Synergistic Use of Optical and SAR Imagery for Near Real-Time Green Area Index Retrieval in Maize

The green area index (GAI) is a key biophysical variable for crop monitoring. Its retrieval often relies on optical remote sensing, which can be hampered by frequent cloud cover. In this context, synthetic aperture radars (SARs) offer the advantage of being able to provide dense time series that can be used to complement sparse GAI time series retrieved from optical data. In this study, a transformer encoder is implemented to perform synergistic SAR-optical GAI retrieval in maize fields, in near real-time, from past and current values of SAR backscatter coefficient and interferometric coherence, as well as past values of the GAI when available. Sentinel-1 and Sentinel-2 data acquired from 2018 to 2021 over the Hesbaye region of Belgium are used for cross-validation. The results show that the model can successfully retrieve the GAI at the parcel level on an unseen growing season with a mean R2 of 0.88 and RMSE of 0.71. The performance of the model is further assessed through external validation using in situ data collected from 10 maize fields in Belgium in 2018. These promising results pave the way for the generation of accurate, dense GAI time series throughout the crop growing season, ultimately improving the capability of operational crop monitoring systems in cloud-prone regions for which the timely delivery of information can be crucial.

Affiliations

UCLouvainSST/ELI/ELIE - Environmental Sciences

Citations

APA
Chicago
FWB

Bouchat, J., Deffense, Q., De Maet, T., & Defourny, P. (2025). Synergistic Use of Optical and SAR Imagery for Near Real-Time Green Area Index Retrieval in Maize. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 18, 28515-28530. https://doi.org/10.1109/jstars.2025.3622750 (Original work published 2025)