A bibliometric overview of conversational AI in hydrology and environmental sciences

Abstract

The application of large language models (LLMs) in hydrology and environmental sciences is expanding rapidly, but a comprehensive understanding of their potential, best practices and application areas are not studied extensively. This study conducts a bibliometric analysis of recent scientific literature to evaluate publication trends, citation impact, and the key contributors in LLM-related research within these fields and offers insights and suggestions for best practices. We focus on extracting and analyzing critical metadata, including citations, publication dates, journals, authors, affiliated countries, impact factors, cite scores, research types, domains, and keywords. Additionally, we assess the purpose, use cases, and applications of LLMs, as well as the ethical considerations surrounding cost, scalability, data privacy, transparency, and sustainability. Our findings indicate significant growth in LLM-based studies, especially in hydrological modeling, climate forecasting, and enviC privacy and transparency, that need further exploration. This study provides a detailed understanding for the landscape of LLM adoption in environmental sciences, offering valuable insights for future research and policy development in these critical areas.