Building an Intelligent Data Exploring Assistant for Geoscientists

• Published in: Journal of geophysical research. Machine learning and computation Vol. 2; no. 3
• Main Authors: Widlansky, Matthew J., Komar, Nemanja
• Format: Journal Article
• Language: English
• Published: 01.09.2025
• DOI: 10.1029/2025JH000649
• ISSN: 2993-5210

Advances in natural‐language processing and large language models (LLMs) are transforming how geoscientists interact with complex data sets, enabling efficient and intuitive scientific analyses. This study introduces the Intelligent Data Exploring Assistant (IDEA), a prototype software framework that integrates existing LLM technology with domain‐specific instructions, data, analytical tools, and computing resources to support geoscientific research. We demonstrate its application through the Station Explorer Assistant (SEA), a web‐based tool designed for sea level scientists. SEA empowers users to analyze and interpret coastal water level data by addressing challenges such as vertical datum conversions and assessing flooding risks. We also demonstrate the generalizability of building an IDEA, whereby we deploy a local instance of the framework to analyze atmospheric observations from Mars collected by NASA's InSight Mission. By combining LLM capabilities with robust domain‐specific customizations, SEA and the Mars IDEA generate accurate analyses, visualizations, and insights through natural‐language prompts. This study highlights the potential of IDEA frameworks to lower technical barriers, enhance educational opportunities, and transform geoscientific workflows while addressing the limitations and uncertainties of current LLM technology. Artificial intelligence (AI) is transforming how scientists explore and understand our world. At the University of Hawaiʻi Sea Level Center (UHSLC), we are developing tools that use large language models, like what ChatGPT uses, to help scientists study sea level changes. One such tool, called the Station Explorer Assistant (SEA), allows researchers to ask questions in everyday language and receive clear explanations and data analyses in response. SEA uses AI to analyze sea level data, compare water levels to normal conditions, and predict potential flooding, drawing on the UHSLC's extensive database. It even writes and runs its own analysis software, which it shows the user to check that its results are accurate. By making sea level science easier to understand and access, SEA can support communities adapting to rising seas and other coastal challenges. SEA technology is generalizable across geoscience domains through a framework we call an Intelligent Data Exploring Assistant (IDEA), which we demonstrate by asking it to analyze wind observations from Mars. Our work highlights how AI can enhance scientific research and communication, and we envision similar tools being created to support scientists in many fields. Large language models can assist geoscientists by generating data analyses and visualizations from natural‐language prompts A general‐purpose Intelligent Data Exploring Assistant shows the potential of artificial intelligence to enhance geoscience research The Station Explorer Assistant analyzes water level data from tide gauges providing insights into sea level variability and risks