Open jessjaco opened 1 year ago
We might consider using xrspatial hillshade
I think we should test without the hillshade first perhaps. I think this was only added to DE Africa after testing identified issues.
@sopac suggested the same, as terrain is likely not as big an issue in this study area.
This is an issue in a few places (around Noumea for one), but it has mostly been eliminated by my model-based filtering
of "bare_terrain" as I call it
Looks like they get the sun azimuth and elevation from the landsat metadata, then create a hillshade using matplotlib