wtr_qflx_evap_tot column vs patch

[bandre-ucar]

Summary of Issue:

In clm4, qflx evap tot was calculated at the column level. In clm5, it appears to be both a patch and column level variable. Initial porting of data structures brought wtr_qflx_evap_tot in as a patch level variable instead of a column level variable as in clm4. I'm not sure if this is intentional and correct, or a mistake.

For consistency wtr_qflx_evap_tot is being used as patch level in clm5. This needs science review, possible conversion back to column.

Incomplete list of affected routines:

WisoSoilFluxes
WisoSoilSpecificHumidity
WisoHydrologyNoDrainage
WisoBareGroundFluxes
WisoLakeFluxes
WisoHydrologyLake
WisoDrainage
WisoSurfaceRunoff_urban

[bandre-ucar]

After looking at this more, I think I have to create the column level variable now. This still requires science review.