Answers are grounded in this paper and cite their sources.
Normalized difference vegetation index (NDVI) time series provide insights into vegetation dynamics, though its continuity can be disrupted by cloud cover. Synthetic aperture radar (SAR) data can support NDVI modeling; however, the relationship between NDVI and SAR-derived features in agriculture remains insufficiently explored. This study evaluates the potential of Sentinel-1 intensity and polarimetric features for modelling NDVI time series in four crops: wheat, barley, corn and sunflower. Temporal features were derived to describe SAR dynamics. Three regression scenarios were compared using different features: intensity, polarimetric and a combination of both. The combined scenario outperformed others, though the intensity-only achieved close results. The polarimetric scenario performed well for winter cereals. Incorporating temporal information significantly improved NDVI predictions. Irrigated barley, wheat and sunflower showed strong performance across all scenarios (R2 > 0.84), though sunflower underperformed in the polarimetric scenario. The VH/VV ratio, its temporal derivative and VV backscatter were key features in winter crops and corn. Key features for sunflower were VH and VHxVV. NDVI modelling was most accurate in the germination and vegetative growth stages. This study highlights the potential of S-1 intensity features, particularly when temporal dynamics are considered, for enhancing crop-specific NDVI time series modelling.
No takes yet. Share an insight, caveat, or question.
Aranguren et al. (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: