Reconstructing winter wheat NDVI time series using Sentinel-1 radar data

D.L. Koshkin; A.V. Panfilova

doi:10.32848/agrar.innov.2026.38.15

D.L. Koshkin Mykolaiv National Agrarian University https://orcid.org/0000-0002-6927-8487
A.V. Panfilova Mykolaiv National Agrarian University https://orcid.org/0000-0003-0006-4090

DOI: https://doi.org/10.32848/agrar.innov.2026.38.15

Keywords: winter wheat, Sentinel‑1, Sentinel‑2, NDVI, SAR, phenologically normalized time, GAM, time series reconstruction

Abstract

Purpose. This study aimed to assess the accuracy of reconstructing winter wheat normalized difference vegetation index (NDVI) time series using Sentinel‑1 radar data and phenologically normalized time under incomplete Sentinel‑2 optical coverage, and to evaluate the robustness of the approach across growing seasons. Methods. The study used data from three experimental plots at the Educational, Scientific and Practical Center of Mykolaiv National Agrarian University over four growing seasons (2021–2025). Phenologically normalized time was used to establish a common temporal scale across seasons. Sentinel‑1 VV and VH backscatter coefficients, the polarization ratio Q, and the Radar Vegetation Index (RVI) were used as predictors. Sentinel‑1 and Sentinel‑2 observations were matched according to their closest phenological timing. Seasonal NDVI dynamics were reconstructed using a Generalized Additive Model (GAM) and evaluated under two scenarios: random withholding of individual observations and block-gap validation simulating a prolonged absence of optical observations caused by persistent cloud cover. Results. The relationship between NDVI and SAR predictors was strongly phase dependent. RVI and Q showed the highest correlations with NDVI during spring regrowth, whereas VV showed the weakest relationship. In both validation scenarios, the model reproduced the main seasonal NDVI trajectory, including autumn canopy development, winter decline, spring regrowth, and maximum values around heading. Reconstruction was more accurate under random withholding, while block-gap validation remained more challenging but preserved the seasonal NDVI dynamics across the continuous optical gap. Findings. Combining phenologically normalized time with Sentinel‑1 SAR predictors within a GAM provides a suitable basis for reconstructing winter wheat NDVI under incomplete Sentinel‑2 coverage. The results demonstrate the potential of Sentinel‑1 and Sentinel‑2 integration to improve the temporal continuity of NDVI time series for winter wheat monitoring. Further research should include additional growing seasons and agroclimatic regions and apply spatially independent validation.

References

1. Vidican R., Mălinaș A., Ranta O., Moldovan C., Marian O., Ghețe A., Ghișe C. R., Popovici F., Cătunescu G. M. Using Remote Sensing Vegetation Indices for the Discrimination and Monitoring of Agricultural Crops: A Critical Review. Agronomy. 2023. Vol. 13, No. 12. P. 3040. DOI: 10.3390/agronomy13123040.
2. Drusch M., Del Bello U., Carlier S., Colin O., Fernandez V., Gascon F., Hoersch B., Isola C., Laberinti P., Martimort P., Meygret A., Spoto F., Sy O., Marchese F., Bargellini P. Sentinel-2: ESA's Optical High-Resolution Mission for GMES Operational Services. Remote Sensing of Environment. 2012. Vol. 120. P. 25–36. DOI: 10.1016/j.rse.2011.11.026.
3. Torres R., Snoeij P., Geudtner D., Bibby D., Davidson M., Attema E., Potin P., Rommen B., Floury N., Brown M., Navas Traver I., Deghaye P., Duesmann B., Rosich B., Miranda N., Bruno C., L'Abbate M., Croci R., Pietropaolo A., Huchler M., Rostan F. GMES Sentinel-1 Mission. Remote Sensing of Environment. 2012. Vol. 120. P. 9–24. DOI: 10.1016/j.rse.2011.05.028.
4. Veloso A., Mermoz S., Bouvet A., Le Toan T., Planells M., Dejoux J.-F., Ceschia E. Understanding the Temporal Behavior of Crops Using Sentinel-1 and Sentinel-2-Like Data for Agricultural Applications. Remote Sensing of Environment. 2017. Vol. 199. P. 415–426. DOI: 10.1016/j.rse.2017.07.015.
5. Nasrallah A., Baghdadi N., El Hajj M., Darwish T., Belhouchette H., Faour G., Darwich S., Mhawej M. Sentinel-1 Data for Winter Wheat Phenology Monitoring and Mapping. Remote Sensing. 2019. Vol. 11, No. 19. P. 2228. DOI: 10.3390/rs11192228.
6. Mercier A., Betbeder J., Baudry J., Le Roux V., Spicher F., Lacoux J., Roger D., Hubert-Moy L. Evaluation of Sentinel-1 & 2 Time Series for Predicting Wheat and Rapeseed Phenological Stages. ISPRS Journal of Photogrammetry and Remote Sensing. 2020. Vol. 163. P. 231–256. DOI: 10.1016/j.isprsjprs.2020.03.009.
7. Szigarski C., Jagdhuber T., Baur M., Thiel C., Parrens M., Wigneron J.-P., Piles M., Entekhabi D. Analysis of the Radar Vegetation Index and Potential Improvements. Remote Sensing. 2018. Vol. 10, No. 11. P. 1776. DOI: 10.3390/rs10111776.
8. Wood S. N. Generalized Additive Models: An Introduction with R. 2nd ed. Boca Raton: Chapman and Hall/CRC, 2017. 496 p. DOI: 10.1201/ 9781315370279.
9. Hird J. N., McDermid G. J. Noise Reduction of NDVI Time Series: An Empirical Comparison of Selected Techniques. Remote Sensing of Environment. 2009. Vol. 113, No. 1. P. 248–258. DOI: 10.1016/j.rse.2008.09.003.
10. Pipia L., Muñoz-Marí J., Amin E., Belda S., Camps-Valls G., Verrelst J. Fusing Optical and SAR Time Series for LAI Gap Filling with Multioutput Gaussian Processes. Remote Sensing of Environment. 2019. Vol. 235. Article 111452. DOI: 10.1016/j.rse.2019.111452.
11. Pedersen E. J., Miller D. L., Simpson G. L., Ross N. Hierarchical Generalized Additive Models in Ecology: An Introduction with mgcv. PeerJ. 2019. Vol. 7. Article e6876. DOI: 10.7717/peerj.6876.