wgrib2: grid
Introduction
The grid option prints out the grid information.
$ wgrib2 gep19.t00z.pgb2af180 grid d 1
1:0:grid_template=0:winds(N/S):
latlon grid:(360 x 181) units 1e06 input WE:NS output WE:SN res 48
lat 90.000000 to 90.000000 by 1.000000
lon 0.000000 to 359.000000 by 1.000000 #points=65160
$ wgrib2 nam.683 grid
1:0:grid_template=30:winds(grid):
Lambert Conformal: (1473 x 1025) input WE:SN output WE:SN res 8
Lat1 12.190000 Lon1 226.541000 LoV 265.000000
LatD 25.000000 Latin1 25.000000 Latin2 25.000000
LatSP 0.000000 LonSP 0.000000
$ wgrib2 .t00z.master.grb2f048 grid d 1
ebis@landing2:~/grib2_examples$ wgrib2 gfs.t00z.master.grb2f048 grid  more
1:0:grid_template=40:winds(N/S):
Gaussian grid: (3072 x 1536) units 1e06 input WE:NS output WE:SN
number of latitudes between poleequator=768 #points=4718592
lat 89.909340 to 89.909340
lon 0.000000 to 359.882813 by 0.117188
$ wgrib2 merc.g2 grid d 1
1:0:grid_template=10:winds(N/S):
Mercator grid: (73 x 23) LatD 22.500000 input WE:SN output WE:SN res 48
lat 48.090000 to 48.090000 by 513669.000000 m
lon 0.000000 to 0.000000 by 513669.000000 m
orientation 0.000000
The four previous examples are for a grid definitions of a latlon, Lambert Conformal,
Gaussian and Mercator grids. These are the most common grids that are commonly distributed
from NCEP. Other commonly used grids are: polar stereographic, rotated latlon and
thinned Gaussian. Radar and satellites often use different grids.
Understanding the grid definitions
The grid definitions are based on the grib grib defintions as published
by the WMO and copied by the NCEP's grib documentation. For example, the
latlon grid defintion is given by
https://www.nco.ncep.noaa.gov/pmb/docs/grib2/grib2_doc/grib2_temp30.shtml.
Wgrib2's version of the latlon grid is given by
$ wgrib2 gep19.t00z.pgb2af180 grid d 1
1:0:grid_template=0:winds(N/S):
latlon grid:(360 x 181) units 1e06 input WE:NS output WE:SN res 48
lat 90.000000 to 90.000000 by 1.000000
lon 0.000000 to 359.000000 by 1.000000 #points=65160
template=N .. grid defintion template 3.N
winds(N/S) .. winds are earth relative
winds(grid) .. winds are grid relative
(NX x NY) .. grid dimensions (for rectangular grid)
units 1e06 .. scaling of angles, scaled angles are stored as 4byte integers
input WE:NS .. grib has the data in WE:NS scan order
output WE:SN .. wgrib2 internally has the data is in WE:SN scan order
if wgrib2 writes a bin/ieee/text file, it will be in the output scan order
The output scan order has to be WE:SN for geolocation to work.
The default output scan order is WE:SN.
output WE:NS .. wgrib2 internally has the data is in WE:NS scan order
The is enabled by order we:ns, and is used for writing binary data
in WE:NS order.
output raw .. wgrib2 internally has the data in the input scan order
This scan order is needed for new_grid to work.
res N .. value of the resolution and component flags octet (byte)
lat X to Y by Z .. latitudes start at X goes to Y by steps of Z
lon X to Y by Z .. longitudes start at X goes to Y by steps of Z
note: grib2 uses longitude is in [0,360) model
#points N .. number of points in the grid including grid points with undefined values
dlat and dlon in latlon grids
An above example showed the dlat and dlon for latlon grids. The dlat and dlon were
saved to the nearest 1e6 of a degree. For calculating the lat/lon of the grid points,
wgrib2 calculates a more accurate version of dlat and dlon from the end points
of the grid. For all the angle calculations, wgrib2 uses double precision. The only
exceptions is when angles are saved to rpn registers which are single precision, and if
wgrib2 is configured to call the old (single precision) iplib library.
Staggered Grids, wgrib2 2.0.8+
Staggered grids are often used in grid point (as opposed to spectral)
atmospheric models. (Arakawa, A.; Lamb, V.R. (1977). "Computational design of the
basic dynamical processes of the UCLA general circulation model". Methods in Computational Physics:
Advances in Research and Applications. 17: 173–265.) There are advantages in
storing the model grids in grib for both the modeler and the user. The advantages
are compactness, and a standard format. The user also has the advantage of
eliminating an extra interpolation step. The staggering information is stored
in the last 4 bits of flag table 3.4. If these bits are all zero, there is no
staggering.
Staggered grids in GRIB work by
 A "fundamental" grid is defined
 staggered grid could have an 0 dx offset in the X direction
 or staggered grid could have an 1/2 dx offset in the X direction
 or staggered grid could have 0 dx offset for odd rows and 1/2 dx for even rows (all in the X direction)
 staggered grid could have an 0 dy offset in the Y direction
 or staggered grid could have an 1/2 dy offset in the Y direction
 staggered grid could have fewer points in the row or column if the offset is nonzero
This scheme allows encoding the Arakawa AE Egrids. The wgrib2 grid will show
the staggering. The storage description was updated in v2.0.8 to be
 nx*ny: length of the row is nx, there are ny rows
 nx*(ny1): length of the row is nx, there are ny1 rows
 trimx*ny: length of the row is either nx or nx1, there are ny rows
 trimx*(ny1): length of the is either nx or nx1, there are ny1 rows
Usage
grid
See also:
nxny,
nlons
