[Wrf-users] Bad SMSTAV Message in NMM core execution

[Eric Altshuler]

Hello Mac,

I also think your time step (72 seconds) is much too large. For NMM, the rule of thumb is to set the time step (in seconds) to about 2.25 times the horizontal resolution in km. Your resolution is 12 km so your time step should be about 27 seconds (it's better to use 26 2/3 so there will be an integer number of time steps in one hour). On the other hand, for ARW the rule is 6*DX so 72 seconds would be the recommended time step for 12 km resolution.

Best regards,

Eric L. Altshuler
Assistant Research Scientist
Center for Ocean-Land-Atmosphere Studies
4041 Powder Mill Road, Suite 302
Calverton, MD 20705-3106
USA

E-mail: ela at cola.iges.org
Phone: (301) 902-1257
Fax: (301) 595-9793

----- Original Message -----
From: "Eric Altshuler" <ela at cola.iges.org>
To: "mmacleod" <mmacleod at scotiaweather.com>
Cc: wrf-users at ucar.edu
Sent: Thursday, November 13, 2008 9:44:30 PM GMT -05:00 US/Canada Eastern
Subject: Re: [Wrf-users] Bad SMSTAV Message in NMM core execution

Hello Mac,

The 'bad SMSTAV' message is produced by subroutine SURFCE found in the source code file module_sf_lsm_nmm.F in directory 'phys'. The value of SMSTAV is obtained from a call to subroutine SFLX in the same source file (SMSTAV=SOILQW; the corresponding dummy argument is SOILW in the declaration of subroutine SFLX). It seems to be a soil wetness variable. Something is going wrong in the calculation of SOILW in subroutine SFLX. That's all I can determine without doing a lot more work.

Best regards,

Eric L. Altshuler
Assistant Research Scientist
Center for Ocean-Land-Atmosphere Studies
4041 Powder Mill Road, Suite 302
Calverton, MD 20705-3106
USA

E-mail: ela at cola.iges.org
Phone: (301) 902-1257
Fax: (301) 595-9793

----- Original Message -----
From: "mmacleod" <mmacleod at scotiaweather.com>
To: wrf-users at ucar.edu
Sent: Thursday, November 13, 2008 11:12:39 AM GMT -05:00 US/Canada Eastern
Subject: [Wrf-users] Bad SMSTAV Message in NMM core execution

Good morning.

We are setting up the WRF for use by our company.

In testing the NMM core on a Linux box, we continue to encounter the 
error message such as follows:
 
  bad SMSTAV:           14          16            NaN
  bad SMSTAV:           16          16            NaN
 Flerchinger used in NEW version. Iterations=          10
  bad SMSTAV:           19          17            NaN
  bad SMSTAV:           16          11            NaN
  bad SMSTAV:           15          18            NaN
  bad SMSTAV:           19          12            NaN
  bad SMSTAV:           17          13            NaN
  bad SMSTAV:           18          19            NaN
  bad SMSTAV:           20          19            NaN
  bad SMSTAV:           20          14            NaN
  bad SMSTAV:           16          20            NaN
  bad SMSTAV:           18          15            NaN
  bad SMSTAV:            2          21            NaN
  bad SMSTAV:           20          15            NaN
  bad SMSTAV:           16          21            NaN
  bad SMSTAV:            2          22            NaN
  bad SMSTAV:            2          23            NaN
  bad SMSTAV:            2          24            NaN
  bad SMSTAV:            7          24            NaN
  bad SMSTAV:            2          25            NaN
  bad SMSTAV:            2          26            NaN
 Flerchinger used in NEW version. Iterations=          10
  bad SMSTAV:            6          26            NaN

The grid used is a 12 km resolution 30X30 28 level at 72 second time 
steps.  The driving model is the NAM 12 km resolution in one hour 
outputs. We have run the model to 6 hours and this message appears with 
every 2nd time step as we have set the micro-physics call to 2. We also 
tested with the micro-physics call set to 6 and this appeared every 6th 
time step.

I have conducted a search of the source code for the combination "bad 
SMSTAV".  I can find lots of "bad" and lots of "SMSTAV" but none that 
come together, so the source of this message is a mystery to us. We ran 
the same grid on the same NAM input files using the ARW core and did not 
get this message.

If anyone can provide insight into this , we would be grateful.   To 
assist those who want to investigate this message, I am attaching text 
files of the "namelist.input" used and a text file titled 
"WRFRUNLOG.OUT" which shows the model output messages to the point where 
I kill it with "CTL" Z.

Thank you in advance for your help

Mac
 

-- 
M.A. (Mac) MacLeod
President and General Manager
Scotia Weather Services Inc
192 Wyse Road, Suite 8,
Dartmouth, N.S. B3A 1M9

Tele: 902-468-3866
Fax:  902-461-1768
E-mail: mmacleod at scotiaweather.com
Visit us: www.scotiaweather.com