QNCLOUD in morr_double_moment_aero scheme

[Mercy]

I am using the Morrison double-moment aero microphysics scheme (mp=40). This scheme allows for initial droplet prediction through a variable INUM (line 354). By default, INUM=1.There is also a variable QDCNST=250, which is the initial constant cloud droplet concentration at Line 359.

Since I am interested in the droplet prediction, I have set the INUM=0. Hence, now QDCNST is not required as the scheme will predict droplet concentration depending on the aerosol input given and activation scheme selected, i.e., IACT =1/2/3/4 at line 371.

The QCLOUD in the output files appears sane; however, the QNCLOUD, which is the cloud droplet number concentration in the wrfout_d01*.nc file is
1) 250 if line 359 is not commented
2) missing value if line 359 is commented

Is it because the droplets formed from activation are not passed on to the output?
or is it because the cloud water content is calculated without any relation to the to the activated droplets.

I have attached the microphysics code used for the runs

 

[Ming Chen]

I take a quick look at this code, in which NC3D is calculated for determining QNCLOUD. Please let me know if I am wrong.

The default option of INUM (=1) makes sure that NC3D always has a physically meanful value. However, with the option INUM = 0, NC3D is only gievn value under certain conditions, for example,

1690        IF (QC3D(K).LT.QSMALL) THEN
1691          QC3D(K) = 0. 
1692          NC3D(K) = 0.
1693          EFFC(K) = 0. 
1694        END IF

Otherwise NC3D is not calculated and is not initialized. This explains why its value is NaN.

It is recommended to use the option INUM =1, which is hardwired inside the code. If you change it to INUM =0, some codes may need to change accordingly. For your case, probably an initialization of NC3D need to be added.