This statement is confirmed by the fact that, through Panoply, a GCTP_GEO projection is set in the metadata. Sadly GDALInfo is reporting something weird (I'm aware that the best option would be to use the conversion tool from NASA but for specific needs, I've to rely on GDAL )
Here below a metadata extraction of the 'projection' component:
...GEOGCRS["Unknown datum based upon the Clarke 1866 ellipsoid",
DATUM["Not specified (based on Clarke 1866 spheroid)",
I've been searching quite a lot but there is no documentation that states Clarke as the valid ellipsoid for this product.
Unfortunately, to complicate the situation, not all the products have the same CRS. The GDALInfo for the MCD43D31.061 doesn't have any metadata for the Coordinate System but through the PARAMETERNAME.1 confirm that has a CMG at 1KM
According to the General Cartographic Transformation Package (GCTP), the projection parameters (available here https://modis-land.gsfc.nasa.gov/GCTP.html) of the 'ellipsoid' should be referred to as a sphere of radius 6371007.181 in case of a projection to a sinusoidal. As the original software is written in the 90es has as default the Clarke ellipsoid, if parameters are not properly set, I'm wondering if an incident occurred at a certain point in the production chain.
Could someone please help me to better understand if the metadata in the MCD43D60 is wrongly assigned? Which would be the best solution? over impose the correct information or data have been compromised in the processing chain and only an intervention during it could solve the issue.