Artificial intelligence applications in hydrological studies and ecological restoration of watersheds: A systematic review

Abstract

Water resources management is fundamental to the sustainability of river basins. Water quality is affected by pollution caused by human activities. In this context, the restoration of degraded watersheds helps soil recovery, sustainable water management, reforestation, biodiversity conservation and mitigation of human impacts. Artificial intelligence (AI) innovates data management and analysis processes by optimising decision-making and data analysis in hydrological studies and ecological restoration. This research aims to analyse scientific information related to the integration of AI in studies on hydrogeology and ecological restoration of watersheds by analysing scientific databases for knowledge of the intellectual structure, lines and trends of research. The methodology includes three phases: i) search criteria and data processing (Scopus-Web of Science); ii) analysis of the intellectual and conceptual structure; and iii) application of the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) method. The results indicate that there is a total of 171 records, with a 4.49% growth in scientific production in the last four years, focusing on artificial neural networks (10.53%), artificial intelligence (3.51%), genetic algorithms (1.17%) and machine learning (1.17%). This increase is due to the climatic variation generated in recent years, driven by anthropogenic pressures, especially in the agricultural sector due to the high demand for fertiliser and pesticide pollution. This problem has prompted the search for more far-reaching environmental management technologies, making it a potential niche for study. China (72.51%) and the United States (25.73%) are the most outstanding contributors to production in this area. On the other hand, there is less research in this area in developing countries such as South Africa (2.92%), Colombia (1.17%), and Argentina (0.58%), among others. This analysis identifies opportunities and challenges in applying AI for water resource optimisation and water quality prediction, providing an innovative conceptual framework for sustainable watershed management.

Keywords

• Hydrogeology
• Bibliometrics
• Neural network
• Genetic algorithm
• Machine learning