FVC and Land-Cover Dataset of the Aiximan Lake Region, Aksu Prefecture, Xinjiang (2000–2025)

Training samples were derived from field survey points and manually interpreted samples based on high-resolution remote sensing imagery, thereby improving the regional adaptability and discrimination of different land-cover classes. Independent validation samples yielded an overall classification accuracy of 91% and a Kappa coefficient of 0.88, indicating reliable classification performance.The annual FVC dataset covers 2000–2025 and provides continuous year-by-year information on vegetation cover across the study area.

FVC was derived from Landsat imagery using the normalized difference vegetation index (NDVI) and the pixel dichotomy model. Landsat 7 ETM+ imagery was used for 2000–2012, while Landsat 8 OLI imagery was used for 2013–2025. Based on the calculated FVC values, vegetation cover was classified into five levels: extremely low, low, medium, medium-high, and high.

The annual FVC classification data provide a continuous spatial representation of interannual vegetation-cover dynamics and complement the multi-temporal land-cover classification data.By integrating multi-temporal land-cover maps with continuous annual FVC classification data, this dataset provides a comprehensive basis for investigating long-term land-cover change and vegetation dynamics in the Aiximan Lake Area. It can support research on land-use/land-cover change, vegetation dynamics, ecological restoration assessment, ecological risk assessment, ecosystem service evaluation, and ecological and environmental evolution in the Aiximan Lake Area and surrounding watersheds.