Saturation vapor pressure calculation does not change at negative temperatures

[bsu-wrudisill]

Hello,

I would like to raise a question — it appears that the saturation vapor pressure of ice and liquid water are treated the same in summa.
There's only a formula for the saturation vapor pressure with respect to water in the conv_funcs.f90 file:

summa/build/source/engine/conv_funcs.f90

Line 72 in 19a7f9c

subroutine satVapPress(TC, SVP, dSVP_dT)

And, in the vegNrgFlux code, the saturation vapor pressure of the ground is computed this way:

summa/build/source/engine/vegNrgFlux.f90

Line 1089 in 19a7f9c

call satVapPress(TG_celcius, scalarSatVP_GroundTemp, dSVPGround_dGroundTemp)

TG_celcius = groundTemp - Tfreeze
call satVapPress(TG_celcius, scalarSatVP_GroundTemp, dSVPGround_dGroundTemp)

So, I don't think that the saturation vapor pressure with respect to ice is ever used for the snow surface? I think that TG_celcius is just the snow temperature of the uppermost layer in this case (if there is snow on the ground).

I actually care about the atmospheric water deposition process (which happens at cold temperatures) so I think that this might be an important process that depends quite strongly on how the surface vapor pressure is calculated!

Thank you. Please let me know if I'm mistaken about this.

[wknoben]

Hi Will,

Thanks for bringing this up. The underlying assumption here is that the difference between either equation is not huge, though you're right that the current implementation can be improved and that this can be important in certain cases.

If you're interested in implementing a more robust version we'd be happy to get a pull request. If you do, can I ask you to base this of the develop_sundials branch and direct the PR into that as well? That way your change will be most easily incorporated in the upcoming SUMMA v4 release.

[bsu-wrudisill]

Awesome, thanks for getting back to me so quickly! I have already started on a modification and will create a pull request after I do some more testing. Believe it or not this small change can make a pretty sizable difference in sublimation rates (I will provide some plots and test cases to show this).

Thanks
-Will

[wknoben]

Cool, looking forward to seeing it!

We do try to keep to a few PR guidelines. If you haven't already, it might be good to have a look at an existing PR (e.g., #588). The 6 tick boxes at the top are what we'd like to see (though more is obviously welcome too). Let me know if you have any questions in the meantime.