MPAS derived fields in diag files (e.g., LCL, CAPE) are zero everywhere

[emurillo]

In my 3-km regional MPAS run over the CONUS, I get reasonable values for basic variables like MSLP, height_500hpa, surface temperature and humidity that look close to observations for that day. I'm using ERA5 for initial and boundary conditions. But the derived fields like refl10cm_max, cape, cin, lcl, etc. are all zero for the full 2 hours I'm testing.

I also can't find any documentation as to how CAPE and associated parcel variables are calculated (.i.e. surface-based, most unstable).

I'm attached my streams and namelist files here. Thanks!
Elisa

 

[Ming Chen]

Did you include these diagnostic variables in your stream_list.atmosphere.diagnostics?

Also, what severe weather existed during the 2-hr integration period?

if you run over a longer period, initialize the model 12-hr ahead of the severe weather occurrence, can you get reasonable variables of CAPE etc.?,

 

[emurillo]

Yes, these variables are included in my stream_list.atmosphere.diagnostics file.

The MCS I'm trying to simulate was over the south-central US that initialized around 19 UTZ June 9 2022.

I'm now running 00 UTZ June 9 2022 through 00 UTZ June 11 2022 to see if I can get reasonable values for some of these variables. Still, I'm not sure why CAPE, or especially LCL, would be uniformly zero across my entire CONUS domain for the 2 hours I was testing (00-02 UTZ June 9 2022).

 

[Ming Chen]

It is probably because convection hasn't well developed yet in the short period of 2-hour, although I am not 100% sure. It takes time for the model to spin-up and allow convection to develop. Please keep me updated of your results from the simulation over a longer period. Thanks.

 

[Ming Chen]

Hi Elisa,

I just took a look at your data saved in /glade/derecho/scratch/emurillo/regional_3km_uniform.

QV == 0 in your LBC file, which leads to extremely weak convective activities. Errors in your LBC data are derived from the ERA5 used to produce LBC files.

Please follow the steps below to rerun your case:

(1) download the latest version of era5_to_int.py

(2) run ear5_to_int.py to process ERA5 pressure level data

(3) rerun init_atmosphere_model to produce initiial and boundary files

(4) run atmosphere_model for at least 24 hours, with 9-12 hr of spin-up time for convection to fully develop

Let me know if you still have any issues. Thanks.

 

[emurillo]

Thank you for looking into this further. I completed steps 1-3 and then checked my new LBC files before rerunning the model. I found that QV still == 0 in the the LBC files. Do you know if there is something incorrect in my namelist.init.atmosphere or streams.init.atmosphere files?

Thank you

 

[Ming Chen]

Hi Elisa,

I run a quick test again. All work fine and I am able to get reasonable qv in the lateral forcing data.

I am perplexed by your result that shows zero qv in LBC files ....

Please let me know where your office is located and what days you come to office. I will drop by and go through the process with you together.

Ming

 

[Ming Chen]

I will be in office Monday-Thursday and most of Fridays. My office is FL3 3015.

You are also welcome to drop by for any MPAS issues.

 

[Ming Chen]

Just double check your namelist.init_atmopshere, and it looks fine.

 

[emurillo]

Sorry for the delayed response! I'm a fully remote postdoc, so we can set up a Google Meet to look through what might be going on. My Google Calendar is up to date and my working hours are 8am-3pm Eastern. Let me know if I may set up this meeting or if you would like to. Thanks!

 

[Ming Chen]

Hi Elisa,
This week is MPAS-WRF workshop. I will schedule a meeting next week. See you soon !

 

[Ming Chen]

Hi Elisa,

It is nice talking to you and thank you for sharing the plots that demonstrate the problems in LBC files.

Michael (our chief software engineer) fixed several bugs in the code mpas_init_atm_cases.F. Attached is the moisture field in the lowest level of the model, which I created using the modified code. I believe it is reasonable.

Please copy the code saved at /glade/derecho/scratch/chenming/MPAS-V8.2.3/src/core_init_atmosphere/mpas_init_atm_cases.F

Then use this code to replace the original one and recompile init_atmosphere, and rerun init_atmosphere to create LBC files. But you don't need to re-produce initial condition.

Please Let me know how it works. Thanks.

Ming