Hierarchies: Cloud-less aquaplanet driver

[LenkaNovak]

Following Frierson et al 2006

also see Frierson 2007 for reference.

Components

• add driver
• implement in CI
• implement in hierarchy tests
• enable time varying albedo in slab ocean

QA

• the above circulation is closer to the 10x moisture case in Frierson, with weak eddies and Ferrel cell, too strong tropical updrafts, lack of the cold point. Also it crashes after 180d. So this will require fine tuning.

Notes for fine tuning

• constant albedo crashes at 180d (with FriersonDiffusion and 160d standard diffusion), varying albedo after 100d. Debug plots show strong grid scale storms and not enough mixing, too moist near the surface, so consider 1) increasing the resolution, 2) increasing diffusion coefficient, 3) applying an upper tropo sponge, 4) revisit moisture removal, 5) add Betts Miller, 6) consider using parameters from this Atmos run.