<p>Recent advances in satellite technology and cloud computing have enabled global-scale monitoring of long-term surface water changes. The dynamic nature of surface water, driven by seasonal fluctuations and climatic events, presents challenges for accurately interpreting these dynamics. Here, we introduce the first global dataset that identifies the timing, at annual resolution, of surface water advance or recession from 1984 to 2022. Our approach focuses on identifying persistent changes in surface water features by filtering out seasonal or shorter-term fluctuations. Using a novel algorithm, we mapped the timing of surface water transitions globally, including rivers, lakes, reservoirs, flooded agriculture, and coastal regions. In the dataset each 30 m × 30 m pixel records whether water advance or recession occurred and specifies the year of transition. This dataset enables users to visualize the location, type, and magnitude of changes, while its focus on timing provides new insights into the drivers of water dynamics. Designed for accessibility, the dataset supports scientific research as well as NGOs, policymakers, and water managers in addressing surface water-related challenges.</p>