[Wrf-users] WRFDA and PSOTs

[Crowell, Sean M.]

Hi everyone,

I've emailed about this before, but will those of you who have a chance try to run a pseudo-single observation test (PSOT) with WRFDA by setting the variables in your namelist as:

&wrfvar15
num_pseudo = 0,
pseudo_x = 10,
pseudo_y = 10,
pseudo_z = 10,
pseudo_val = 1.0,
pseudo_err = 1.0,

&wrfvar19
pseudo_var = "u"

You'll need a wrfout file as a first guess and a background error file be.dat.  What I'm curious about is whether you get a response when you change the "u" to "v", "t" or "q".  I get a reasonable response when I set the variable to "u", but none at all when I set it to "v", "t" or "q".  The output shows that the observation is seen by WRFDA, but it doesn't contribute to the cost function.  If you have a chance, please let me know what you get when you give this a try.  I'm doing some work that is pretty heavily dependent on the length scales in the background error covariance, so I need this feature to work.

Sean



======================================
Sean Crowell
Postdoctoral Research Associate
College of Atmospheric and Geographic Sciences
University of Oklahoma
Office: NWC 5420
Phone: (405)325-1089
scrowell at ou.edu<mailto:scrowell at ou.edu>





-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.ucar.edu/pipermail/wrf-users/attachments/20120907/8fc5a390/attachment.html