The 10 m artificial aquaculture ponds dataset in China from 2015 to 2025

This dataset provides a national-scale, 10 m resolution spatial dataset of artificial aquaculture ponds (AAPs) in China for the years 2015, 2020, and 2025. The dataset was generated using Sentinel-2 imagery and a deep learning–based semantic segmentation framework, aiming to support large-scale analysis of aquaculture dynamics and related environmental studies. Technical notes: 1. All files are provided in Shapefile format and represent artificial aquaculture ponds extracted at a spatial resolution of 10 m.2. All vector layers preserve complete georeferencing information and can be directly used in standard GIS and remote sensing software.3. The coordinate reference system is consistent across all years to facilitate inter-annual comparison.4. Visual inspection and manual correction were conducted to ensure the quality and reliability of the final dataset.5. Annual datasets for the period 2015–2025 will be progressively updated and released in future versions.