Evaluation of sentinel-1 and sentinel-2 feature sets for delineating agricultural fields in heterogeneous landscapes

The Group on Earth Observations Global Agricultural Monitoring Initiative (GEOGLAM) considers agricultural elds as one of the essential variables that can be derived from satellite data.We evaluated the accuracy at which agricultural elds can be delineated from Sentinel-1 (S1) and Sentinel-2 (S2) images in different agricultural landscapes throughout the growing season.We used supervised segmentation based on the multiresolution segmentation (MRS) algorithm to rst identify the optimal feature set from S1 and S2 images for eld delineation. Based on this optimal feature set, we analyzed the segmentation accuracy of the elds delineated with increasing data availability between March and October of 2018. From the S1 feature sets, the combination of the two polarizations and two radar indices attained the best segmentation results. For S2, the best results were achieved using a combination of all bands (coastal aerosol, water vapor, and cirrus bands were excluded) and six spectral indices. Combining the radar and spectral indices further improved the results. Compared to the single-period dataset in March, using the dataset covering the whole season led to a signicant increase in the segmentation accuracy. For very small elds (< 0.5 ha), the segmentation accuracy obtained was 27.02%, for small elds (0.5 1.5 ha), the accuracy was 57.65%, for medium elds (1.5 ha 15 ha), the accuracy was 75.71%, and for large elds (> 15 ha), the accuracy stood at 68.31%. As a use case, the segmentation result was used to aggregate and improve a pixel-based crop type map in Lower Saxony, Germany.