[Wrf-users] Adaptive Time-step output times mismatch/incorrect

[Hutchinson, Todd]

Kevin ,

 

Thanks--I agree with your analysis of the code.  The adapt_timestep
function needs to be called for all domains.  It appears that the change
between WRF 3.1.1 and WRF 3.2 was instituted in order to handle
restarts, however, it looks like it broke the nested adaptive time step.


 

Assuming that you're not doing restarts, I agree and I would modify the
section in the solve_em.F WRF code to the version 3.1 version as you
have described.  

 

Todd

 

 

 

 

From: wrf-users-bounces at ucar.edu [mailto:wrf-users-bounces at ucar.edu] On
Behalf Of Kevin Matthew Nuss
Sent: Friday, October 22, 2010 5:37 PM
To: Zulauf, Michael; rahulm at uw.edu; wrf-users at ucar.edu
Subject: Re: [Wrf-users] Adaptive Time-step output times
mismatch/incorrect

 

Hi All,

I checked the code. The subroutine to adapt time steps is no longer
called for NESTED domains unless at a boundary update time, so it is not
able to adapt the time step to hit the output time, except for the outer
domain.

Here is the 3.1.1 code in solve_em.F when it worked:

    ! Adaptive time step: Added by T. Hutchinson, WSI  3/5/07
    !   In this call, we do the time-step adaptation and set
time-dependent lateral
    !   boundary condition nudging weights.
    !
       IF (config_flags%use_adaptive_time_step) THEN
        CALL adapt_timestep(grid, config_flags)
         adapt_step_flag = .TRUE.
       ELSE
         adapt_step_flag = .FALSE.
       ENDIF
    ! End of adaptive time step modifications

And here is the equivalent code in 3.2.1:

    ! Adaptive time step: Added by T. Hutchinson, WSI  3/5/07
    !   In this call, we do the time-step adaptation and set
time-dependent lateral
    !   boundary condition nudging weights.
    !
       IF ( (config_flags%use_adaptive_time_step) .and. &
            ( (.not. grid%nested) .or. &
            ( (grid%nested) .and. (abs(grid%dtbc) < 0.0001) ) ) )THEN
          CALL adapt_timestep(grid, config_flags)
          adapt_step_flag = .TRUE.
       ELSE
          adapt_step_flag = .FALSE.
       ENDIF
    ! End of adaptive time step modifications

The variable dtbc is descripbed as:  "TIME SINCE BOUNDARY READ." Since I
don't know the purpose of the changes, I won't suggest changing 3.2.1 to
look like 3.1.1

I also looked at the code related to the configuration variable
"adjust_output_times". To me it looks like the output string for the
file name is changed when this is "true" but not the timestep or
anything else.

Kevin Nuss



On Mon, Jun 28, 2010 at 3:05 PM, Zulauf, Michael
<Michael.Zulauf at iberdrolausa.com> wrote:

FYI, I've seen this exact issue with 3.2 also (and _only_ 3.2).  I've
had a couple emails with wrfhelp (the last being about a month ago), but
haven't heard back as to whether they were able to reproduce or
otherwise shed some light on this issue.  I expect they're busy.  But
I'll ping them again here.   ;-)

I've been planning to try a different set of compilers, and see if that
makes any difference.  But I've been busy too.

So, no help, but at least you're not alone.

Mike



-----Original Message-----
Date: Fri, 25 Jun 2010 11:17:35 -0700
From: Rahul Mahajan <rahulm at atmos.uw.edu>
Subject: Re: [Wrf-users] Adaptive Time-step output times
       mismatch/incorrect
To: David Ovens <ovens at atmos.washington.edu>
Cc: wrf-users at ucar.edu
Message-ID:
       <AANLkTikc8pDwNWZhHCtmbP0LrJ6m8Sy0chrsCp7pzqGC at mail.gmail.com>
Content-Type: text/plain; charset="iso-8859-1"

David,
It does look like the way you say. So, I suppose the issue is someplace
else.

Thanks,
Rahul.

On Fri, Jun 25, 2010 at 11:09 AM, David Ovens
<ovens at atmos.washington.edu>wrote:

> Rahul,
>
> I am not sure if this bug is the same that I found earlier in 3.2,
> since I noticed an output bug with a time_step of 216 seconds and
> adaptive time stepping was turned off.  But, take a look at your
> external/esmf_time_f90/ESMF_Clock.F90 around line 1024.  It should
> look like this:
>
>              IF ( ( .NOT. ( pred1 ) ) .AND. &
>                   ( ( pred2 ) .OR. ( pred3 ) ) ) THEN
>                 alarm%alarmint%Ringing = .TRUE.
> !                 alarm%alarmint%PrevRingTime =
clock%clockint%CurrTime
>                 alarm%alarmint%RingTimeSet = .FALSE.  !it is a one
time
> alarm, it rang, now let it resort to interval
> #if 1
> ! logic above is sufficient to set the PrevRingTime, it is now.
>                 IF ( positive_timestep ) THEN
>
> I hope that does it for you.
>
> David

*****************************




This message is intended for the exclusive attention of the address(es)
indicated.  Any information contained herein is strictly confidential
and privileged, especially as regards person data,
which must not be disclosed.  If you are the intended recipient and have
received it by mistake or learn about it in any other way, please notify
us by return e-mail and delete this message from
 your computer system. Any unauthorized use, reproduction, alteration,
filing or sending of this message and/or any attached files to third
parties may lead to legal proceedings being taken. Any
opinion expressed herein is solely that of the author(s) and does not
necessarily represent the opinion of Iberdrola. The sender does not
guarantee the integrity, speed or safety of this
message, not accept responsibility for any possible damage arising from
the interception, incorporation of virus or any other manipulation
carried out by third parties.


_______________________________________________
Wrf-users mailing list
Wrf-users at ucar.edu
http://mailman.ucar.edu/mailman/listinfo/wrf-users

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.ucar.edu/pipermail/wrf-users/attachments/20101025/1a12b1a9/attachment-0001.html