Fig. 3 | Scientific Data

Fig. 3

From: A three-decade lake dataset on the Mongolian plateau tracking water area and quality dynamics (1990–2020)

Fig. 3

Workflow for annual lake-area mapping (example: Uvs Lake). A two-step workflow comprising (1) spatial extraction and refinement—classifying yearly water with the JRC Yearly Water Classification History and intersecting with the Maximum Water Extent (MWE), then eliminating small fragments and dissolving polygons to obtain coherent annual lake boundaries; and (2) temporal screening and gap filling—building an area time series to flag abrupt-change years, inspecting and correcting anomalies with historical imagery, and filling missing observations using the GLAD surface-water dataset, resulting in a high-quality spatiotemporal record of annual lake areas for 1990–2020.

Back to article page