[Wrf-users] WRF sea breeze

[Dorita Rostkier-Edelstein]

Hi Mikhail,
 
From my experience simulating sea-land breeze with WRF over Israel over complex topography (but with versions 2 of WRF) I suggest that you use 2-way nesting WITH feedback. My experience is that if large scale flow is strong unless you use 2-way with feedback the large scale will erase the high resolution mesoscale effects very quickly, in particular land-breeze. Also, re-start the model often, the range of predictability is very short at 1 km resolution (the uptake of the dynamics by the lateral boundaries, i.e., the large scale). I re-start 4 times a day, and spin up is very short, ~3 hours. I have not noticed significant differences between PBL parameterizations as compared to the differences caused by the nesting strategy and lead time. 
 
I will be very interested to read about your new results if you try my advice.
 
Dorita
 



----- Original Message ----
From: Mikhail Titov <mikhail.titov at canterbury.ac.nz>
To: wrf-users at ucar.edu
Cc: Mike Green <mike.green at canterbury.ac.nz>
Sent: Tuesday, September 30, 2008 3:32:17 AM
Subject: [Wrf-users] WRF sea breeze


Dear All,
 
I model diurnal-nocturnal cycle of see breeze for Australia (desert pattern) using WRF v3.0 with ETA levelling.
But in the modelled one year mean wind off-show night breeze is much better developed than on-shore day breeze. 
In fact WRF (4-grid run: 27-9-3-1 km with local topography) fails to reproduce on-shore day breeze to compare with observations.
 
My physical parameterisation is the next one:
=============================================
 &physics
 mp_physics                           = 6,     6,     6,    6,
 mp_zero_out                          = 2,
 ra_lw_physics                        = 1,     1,     1,    1,
 ra_sw_physics                       = 1,     1,     1,    1,
 radt                                       = 27,    27,    27,   27,
 sf_sfclay_physics                   = 1,     1,     1,    1,
 sf_surface_physics                 = 2,     2,     2,    2,
 bl_pbl_physics                       = 1,     1,     1,    1,
 bldt                                       = 0,     0,     0,    0,
 cu_physics                            = 1,     1,     0,    0,
 cudt                                      = 3,     3,     3,    0,
 isfflx                                      = 1,
 ifsnow                                   = 0,
 icloud                                   = 1,
 surface_input_source             = 1,
 num_soil_layers                    = 4,
 ucmcall                                = 0,
 mp_zero_out                        = 0,
 maxiens                              = 1,
 maxens2                             = 3,
 maxens3                             = 16,
 ensdim                               = 144,
 slope_rad                            = 0,
 topo_shading                      = 0,
 omlcall                               = 1,
 oml_hml0                           = 50.,
 oml_gamma                       = 0.13,
===========================================
 
I use physical parameterization applied to desert places of Africa and Arabia (WRF workshop, Boulder,  June 2008). 
Also I use two-way nesting (feedback = 0).
 
Any suggestions about what kind of physical parameterization is better to reproduce see breeze in WRF is highly appreciable.
Any experience how is better to model see breeze using WRF will be very useful. 
 
Many thanks,
Michael
 
--------------------------------------
Dr. Mikhail Titov 
Connell Wagner Ltd.
VP of the NZ Met.Society
Universityof Canterbury
Christchurch, New Zealand
E-mail: Mikhail.Titov at canterbury.ac.nz
--------------------------------------
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.ucar.edu/pipermail/wrf-users/attachments/20080930/f568a18e/attachment-0001.html