SAR Backscatter, not Optical Indices or Phenology, Drives Cropland Classification Accuracy: A Multi-Sensor Feature Ablation Study in the Abuja Municipal Area Council, Nigeria

Cropland mapping in rapidly urbanizing tropical regions increasingly relies on multi-sensor Earth observation, yet the individual contribution of different sensor and feature types to classification accuracy is rarely quantified explicitly. Building on a prior comparative study of Random Forest (RF) and Gradient Tree Boosting classifiers using a 20-band Sentinel-1/Sentinel-2 feature stack over the Abuja Municipal Area Council (AMAC), Nigeria, this study conducts a systematic feature ablation to isolate the marginal contribution of spectral indices, NDVI-derived phenological metrics, and Synthetic Aperture Radar (SAR) backscatter to cropland classification accuracy. Six feature-stack combinations, ranging from optical-only to the full fused stack, were classified using an RF classifier with fixed hyperparameters and the same stratified reference dataset (373 validation points) used in the original study. Overall accuracy, Kappa coefficient, and cropland-specific precision, recall, and F1-score were compared across combinations, and McNemar's test was used to assess whether accuracy differences between stacks were statistically significant. Results show that SAR backscatter, not spectral indices or phenological metrics, was the dominant driver of accuracy gains: the optical-plus-SAR combination achieved the highest overall accuracy (74.8%) and Kappa (0.685) of all six combinations tested, exceeding even the full fused stack (73.5%, κ=0.668), while adding spectral indices and phenological metrics without SAR reduced accuracy below the optical-only baseline (67.0% vs. 68.9%). McNemar's tests confirmed the full stack was not significantly different from the optical-only baseline or the optical-plus-SAR combination (p > 0.05), but was significantly more accurate than any combination lacking SAR (p < 0.05). Gini-based variable importance ranked VH backscatter and VH temporal variance as the two most important predictors among all 20 bands, and this ranking was corroborated by a permutation-importance cross-check on an independently trained model, which agreed with Gini importance on both top-ranked SAR bands despite the two methods diverging further down the ranking  indicating SAR's dominance is a robust finding rather than an artifact of Gini importance's known bias toward continuous variables. These findings indicate that, in this tropical savanna setting, radar data delivers the largest return on investment for operational cropland monitoring, while spectral indices and phenological metrics contribute comparatively little value on their own.

Feature Ablation, Random Forest, Sentinel-1, Sentinel-2, SAR-Optical Fusion, Variable Importance, Cropland Mapping

1. Azadi, H., Taheri, F., Burkart, S., Mahmoudi, H., De Maeyer, P., & Witlox, F. (2021). Impact of agricultural land conversion on climate change: H. Azadi et al. Environment, Development and Sustainability, 23(3), 3187-3198.. [Google Scholar] [Crossref]

2. Belgiu, M., & Drăguţ, L. (2016). Random forest in remote sensing: A review of applications and future directions. ISPRS journal of photogrammetry and remote sensing, 114, 24-31.Breiman, L. (2001). Random forests. Machine Learning, 45(1), 5-32. [Google Scholar] [Crossref]

3. Du, J., Watts, J. D., Jiang, L., Lu, H., Cheng, X., Duguay, C., & Tarolli, P. (2019). Remote sensing of environmental changes in cold regions: Methods, achievements and challenges. Remote Sensing, 11(16), 1952. [Google Scholar] [Crossref]

4. Dubey, A., Singh, P. K., Singh, S., Manzoor, U., Sahoo, C., & Saini, A. (2025). Impact of Climate Change and Land Degradation on Agriculture. In Eco-Resilience: Climate Change, Land Degradation and Sustainable Solutions (pp. 49-95). Cham: Springer Nature Switzerland. [Google Scholar] [Crossref]

5. FCTA/ACReSAL. (2025). FCT: Agro-Climatic Resilience in Semi-Arid Landscapes. Federal Capital Territory Administration. Retrieved from https://acresal.gov.ng/states/fct/ [Google Scholar] [Crossref]

6. Foody, G. M. (2004). Thematic map comparison. Photogrammetric Engineering & Remote Sensing, 70(5), 627-633. [Google Scholar] [Crossref]

7. Gorelick, N., Hancher, M., Dixon, M., Ilyushchenko, S., Thau, D., & Moore, R. (2017). Google Earth Engine: Planetary-scale geospatial analysis for everyone. Remote sensing of Environment, 202, 18-27. [Google Scholar] [Crossref]

8. Ibrahim, I., Khalid, S. S., Musa, H. H., & Ahmed, N. A. (2026). A comparative analysis of Random Forest and Gradient Tree Boosting for cropland mapping using multi-sensor Sentinel data: A case study of Abuja Municipal Area Council, Nigeria. International Journal of Latest Technology in Engineering, Management & Applied Science (IJLTEMAS), 15(4), 156–164. https://doi.org/10.51583/IJLTEMAS.2026.150400014 [Google Scholar] [Crossref]

9. Li, Y., & Xiao, X. (2025). Deep learning-based fusion of optical, radar, and LiDAR data for advancing land monitoring. Sensors, 25(16), 4991. [Google Scholar] [Crossref]

10. Van Tricht, K., Gobin, A., Gilliams, S., & Piccard, I. (2018). Synergistic use of radar Sentinel-1 and optical Sentinel-2 imagery for crop mapping: A case study for Belgium. Remote Sensing, 10(10), 1642. [Google Scholar] [Crossref]

• Enhancing Formation Control of Multi Agent Systems Using Ann Based Technique
• Improving Sliding Mode Control with Chattering Reduction using Fuzzy Based Technique
• Cooking Quality, Fasting Blood Glucose, Glycemic Index and Load of High–Fiber Noodles Made from Wheat, Tiger Nut Residue and Cassava Flour Blends
• Matrix Rhythm Therapy Versus Interferential Therapy Combined with Lumbar Stabilization Exercises in Chronic Non-Specific Low Back Pain: A Randomized Comparative Trial
• Formulation and Sensory Evaluation of Functional Cake Prepared from Sweet Potato Powder