Hi @Guille,
Thanks for sharing your data. It looks like the negative values in QF.tif are occurring on the edge of the raster where the DEM is nodata. Here’s a screenshot of what I’m seeing, where red pixels are less than 0 (all -11 QF values in this case ).
I’ll look into why this might be happening. Likely a nodata masking bug or alignment issue. But I don’t think it has a huge effect on the outputs. It seems to be a boundary issue.
I’ll update here what turns up, but let us know if you have any other concerns with the outputs you’re seeing.
Cheers,
Doug