Artificial neural modeling for precision agricultural water management practices

Recent advances in neural computing and numerical modeling have given rise to artificial intelligence (AI) in various fields of research. Machines with AI capabilities in a broader scope can perform tasks mimicking human brains. In this book chapter, a practical guide is provided to use AI-based algorithms in hydrological studies. Firstly, common hydrological modeling problems have been identified based on their modeling nature and categories such as regression, classification, sequence prediction, and reinforcement learning tasks. An introduction of the commonly used machine learning (ML) algorithms and libraries are discussed in the following sections to familiarize readers with the recent advancement in this specific domain. Important steps from training to implementation of ML algorithms in hydrological studies have also been discussed. A detailed review of common hydrological domains such as rainfall-runoff modeling, groundwater modeling, evapotranspiration modeling, and water resource management is summarized accordingly. In the final section of this book, some challenges implementing ML algorithms in hydrological-related studies are discussed. The chapter also presents a need to establish some ethical rules and responsible use of AI in water-related studies. The applications of ethical principles will help the water resource managers, policymakers, and government sectors to layout the framework for responsible use of AI in the water sector.