AI Education for Beginners in the Water Resources Field

About This Course

Artificial intelligence is rapidly becoming part of everyday practice in hydrology and water management — from forecasting river water levels to detecting leaks in distribution networks. Yet surveys conducted by the UNESCO IHP National Committee of the Republic of Korea found that stakeholders' understanding of AI in the water sector remains relatively low, while more than 80% of respondents expressed a clear need for introductory, water-specific AI learning materials. This course was created to answer that need.

The course is developed under Sub-theme 1.9 of UNESCO IHP Phase IX — "Development and Sharing of New Technologies Using AI and the Internet of Things (IoT) for the Capacity Building of Water Resource Stakeholders" — for which the Republic of Korea serves as the lead country. It is designed to lower the entry barrier to AI for newcomers to the water field: undergraduate and graduate students, early-career researchers, and practitioners in government agencies, water utilities and consultancies who have little or no background in machine learning or programming.

Part of the "AI in the Water Resources Field" series, the curriculum is organized into three complementary tracks. Theoretical Learning explains what AI is, why it matters for water resources, and how the main model families used in the sector (ANN, CNN, RNN) differ from one another, together with recent research trends. Practical Learning guides learners through a complete modelling pipeline — data acquisition, Python set-up, coding, model training, and interpretation of results — using real observed data: river water level prediction with an ANN, and leak detection in a water distribution network with a CNN. Informatic Learning presents interviews with AI experts working in the field, who share practical opportunities, limitations, and advice for beginners.

All course materials — lecture notes, narration and subtitles — are delivered entirely in English, and each video is divided into short segments so that learners anywhere can progress at their own pace. No prior AI experience is required. The course is also linked to UNESCO's IHP Water Information Network System (IHP-WINS): upcoming modules apply time-series models (RNN/LSTM) to the openly published water quality record of the Taehwa River Ecohydrology Demonstration Site in the Republic of Korea, available at IHP-WINS. Learners are encouraged to download the dataset and experiment with it directly.