The 2022 global annual maximum land surface water cover product, with a spatial resolution of 30 meters, was generated from Landsat satellite data using a prior-knowledge-based rapid extraction algorithm for land surface water. The input data used to produce GAMLSW include Landsat data and Shuttle Radar Topography Mission (SRTM) Digital Elevation Model (DEM) data. Global Multi-resolution Terrain Elevation Data 2010 (GMTED2010) was used in high-latitude areas due to lack of SRTM-DEM data. Firstly, an annual Landsat top-of-atmosphere reflectance dataset was created and cloud masking was applied. Feature indices such as the NDVI, NDWI, and NDSI were calculated, while terrain indices were obtained from the DEM. Subsequently, a set of land surface water extraction rules based on multiple indices thresholds were established using prior knowledge (Liu et al., 2021). Finally, the established land surface water extraction rules were applied to multi-temporal Landsat data of the same region within the annual cycle to extract land surface water information on a per-pixel basis, and the extracted results were synthesized via a spatial logical OR operation to generate the annual maximum land surface water cover product.