Reply to thread

When I had looked into detecting an overcast, cloudy sky, as it is sunlight through thick water vapor, the absorption spectra seems to be nothing special.

That leaves light intensity or scattering, but in a small, low cost module as OP surely wants, this seems not easy.

Campbell Scientific they use LIDAR (light detection and ranging) to find clouds in ceilometers. It uses a laser and backscattering.

For their pyranometers, as an example the graph shows water i.e. fog doesn't do a lot for infra-red it appears.

[ATTACH=full]198480[/ATTACH]