Sentinel-2 Biophysical Processing, Low LAI in Forests


I wanted to check to make sure that I haven’t made mistakes using the S2 Thematic Land Processing / Biophysical Processor for derivation of an LAI product. I followed the precise steps I found on a different SNAP forum post, here: S2 LAI Processing, in order to create the LAI product. Unfortunately after processing it appears that the LAI values in the output NETCDF files are very low for summer forests. Ground truth in our deciduous plots of oaks and maples, and common sense, suggest an LAI of ~ 5-7 would be correct in summer, but the LAI output from SNAP returns values of ~2-3 for these areas. Does anyone have any thoughts regarding potential mistakes I could be making, or if the S2-derived LAI product may be more oriented toward agricultural measurements? I am happy to provide files as needed - I was unsure of which files may be desired. I’ll add that it appears that croplands around the forests of interest appear to show much larger increases in LAI from spring to summer (e.g., 0.2 to 6.7 LAI) relative to the forests (e.g., 0.2 to 2.7 LAI).Thank you!

Does anyone have any thoughts about this or might be able to confirm these discrepancies that I am seeing?