Scopus bibliometric review maps 2015–2025 growth in AI research for sustainable water management

A bibliometric review of 3,280 Scopus-indexed research articles published from 2015 through 2025 found rapid growth in artificial intelligence research related to sustainable water management, with publication and citation trends increasing sharply over the period. The review mapped the field’s research themes, methods, journals, institutions, authors, funders and international collaborations; it did not directly evaluate the performance of individual AI-enabled water-management systems.

The researchers searched Scopus on March 4, 2026, with a structured query combining “artificial intelligence,” “AI,” “machine learning” and “deep learning” with terms covering sustainability, water management, water conservation, water use, water supply and hydrology. The initial search returned 5,215 records. Following title-and-abstract screening, 1,935 entries were excluded, leaving 3,280 articles for analysis. The 2015–2025 window was intended to capture a period of rapid development in AI, big-data analytics, cloud computing and Internet of Things-enabled environmental monitoring, while ending with the most recent complete publication year available at the time of data collection.

Only English-language research articles were eligible. Reviews, book chapters, short surveys, notes, books and retracted papers were excluded. The authors acknowledged that restricting the language and publication format could omit influential work, while dependence on the chosen search terms could miss relevant studies that used different terminology. The review was also limited to Scopus and did not search sources such as Web of Science or Google Scholar.

VOSviewer version 1.6.20 was used to construct bibliometric maps and visualize annual publication and citation trends, co-authorship, keyword co-occurrence and co-citation relationships. Publication and citation indicators, together with h-index and g-index measures, were used to assess research output and citation impact. These measures do not independently establish scientific quality, practical effectiveness or whether AI systems outperform conventional water-management approaches.

The 3,280 articles were distributed across 26 subject categories. Environmental science accounted for 36% of the corpus, while Earth and Planetary Sciences, engineering and Agricultural and Biological Sciences each represented 9%; computer science represented 6%. The articles appeared in 710 journals, led by Science of the Total Environment with 207 publications, followed by Water with 154 and the Journal of Environmental Management with 151.

China was the largest national contributor, with 1,033 publications, or 31.4% of the corpus. The United States followed with 560 publications, representing 17.0%; India had 460, or 14.0%; Iran had 296, or 9.0%; and Saudi Arabia had 194, or 5.91%. The publications came from 134 countries and regions, and the review’s collaboration analysis associated Chinese researchers with more than 45% of the field’s co-authored output. The discussion links China’s publication lead to investment in AI and environmental monitoring, national research programs, research infrastructure and collaboration among universities, research institutes and industry.

Government and institutional funding was also prominent. The National Natural Science Foundation of China was associated with 531 publications, followed by China’s National Key Research and Development Program with 199. The U.S. National Science Foundation funded 80 articles and the European Commission 53, while the Fundamental Research Funds of Central Universities, Chinese Academy of Sciences and National Research Foundation of Korea were associated with 55, 52 and 39 publications, respectively. The U.S. Geological Survey was associated with 37 publications, the China Postdoctoral Science Foundation with 36, and Brazil’s Conselho Nacional de Desenvolvimento Científico e Tecnológico with 35.

Among 7,671 contributing research centers, the University of Chinese Academy of Sciences in Beijing led with 56 articles, followed by Beijing Normal University with 33. The corpus included 14,348 authors. Zaher Mundher Yaseen was the most prolific with 21 articles, followed by Sani Isah Abba and Ahmed Elbeltagi with 18 each and Ozgur Kisi with 15. A total of 139 authors had co-authored at least five papers, and these authors were distributed across 11 collaboration clusters.

Keyword analysis identified 22,646 distinct terms. Machine learning, water management, water supply and water quality were among the most frequent, while artificial intelligence, environmental monitoring and prediction also appeared prominently. The main themes included water-quality prediction, hydrological modelling, environmental monitoring and decision-support systems. Co-citation analysis identified random forests, neural networks and boosting algorithms as foundational machine-learning methods. Applications described in the reviewed literature included forecasting pollutants in surface and groundwater, modelling rainfall-runoff, evapotranspiration, groundwater levels and river flow, and analyzing wastewater-treatment operations for aeration control and energy use. These were reported applications in the literature, not systems directly tested by the bibliometric review.

The authors identified hybrid models combining machine learning with physics-based approaches, explainable artificial intelligence, integration with Internet of Things sensors, cloud computing and digital twins, and improved data quality and standardization as future research directions. They also noted that many water-management studies use small datasets, which may limit the applicability of machine-learning models. Because the review depended on Scopus coverage, English-language research articles and selected search terms, its maps describe a bounded bibliometric corpus rather than the complete global evidence base.

Subject of Research: Bibliometric analysis of artificial intelligence research for sustainable water management

Article Title: Mapping the knowledge structure and evolution of artificial intelligence for water reuse and sustainable water management

Article References: Kumar, N., Nagar, M. R., Dharsandia, I., Tiwari, A., Mishra, S. K., & Kumar, M. (2026). Mapping the knowledge structure and evolution of artificial intelligence for water reuse and sustainable water management. Discover Artificial Intelligence, 6(1), Article 1442. https://doi.org/10.1007/s44163-026-02389-4

Image Credits: AI Generated

DOI: 10.1007/s44163-026-02389-4

Keywords: artificial intelligence, sustainable water management, bibliometric analysis, machine learning, water quality prediction, hydrological modelling, wastewater treatment, research collaboration, Scopus, VOSviewer, Mapping, knowledge