Update saturation calculation based on model. #61
Conversation
…analysis: Use functions RSLF and RSIF from fv3 model (module_mp_thompson.F90) to calculate the liquid saturation vapor mixing ratio and the ice saturation vapor mixing ratio. This is better algorithm to avoid the Qs break down for very high temperatures at very low pressures.
hu5970
requested review from
christinaholtNOAA,
christopherwharrop-noaa,
chunhuazhou,
daviddowellNOAA and
guoqing-noaa
as code owners
| @@ -333,3 +342,72 @@ function ruc_saturation(Temp,pressure) | |||
| ruc_saturation= (watwgt*qvs1 + (one-watwgt)*qvi1) ! kg/kg | |||
| ! | |||
| end function ruc_saturation | |||
|
|
|||
| !+---+-----------------------------------------------------------------+ | |||
| !>\ingroup aathompson | |||
There was a problem hiding this comment.
This is now consistent with Thompson microphysics, but the doxygen comments should be modified.
|
|
||
| END FUNCTION RSLF | ||
| !+---+-----------------------------------------------------------------+ | ||
| !>\ingroup aathompson |
| ! | ||
|
|
||
| temp4=Temp ! convert to real |
There was a problem hiding this comment.
temp4 is only used as input to 2 functions. Do we need this variable or can the conversion be done in the function call?
There was a problem hiding this comment.
changed temp4 to temp_real for clarification. more detail below.
| ! | ||
|
|
||
| temp4=Temp ! convert to real | ||
| press4=pressure*100.0 ! covert to Pa |
There was a problem hiding this comment.
For clarity: pressure_pa = pressure*100.0
| qvs1 = 0.62198_r_kind*evs/(pressure-evs) ! qvs1 is mixing ratio kg/kg | ||
| !ruc evs = svp1*exp(SVP2*(Temp-273.15_r_kind)/(Temp-SVP3)) | ||
| !ruc qvs1 = 0.62198_r_kind*evs/(pressure-evs) ! qvs1 is mixing ratio kg/kg | ||
| qvs1 = RSLF(press4,temp4) |
There was a problem hiding this comment.
Maybe something similar to this: qvs1 = RSLF(pressure_pa, real(Temp, kind=r_kind))
There was a problem hiding this comment.
I tried "real(Temp, kind=r_kind)"; "real(Temp, kind=4)" and "real(Temp)". They all hang the cloud analysis.
So, go back to use:
temp_real=Temp
| qvi1 = 0.62198_r_kind*eis/(pressure-eis) ! qvi1 is mixing ratio kg/kg, | ||
| !ruc eis = svp1 *exp(22.514_r_kind - 6.15e3_r_kind/Temp) | ||
| !ruc qvi1 = 0.62198_r_kind*eis/(pressure-eis) ! qvi1 is mixing ratio kg/kg, | ||
| qvi1 = RSIF(press4,temp4) |
There was a problem hiding this comment.
See comment for function call above
Update the algorithm of the saturation mixing ratio calculation in non-var cloud analysis:
Use functions RSLF and RSIF from fv3 model (module_mp_thompson.F90) to calculate the liquid saturation vapor mixing ratio and the ice saturation vapor mixing ratio.
This is better algorithm to avoid the Qs break down for very high temperatures at very low pressures.
Tested with a single case. The difference between non-var cloud analysis with new and old saturation scheme in T, q, qi, qc are very small.