Scheduled Downtime
On Friday 21 April 2023 @ 5pm MT, this website will be down for maintenance and expected to return online the morning of 24 April 2023 at the latest

High Urban Moisture with NOAH-MP in MPAS-Atmosphere 8.2.3

C

csteele

New member
I am running a limited domain MPAS run, clipped from the 3.75 km global mesh (which I refreshed last week to get the newer soil fields). I use the convection_permitting suite, but with mp_thompson_aerosols and sf_noahmp for lsm. I noticed that urban areas quickly develop high q2, despite qflx being lower than surrounding areas. This does not occur with just sf_noah. The images below show the same forecast with noahmp and noah for Alberta, Canada. Note the cities of Calgary and Edmonton show high dewpoints (from high q2). The ivgtyp looks good, at 13 for those areas (urban/built-up). Just wondering if noahmp is still not quite so ready yet, or are there many other steps needed to get it to run well?


mpas_noahmp_dewpoint_winds_2024071400.pngmpas_noah_dewpoint_winds_2024071400.png
 
Would you please give me more information about this case:
(1) which version of MPAS did you run?
(2) what data did you use to drive this case?
(3) please upload your namelist.init_atmosphere and namelist.atmosphere for me to take a look.
Thanks!
 
MPAS-Atmosphere Version 8.2.3
I closed my eyes and used GFS 0.25 hourly for this particular case.

namelist files attached
 

Attachments

  • namelist.atmosphere.txt
    2.1 KB · Views: 1
  • namelist.init_atmosphere.txt
    461 bytes · Views: 2
Is this the hotfix-v8.2.3 branch code for MPAS? Can you try the latest release-v8.3.0 branch code to see if the issue is still there? We will check our MPAS run to see if this q2 issue is there.
 
I just take a look at the codes in the official release of MPAS-V8.2.3. q2 diagnostics in NoahMP is significantly related to the fraction of bare soil and vegetation at an urban point. Can you check vegfra at a few urban points ( please look at this variable in both your initial data and diagnostic output files)? Please let know how they look. Thanks.
 
You may also want to output the vegetation fraction variable in your output file in addition to check the input file, because for urban grids, MPAS/NoahMP should modify the input vegetation fraction on the fly and assign vegetation fraction = 0 to those grids. I did not find any obvious bug related to this q2 issue.
 
OK, we also saw this unrealistic high q2 issue in our MPAS/NoahMP run. There may be a bug in the q2 diagnostics in the coupling between NoahMP and MPAS interface code. I am looking into this issue.
 
Thanks anyway. Derecho is down and I cannot repeat the case this week. I will try next week.

Please keep me updated for any progress...
 
Okay. I have confirmed this in a different domain with a shiny new 8.3.0 build. I will make those changes and recompile, and try again.
mpas_noahmp_dewpoint_winds_2025060421.png
 
You must log in or register to reply here.
Top