Cross-sensor cropland mapping: evaluating Chinese satellites and Sentinel-2 in the Nile Delta

Expanding the global application of Chinese Earth Observation satellites (CEOSs) data can enhance Earth observation capabilities, especially in data-scarce regions. This study mapped cropland in Egypt's Nile River Delta using datasets from CEOSs (Gaofen-1 B/C/D, Gaofen-2, Gaofen-7, and Ziyuan-3) and compared the results with those from Sentinel-2 under similar spatial and temporal conditions. Results showed that: (1) High-resolution CEOS data showed advantages in delineating field boundaries and capturing texture features, while Sentinel-2 is more effective for monitoring vegetation dynamics and ensuring spectral consistency due to its spectral characteristics; (2) While CEOS achieved higher overall accuracy ([Formula: see text]), Sentinel-2 performed relatively better in vegetation index-based and non-cropland classifications ([Formula: see text]), likely due to differences in spatial resolution and sensor characteristics; (3) CEOS data tend to perform better in fragmented croplands, whereas Sentinel-2 is advantageous for broad, spectrally uniform areas. Addressing inter-sensor spectral differences is crucial to improving classification reliability. These findings demonstrate the potential of CEOSs for global cropland monitoring and their value in supporting the assessment of Sustainable Development Goals (SDGs).