Unlimited dew (in theory)

The land can pull water from the atmosphere as negative latent heat flux - basically dew, latent heat flux from the atmosphere to the land. Most latent heat flux goes land-to-atmosphere, but a little bit can and does go the other direction. Right now, the magnitude of evap*dt when evap*dt < 0 isn't capped by the amount of moisture available in the bottom level of the atmosphere. So in theory, the atmosphere could be completely dry, and the land could invent a nice big flux of dew from thin air and do land-hydrology with it. The atmosphere has a check to correct this (the qneg4 function, I think, or maybe qneg3?), which I started trying to implement years ago but had trouble getting it working, and I ended up abandoning it as (a) I think this only ever has large-ish values early in spinup and (b) I don't think CLM bothers with this check. However, if we start pushing the model into weirder configurations (like moving continents around), I could see this potentially being a bit of an actual issue. So I should take another stab at this check. Variable to look for where I started trying to do this is `is_qneg` .