2026-10-11T19:13:56Z https://www.scidb.cn/oai

10.57760/sciencedb.36636 2026-05-11T16:13:03Z

2026-05-11 10m bi-annual spatial distribution dataset of China's tidal flats (2016-2023) doi:10.57760/sciencedb.36636 en Tidal flats are highly valuable coastal ecosystems. However, large-scale, high-precision, and long-term remote sensing mapping of tidal flats faces tremendous challenges due to periodic tidal inundation and frequent cloud cover. This dataset provides a high-precision (10-m) bi-annual spatial distribution map of China's tidal flats from 2016 to 2023. Generated on the Google Earth Engine (GEE) cloud platform, the dataset utilizes high-frequency Sentinel-2 time-series imagery covering the coastal zone of China. Methodologically, a novel tide-and-phenology-aware time-series feature reconstruction framework was applied. By employing the Harmonic Analysis of Time Series (HANTS) algorithm to precisely fit the temporal trajectories of MNDWI and NDVI, the severe interferences from instantaneous highly turbid waters and saltmarsh vegetation phenology were successfully decoupled. The extracted robust time-series spectral fingerprints were then fed into a Random Forest classifier for pixel-wise identification.The dataset covers four bi-annual periods (2016-2017, 2018-2019, 2020-2021, and 2022-2023). Extensive evaluations based on large-scale independent validation samples (401,829 points) demonstrate that the dataset achieves an Overall Accuracy of 99.00% and an F1 score of 0.97, exhibiting exceptional mapping robustness in extremely turbid estuaries and complex mosaic habitats. Overcoming the omission defects of traditional static compositing methods in extracting low-tide flats, this dataset provides fundamental and highly reliable spatial data support for coastal ecological protection and restoration, natural resource inventories, nearshore hydrodynamic modeling, and the assessment of Sustainable Development Goals (SDGs). tidal flats; Google Earth Engine; bi-annual spatial distribution map; high resolution; China Gang Yang Chunchen Shao Weiwei Sun Lihua Wang Zhaoliang Song PUBLIC https://creativecommons.org/licenses/by-nc-sa/4.0/ dataset Science Data Bank