The data for the two scans in the figures are very similar. Even though there is a significant amount of variation in the 1-degree FOV data for the sunflowers, the variation is very similar for the two scans. This indicates that the variation is real and due to the sunflower canopy.
The correlation of two scans of the 1-degree FOV data for one of the bands was around 0.9. This analysis was done by matching up the nearest view zenith angles for the two scans.
The data in the figures represent all of the data collected on July 25 for Replication 1 of Set 1 with a relative view angle of 0 degrees.
The data also indicate that the FOV's for the band triplets are 'close', otherwise the variations in the data would not track very well. A key question though is, are the 1-degree FOV's close enough for an accurate determination of the polarizing properties of the canopy for 1-degree FOV data. The figures illustrate that there are significant changes in the reflectance factor of the 1-degree FOV data in some cases across just 2 degrees. If the alignment of the FOV's of the triplet bands are off by a few tenths of a degree in these cases, the resulting calculation of percent polarization could off by several percent of value.
The plots of the 1-degree FOV sky data also indicate that there is very little 'noise' in these data because the measurements are very smooth. The sky represents a very uniform scene - at least when there are no clouds.
The figures represent the data collected for the red and near infrared bands on July 25 for Replication 1 of Set 1 for a relative view angle of 0 degrees (sunflower canopy) and Replication 2 for a relative view angle of 90 degrees (sky).