Re: [ferret_users] Regridding a WRF output using ferret

[Arturo Avila Rosas <aavilar@xxxxxx>]

Dear Hein,
Many thanks for your prompt response. I used the default WRF projection,
but I don't see XLONG and XLAT variables in the netcdf file, Do you mean
GLON and GLAT variables? Shall I use curv_to_rect function to achieve
step 1?
Below there is a WRF output, this includes some parameters such as:
:DX = 0.09629f ;
:DY = 0.096011f ;
:DT = 30.f ;
:CEN_LAT = 24.25f ;
:CEN_LON = -89.45f ;
:MAP_PROJ = 203 ;

Regards,
-- 
Arturo Avila Rosas
Ingeniería de Software
Instituto Mexicano del Petróleo
T. 91.75.73.99
 
netcdf wrfout_d01_2012-06-26_00_00_00 {
dimensions:
	Time = UNLIMITED ; // (1 currently)
	DateStrLen = 19 ;
	west_east = 159 ;
	south_north = 259 ;
	bottom_top = 27 ;
	bottom_top_stag = 28 ;
	soil_layers_stag = 4 ;
variables:
	char Times(Time, DateStrLen) ;
	float GLAT(Time, south_north, west_east) ;
		GLAT:FieldType = 104 ;
		GLAT:MemoryOrder = "XY " ;
		GLAT:description = "Geographic latitude, radians" ;
		GLAT:units = "" ;
		GLAT:stagger = "" ;
	float GLON(Time, south_north, west_east) ;
		GLON:FieldType = 104 ;
		GLON:MemoryOrder = "XY " ;
		GLON:description = "Geographic longitude, radians" ;
		GLON:units = "" ;
		GLON:stagger = "" ;
	float U10(Time, south_north, west_east) ;
		U10:FieldType = 104 ;
		U10:MemoryOrder = "XY " ;
		U10:description = "U at 10 M" ;
		U10:units = "" ;
		U10:stagger = "" ;
	float V10(Time, south_north, west_east) ;
		V10:FieldType = 104 ;
		V10:MemoryOrder = "XY " ;
		V10:description = "V at 10 M" ;
		V10:units = "" ;
		V10:stagger = "" ;
// global attributes:
		:TITLE = " OUTPUT FROM WRF V3.3 MODEL" ;
		:START_DATE = "2012-06-26_00:00:00" ;
		:SIMULATION_START_DATE = "2012-06-26_00:00:00" ;
		:WEST-EAST_GRID_DIMENSION = 160 ;
		:SOUTH-NORTH_GRID_DIMENSION = 260 ;
		:BOTTOM-TOP_GRID_DIMENSION = 28 ;
		:GRIDTYPE = "E" ;
		:DIFF_OPT = 1 ;
		:KM_OPT = 1 ;
		:DAMP_OPT = 1 ;
		:DAMPCOEF = 0.2f ;
		:KHDIF = 0.f ;
		:KVDIF = 0.f ;
		:MP_PHYSICS = 5 ;
		:RA_LW_PHYSICS = 99 ;
		:RA_SW_PHYSICS = 99 ;
		:SF_SFCLAY_PHYSICS = 2 ;
		:SF_SURFACE_PHYSICS = 2 ;
		:BL_PBL_PHYSICS = 2 ;
		:CU_PHYSICS = 2 ;
		:SURFACE_INPUT_SOURCE = 1 ;
		:SST_UPDATE = 0 ;
		:SF_URBAN_PHYSICS = 0 ;
		:FEEDBACK = 1 ;
		:SMOOTH_OPTION = 2 ;
		:SWRAD_SCAT = 1.f ;
		:W_DAMPING = 0 ;
		:DFI_OPT = 0 ;
		:SHCU_PHYSICS = 0 ;
		:WEST-EAST_PATCH_START_UNSTAG = 1 ;
		:WEST-EAST_PATCH_END_UNSTAG = 159 ;
		:WEST-EAST_PATCH_START_STAG = 1 ;
		:WEST-EAST_PATCH_END_STAG = 160 ;
		:SOUTH-NORTH_PATCH_START_UNSTAG = 1 ;
		:SOUTH-NORTH_PATCH_END_UNSTAG = 259 ;
		:SOUTH-NORTH_PATCH_START_STAG = 1 ;
		:SOUTH-NORTH_PATCH_END_STAG = 260 ;
		:BOTTOM-TOP_PATCH_START_UNSTAG = 1 ;
		:BOTTOM-TOP_PATCH_END_UNSTAG = 27 ;
		:BOTTOM-TOP_PATCH_START_STAG = 1 ;
		:BOTTOM-TOP_PATCH_END_STAG = 28 ;
		:DX = 0.09629f ;
		:DY = 0.096011f ;
		:DT = 30.f ;
		:CEN_LAT = 24.25f ;
		:CEN_LON = -89.45f ;
		:TRUELAT1 = 1.e+20f ;
		:TRUELAT2 = 1.e+20f ;
		:MOAD_CEN_LAT = 0.f ;
		:STAND_LON = 1.e+20f ;
		:GMT = 0.f ;
		:JULYR = 2012 ;
		:JULDAY = 178 ;
		:MAP_PROJ = 203 ;
		:MMINLU = "USGS" ;
		:NUM_LAND_CAT = 24 ;
		:ISWATER = 16 ;
		:ISLAKE = -1 ;
		:ISICE = 24 ;
		:ISURBAN = 1 ;
		:ISOILWATER = 14 ;
		:I_PARENT_START = 0 ;
		:J_PARENT_START = 0 ;
}

On Thu, 2012-07-12 at 09:12 +0200, Hein Zelle wrote:
> Dear Arturo,
> 
> > I have to work with an WRF output, basically I have to regrid the U10 and
> > V10 variables to 0.2 degrees and trim the region to get a smaller one
> > (97.85W:80.85W and 18.05N:30.45). Both U1o and V10 will be used in WAM. Any
> > idea will appreciated.
> 
> It's not easy to do this with Ferret.  What projection do you use?
> I used MATLAB initially, later a custom C++ program to achieve this.
> Steps to take:
> 
> - using XLONG and XLAT variables, regrid data fields to a new target
>   grid (regular in lon/lat).
> 
> - Rotate U10 and V10 (using SINALPHA, COSALPHA) to point northward.
> 
> Step 2 is easy (multiplication,addition) in ferret, step 1 is much
> harder.  You could try using the function scat2gridgauss_xy, but I
> haven't succeeded so far (but then, I haven't tried very hard).  
> 
> Kind regards,
>      Hein Zelle
> 
> 

-- 
Arturo Avila Rosas
Ingeniería de Software
Instituto Mexicano del Petróleo
T. 91.75.73.99