WRF-ARW 3.4 winds too strong in nest when it is started after parent domain

[meteoadriatic]

Edit:
Some new important information has been discovered, keep reading below this post, especially on the PAGE 2 of this thread. Issue is still unresolved.

Hello.

I compiled ARW 3.4 and put it into my wrfems installation instead of 3.3.1 that works OK. But I have strange issue. 10m wind speeds are now way much stronger than in 3.3.1. At least a double speeds, everywhere, in all nests and in mother domain. I use all other settings (namelist, domains, etc.) the same, just replaced with new binaries.

Anybody of you tried ARW 3.4 already?

[surgeon]

I use 3.4 without ems, and w10 looks ok.

http://pogoda.wroclaw.pl/wrfarw/w10/

[meteoadriatic]

Can you please share your namelist.input? Thx a lot!

[surgeon]

On Your PM.

[meteoadriatic]

Conclusion.

I don't know where was the problem, but I made tests on different computer with different wrfems installation and there works OK. So I reinstalled wrfems on production computer, put v3.4 there and new TBL files and did few namelist mods and now new ARW is ready for publishing.

[alfe]

did few namelist mods

What kind of modification ? No impact on wind speed ?
Maybe also a TBL file not correct initially ? A wrong file with inapropriate roughness values ?

[meteoadriatic]

Well, I have to correct myself again.

I tested only 6 hours on test system and it was ok. But, in longer simulations still isn't OK. Winds somehow accelerates inside domain during time, especially into nests and toward the centre of domain. Near domain borders, where there is infulence from upper nest boundary conditions, wind is much better simulated then near domain centre where is at least double than realistic.

Example:

http://img204.imageshack.us/img204/8436/windu.png

this is 10m wind speed in m/s. Over 35m/s at maximum, realistic is not more than 15m/s there (as v3.3 gives). You can see how near domain border where is strong influence from outside nest, wind is MUCH slower then near centre.

No idea, I did changed everything in configuration I could. Almost every single entry in namelist I changed to something else. Every runable configuration with 3.4 binaries, gives this issue.

I lastly compiled also WPS, geogrid, metgrid, ungrib. Replaced them, recreated domain again. Replaced also all TBLs and Vtables with those that come with WRF and WPS 3.4 release. Nothing changed, at all. Issue is still there.

This is the last namelist I come up with, however, I don't believe the problem is somewhat in there because I changed almost everything:

Code: Select all

cat static/namelist.wrfm 
&time_control
 start_year                 = 2012, 2012
 start_month                = 04, 04
 start_day                  = 15, 15
 start_hour                 = 06, 18
 start_minute               = 00, 00
 start_second               = 00, 00
 end_year                   = 2012, 2012
 end_month                  = 04, 04
 end_day                    = 17, 17
 end_hour                   = 00, 00
 end_minute                 = 00, 00
 end_second                 = 00, 00
 interval_seconds           = 10800
 input_from_file            = T, T
 history_interval           = 180, 180
 history_outname            = "wrfout_d<domain>_<date>"
 frames_per_outfile         = 1, 1
 io_form_history            = 2
 io_form_input              = 2
 io_form_restart            = 2
 io_form_boundary           = 2
 io_form_auxinput2          = 2
 restart                    = F
 restart_interval           = 5040
 auxhist1_outname           = "sfcout_d<domain>_<date>"
 auxhist1_interval          = 0, 0
 frames_per_auxhist1        = 1, 1
 io_form_auxhist1           = 2
 auxhist2_outname           = "auxhist2_d<domain>_<date>"
 auxhist2_interval          = 0, 0
 frames_per_auxhist2        = 1, 1
 io_form_auxhist2           = 5
 fine_input_stream          = 0, 2
 adjust_output_times        = T
 debug_level                = 0
/

&domains
 time_step                  = 150
 time_step_fract_num        = 0
 time_step_fract_den        = 100
 max_dom                    = 2
 s_we                       = 1, 1
 e_we                       = 160, 103
 s_sn                       = 1, 1
 e_sn                       = 115, 73
 s_vert                     = 1, 1
 e_vert                     = 45, 45
 dx                         = 36448.6484, 12149.5498
 dy                         = 34803.5547, 11601.1846
 grid_id                    = 1, 2
 parent_id                  = 1, 1
 i_parent_start             = 1, 73
 j_parent_start             = 1, 42
 parent_grid_ratio          = 1, 3
 parent_time_step_ratio     = 1, 3
 feedback                   = 0
 smooth_option              = 1
 numtiles                   = 1
 nproc_x                    = -1
 nproc_y                    = -1
 num_metgrid_soil_levels    = 4
 num_metgrid_levels         = 27
 interp_type                = 1
 extrap_type                = 1
 t_extrap_type              = 1
 use_levels_below_ground    = T
 use_surface                = T
/

&physics
 mp_physics                 = 2, 2
 cu_physics                 = 6, 6
 cudt                       = 5, 5
 sf_sfclay_physics          = 1, 1
 sf_surface_physics         = 2, 2
 num_soil_layers            = 4
 num_land_cat               = 24
 num_soil_cat               = 16
 bl_pbl_physics             = 1, 1
 bldt                       = 0, 0
 ra_lw_physics              = 4, 4
 ra_sw_physics              = 4, 4
 radt                       = 30, 30
/

&dynamics                
 rk_ord                     = 3
 w_damping                  = 1
 diff_opt                   = 1
 km_opt                     = 4
 diff_6th_opt               = 0
 diff_6th_factor            = 0.12
 base_temp                  = 290.
 damp_opt                   = 3
 zdamp                      = 5000.
 dampcoef                   = 0.2
 khdif                      = 0
 kvdif                      = 0
 non_hydrostatic            = T
 moist_adv_opt              = 1
 scalar_adv_opt             = 1
/

&bdy_control             
 spec_bdy_width             = 5
 spec_zone                  = 1
 relax_zone                 = 4
 specified                  = T, F
 nested                     = F, T
/

&namelist_quilt          
 nio_tasks_per_group        = 0
 nio_groups                 = 1
/

Also, wrote wrfhelp, still waiting for reply.


Update: I also checked rsl_error.0000 outputs from real and from wrf, they are almost the same in v3.3 and v3.4, no significant differences between versions.

Surgeon: which compiler did you used?


Any further suggestion? Thx

[surgeon]

Surgeon: which compiler did you used?

Any further suggestion? Thx

I used gfortran. Maybe error is in GrADS script?

[meteoadriatic]

Surgeon: which compiler did you used?

Any further suggestion? Thx

I used gfortran. Maybe error is in GrADS script?

I used latest ifort (from parallel studio 2011). I'm now going to try compile with gfortran to exclude possible compiler issues.

GrADS no.. i use simple d mag(ugrd10m,vgrd10m) to compare my test runs... no way GrADS is wrong here, and as you can see the wind is much slower near domain borders so GrADS is ok here.. and wrfpost I think should be OK.

[meteoadriatic]

Progress update:

compiled 3.4 with PGI evaluation version. Same 10m wind results like with Intel. So, compiler is OK, binaries are OK.

Continuing to search for cause of issue....