[Wrf-users] Unusual Velocity Spectra in Ideal LES w/ Wind

Anders A Nottrott anottrot at ucsd.edu
Wed Sep 28 11:18:24 MDT 2011

Hello All,


I'm running some simple ideal LES simulations of the neutral ABL forced with
10 m/s geostrophic wind.  In validating vertical profiles the mean wind,
velocity variance and momentum flux profiles agree well with other LES and
experimental data.  However, the velocity spectra begin to "roll-off" at
unusually low wavenumbers, i.e. there is too much energy content at the
highest wavenumbers (see the plots in the attached PDF).  A comparison with
Mirocha et al. (2010) indicates that this is not typical of WRF-LES.


In an attempt to diagnose the problem I have tried increasing vertical
resolution, 3 different SGS models (3D Smagorinsky, NBA-Stress, NBA-TKE) and
simulations with and without a capping inversion, but the results are
similar each time.  This leads me to believe that the problem is related to
one or more poorly selected options for numerics and as a result unphysical
high frequency waves are generated in the domain.


Note that I am using the 'em_les' test case as the basis for my runs so I
have 5th order RK3 in the horizontal and 3rd order RK3 in the vertical.  I
have set the acoustic timestep to be calculated automatically (i.e. namelist
variable 'time_step_sound=0').  It is worth noting that in purely convective
simulations (no background wind) the velocity and temperature spectra look
fine, so this is more evidence pointing to a problem with the advection


Has anyone else experienced this problem?  Any comments or suggestions will
be very helpful and most appreciated.







Anders Nottrott

PhD Student, Mechanical and Environmental Engineering

University of California, San Diego

Department of Mechanical and Aerospace Engineering

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.ucar.edu/pipermail/wrf-users/attachments/20110928/4fb41de8/attachment-0001.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: WRFLESspec.pdf
Type: application/pdf
Size: 31909 bytes
Desc: not available
Url : http://mailman.ucar.edu/pipermail/wrf-users/attachments/20110928/4fb41de8/attachment-0001.pdf 

More information about the Wrf-users mailing list