Mapping rainfed cereals and olive groves using Sentinel-1, Sentinel-2, and topographic data in northern Morocco
Résumé
Accurate discrimination of agricultural crop categories in heterogeneous, rainfed landscapes remains challenging because crop spectral responses are strongly influenced by environmental variability and complex terrain. This study assessed the contribution of optical (Sentinel-2), radar (Sentinel-1), and topographic variables to the binary discrimination of annual rainfed cereals and perennial olive groves in the Tangier–Tétouan–Al Hoceima region of northern Morocco, a Mediterranean mountain environment characterized by fragmented agricultural systems and rainfed conditions. Using a Random Forest classifier, 38 candidate feature sets were evaluated, representing different combinations of Sentinel-2 spectral bands and vegetation indices, Sentinel-1 radar features, phenological information, and topographic variables to discriminate cereals and olive groves. The results showed that Sentinel-2 variables provided the strongest discriminative power, while Sentinel-1 variables alone yielded comparatively lower performance. The best-performing scenario combined Sentinel-2 spectral bands, Sentinel-1 VV and VH backscatter, Normalized Difference Red Edge Index 2 (NDRE2), Sentinel-1 VV/VH backscatter ratio (RATIO), and topographic variables (elevation, slope, and aspect), reaching an overall accuracy of 0.96 with balanced F1-scores for cereals and olives (0.92 and 0.91, respectively). Sentinel-2-only configurations already achieved high accuracies, particularly when spectral bands were combined with red-edge indices and topographic variables, whereas the addition of Synthetic Aperture Radar (SAR) data produced modest but consistent improvements in classification accuracy. SHAP-based interpretation revealed a hierarchical contribution of predictors, with elevation and slope emerging as the most influential variables, followed by NDRE2 and key spectral bands. These findings indicate that the discrimination between rainfed cereals and olive groves in this rainfed landscape is strongly influenced by topographic and environmental gradients, while spectral variables capture vegetation biophysical properties and SAR contributes complementary structural and moisture-related information. Overall, the study demonstrates that accurate mapping of rainfed cereals and olive groves in heterogeneous Mediterranean agro-ecosystems benefits from the integration of spectral, structural, and topographic information. The proposed framework provides a useful foundation for extending crop mapping and agricultural monitoring approaches in data-scarce and climate-sensitive regions.
Citer ce document
Exporter : BibTeX · RIS (Zotero, Mendeley, EndNote)
Accès au document
Texte intégral archivé sur SPHAERO, en accès ouvert : lisible gratuitement en ligne après connexion. Le document reste aussi disponible à sa source. Se connecter
Voir l'article sur le site de la revueLicence et provenance
Licence : CC BY
Notice moissonnée depuis OpenAlex le 19/10/2026. Le document reste hébergé par sa source.
Voir le document à la source →