvest Agricultural Water). The synthesis shows that IoT systems can increase monitoring frequency for physicochemical parameters and support baseline characterization and anomaly detection, while rapid biosensors and molecular assays offer more direct detection of microbial hazards with reduced time-to-result compared with conventional culture. AI approaches can assist pattern recognition, risk prediction and interpretation of multi-parameter data. However, physicochemical measurements cannot independently confirm microbiological safety, and sensor accuracy, calibration drift, matrix interference, field validation and model transferability remain critical limitations. Integrated digital monitoring may support risk-based water management when combined with validated microbial testing and clear response procedures, but evidence that it directly reduces produce contamination under routine farming conditions remains limited. Recommendations are provided for strengthening field validation and implementation, particularly in resource-limited locations.
Source: Research Square
Author: Ayobami Alli
Language: English
Topic: Smart Agriculture and AI
Subfield: Plant Science
Domain: Life Sciences
Open Access status: green
OpenAlex ID: W7224795260