In two studies Julien Michel (#CNES & #CESBIO) showed:
▶️ that data fusion methods can be used to merge #Sentinel2 with #SentinelHR, providing time series with 2 m resolution and 5 days revisit with rmse errors below 0.01 in reflectance even for pixels with a high gradient

Details: https://labo.obs-mip.fr/multitemp/test-of-spatio-temporal-fusion-of-sentinel-2-new-generation-data-with-a-potential-vhr-mission/

▶️ that these methods still provide good results when the #SentinelHR revisit is reduced from 12 to 30 days.

Details: https://labo.obs-mip.fr/multitemp/how-does-revisit-affect-data-fusion-methods-between-s2-and-a-potential-vhr-constellation/

Test of spatio-temporal fusion of Sentinel-2 New Generation data with a potential VHR mission – Séries Temporelles

At #CESBIO laboratory, we promote a new very high resolution mission, #SentinelHR, to complement #Sentinel2 with 2 m resolution images acquired with a systematic and global revisit every 10 to 30 days and 4 spectral bands.

We are ready to trade revisit for resolution, as we feel that landscape details with high frequency (buildings, trees, hedges, roads) tend to change less than the low frequency details. But demonstrating it needed work.

We made two studies to try to prove it.
🔽