SOLUTION for gfsptiles/gfs not being available or too slow

[windyweek]

Hello All,

My first post in this great forum and community

The thread is a bit old but will use the opportunity to ask something associated with the use of the master grib files.

Do you guys also use Vtable.GFS+TROP file (instead of Vtable.GFS) for GFS to metgrid variables mapping?
It seems that Vtable.GFS+TROP has few more variables and levels as follows:

Code: Select all

-----+------+------+------+----------+---------+-----------------------------------------+-----------------------+
  33 |   6  |   0  |      | UUW      | m s-1   | U                 at max wind           |  0  |  2  |  2  |   6 |
  34 |   6  |   0  |      | VVW      | m s-1   | V                 at max wind           |  0  |  2  |  3  |   6 |
   2 |   6  |   0  |      | PRESW    | Pa      | Pressure of max wind level              |  0  |  3  |  0  |   6 |
  11 |   6  |   0  |      | TTW      | K       | Temperature at max wind level           |  0  |  0  |  0  |   6 |
   7 |   6  |   0  |      | HGTW     | m       | Height of max wind level                |  0  |  3  |  5  |   6 |
  33 |   7  |   0  |      | UUT      | m s-1   | U                 at tropopause         |  0  |  2  |  2  |   7 |
  34 |   7  |   0  |      | VVT      | m s-1   | V                 at tropopause         |  0  |  2  |  3  |   7 |
   2 |   7  |   0  |      | PREST    | Pa      | Pressure of tropopause                  |  0  |  3  |  0  |   7 |
  11 |   7  |   0  |      | TTT      | K       | Temperature at tropopause               |  0  |  0  |  0  |   7 |
   7 |   7  |   0  |      | HGTTROP  | m       | Height of tropopause                    |  0  |  3  |  5  |   7 |
-----+------+------+------+----------+---------+-----------------------------------------+-----------------------+

For the forecasts I'm running I'm primary interested in U at 10m, V at 10m and T at 2m fields and would definitely give it a try some time soon but wanted to see what's the community best practices . Would like to know if you see any improvements (U10, V10 especially) with or without it.

Some more info about my configuration:
WRF ARW v3.5.1 (vanilla WRF, no EMS yet).
5km horizontal resolution, 50 vertical levels, the domain is over SE Europe, no nesting.
Physics options:
mp_physics=5,
ra_lw_physics=1,
ra_sw_physics=1,
radt=10,
sf_sfclay_physics=1,
sf_surface_physics=2,
bl_pbl_physics=1,
bldt=0,
cu_physics=1,
cudt=5,
isfflx=1,
ifsnow=1,
icloud=1,
surface_input_source=1,
num_soil_layers=4,
num_land_cat=21,
sf_urban_physics=0,
prec_acc_dt=60

[windyweek]

The latest post is quite old but worth answering it (yeah, answering my own question )

Vtable.GFS+TROP seems to be working in WRF 3.6.1+ - http://www2.mmm.ucar.edu/wrf/users/wpsv ... 3.6.1.html. Tried it on 3.5 with no luck.
3.7.1 also addresses some issues around tropopause and max wind levels grib fields - http://www2.mmm.ucar.edu/wrf/users/wpsv ... 3.7.1.html.