Bias Correction of Sentinel-2 MSI Vegetation Indices in a Desert Steppe with Original Assembled Field Online Multi-Angle Spectrometers

In desert steppe regions with sparse vegetation, there are discrepancies between vertical and oblique observations made by satellite-based sensors. this study, we developed deployed an online multiangle spectrometer in the area of Inner Mongolia, China, to calibrate vegetation indices. One key components device is a specially designed quarter-arc iron track that holds fixed view angles 30°, 45°, 60°, 75° 90° fixed. These observation positions equipped high-efficiency multichannel sensors can capture reflectance ground objects at visible near-infrared wavelengths. Real-time experiments were conducted multiple error-based angle correction model was constructed reconcile differences angular observations. The calculated results subsequently applied bias-correction process Sentinel-2 index. Across all angles, daily distributions Normalized Difference Vegetation Index (NDVI) Ratio (RVI) exhibited U-shaped pattern nadir occurring noon. Among data, RVI demonstrated superior overall stability compared NDVI. However, growth, NDVI showed less sensitivity, resulting decrease its coefficient variation (CV) from 32.3% 18.2%. To correct bias products, initially path length (PLC) eliminate topographic influence on Band4 Band8 band reflectance. Our findings revealed performed better than mitigating effects topography, as evidenced determination 0.448 0.09. Additionally, corrected error constructing (R 2 =0.84, RMSE=0.03). images significant characteristics areas relatively large angles. study provide valuable scientific support for satellite image products accurate detection through remote sensing low coverage.