I have a large large netcdf file, of which I only need certain data. Therefore, I want to create a subdivision of this netcdf file using ncks. The netcdf file is as following:
Source:
F:\LECOB\Model\20091208_195356.nc
Format:
64bit
Global Attributes:
Model = 's26.bobshelf.20141113'
Title = 'S-NWM_BiP'
Dimensions:
ni_t = 682
nj_t = 712
nk_t = 29
time = 1 (UNLIMITED)
ni_w = 682
nj_w = 712
nk_w = 30
ni_u = 681
nj_u = 712
nk_u = 29
ni_v = 682
nj_v = 711
nk_v = 29
ni_f = 681
nj_f = 711
nk_f = 29
Variables:
time
Size: 1x1
Dimensions: time
Datatype: double
Attributes:
units = 'seconds from 2009-dec-08 17:00:14'
long_name = 'time'
standard_name = 'time'
time_origin = '2009-dec-08 17:00:14'
calendar = 'gregorian'
content = 'T'
axis = 'T'
associate = 'undefined'
ni_t
Size: 682x1
Dimensions: ni_t
Datatype: single
Attributes:
long_name = 'x_grid_index'
standard_name = 'x_grid_index'
content = 'X'
axis = 'X'
nj_t
Size: 712x1
Dimensions: nj_t
Datatype: single
Attributes:
long_name = 'y_grid_index'
standard_name = 'y_grid_index'
content = 'Y'
axis = 'Y'
nk_t
Size: 29x1
Dimensions: nk_t
Datatype: single
Attributes:
long_name = 'z_grid_index'
standard_name = 'z_grid_index'
content = 'Z'
axis = 'Z'
positive = 'up'
ni_w
Size: 682x1
Dimensions: ni_w
Datatype: single
Attributes:
long_name = 'x_grid_index_at_w_location'
standard_name = 'x_grid_index_at_w_location'
content = 'X'
axis = 'X'
nj_w
Size: 712x1
Dimensions: nj_w
Datatype: single
Attributes:
long_name = 'y_grid_index_at_w_location'
standard_name = 'y_grid_index_at_w_location'
content = 'Y'
axis = 'Y'
nk_w
Size: 30x1
Dimensions: nk_w
Datatype: single
Attributes:
long_name = 'z_grid_index_at_w_location'
standard_name = 'z_grid_index_at_w_location'
content = 'Z'
axis = 'Z'
positive = 'up'
ni_u
Size: 681x1
Dimensions: ni_u
Datatype: single
Attributes:
long_name = 'x_grid_index_at_u_location'
standard_name = 'x_grid_index_at_u_location'
content = 'X'
axis = 'X'
nj_u
Size: 712x1
Dimensions: nj_u
Datatype: single
Attributes:
long_name = 'y_grid_index_at_u_location'
standard_name = 'y_grid_index_at_u_location'
content = 'Y'
axis = 'Y'
nk_u
Size: 29x1
Dimensions: nk_u
Datatype: single
Attributes:
long_name = 'z_grid_index_at_u_location'
standard_name = 'z_grid_index_at_u_location'
content = 'Z'
axis = 'Z'
positive = 'up'
ni_v
Size: 682x1
Dimensions: ni_v
Datatype: single
Attributes:
long_name = 'x_grid_index_at_v_location'
standard_name = 'x_grid_index_at_v_location'
content = 'X'
axis = 'X'
nj_v
Size: 711x1
Dimensions: nj_v
Datatype: single
Attributes:
long_name = 'y_grid_index_at_v_location'
standard_name = 'y_grid_index_at_v_location'
content = 'Y'
axis = 'Y'
nk_v
Size: 29x1
Dimensions: nk_v
Datatype: single
Attributes:
long_name = 'z_grid_index_at_v_location'
standard_name = 'z_grid_index_at_v_location'
content = 'Z'
axis = 'Z'
positive = 'up'
ssh
Size: 682x712x1
Dimensions: ni_t,nj_t,time
Datatype: single
Attributes:
units = 'm'
long_name = 'sea surface height above geoid'
standard_name = 'sea_surface_height_above_geoid'
content = 'TYX'
associate = 'time latitude_t longitude_t'
coordinates = 'time latitude_t longitude_t'
_FillValue = -9999
CFL2D
Size: 682x712x1
Dimensions: ni_t,nj_t,time
Datatype: single
Attributes:
units = 's'
long_name = 'CFL2D'
standard_name = 'CFL2D'
content = 'TYX'
associate = 'time latitude_t longitude_t'
coordinates = 'time latitude_t longitude_t'
_FillValue = -9999
tem
Size: 682x712x29x1
Dimensions: ni_t,nj_t,nk_t,time
Datatype: single
Attributes:
units = 'degrees_Celsius'
long_name = 'sea_water_potential_temperature'
standard_name = 'sea_water_potential_temperature'
content = 'TZYX'
associate = 'time depth_t latitude_t longitude_t'
coordinates = 'time depth_t latitude_t longitude_t'
_FillValue = -9999
positive = 'up'
sal
Size: 682x712x29x1
Dimensions: ni_t,nj_t,nk_t,time
Datatype: single
Attributes:
units = '1e-3'
long_name = 'sea water salinity'
standard_name = 'sea_water_salinity'
content = 'TZYX'
associate = 'time depth_t latitude_t longitude_t'
coordinates = 'time depth_t latitude_t longitude_t'
_FillValue = -9999
positive = 'up'
vel_u
Size: 681x712x29x1
Dimensions: ni_u,nj_u,nk_u,time
Datatype: single
Attributes:
units = 'm s-1'
long_name = 'sea_water_x_velocity_at_u_location'
standard_name = 'sea_water_x_velocity_at_u_location'
content = 'TZYX'
associate = 'time depth_u latitude_u longitude_u'
coordinates = 'time depth_u latitude_u longitude_u'
_FillValue = -9999
positive = 'up'
vel_v
Size: 682x711x29x1
Dimensions: ni_v,nj_v,nk_v,time
Datatype: single
Attributes:
units = 'm s-1'
long_name = 'sea_water_y_velocity_at_v_location'
standard_name = 'sea_water_y_velocity_at_v_location'
content = 'TZYX'
associate = 'time depth_v latitude_v longitude_v'
coordinates = 'time depth_v latitude_v longitude_v'
_FillValue = -9999
positive = 'up'
kh
Size: 682x712x30x1
Dimensions: ni_w,nj_w,nk_w,time
Datatype: single
Attributes:
units = 'm2/s'
long_name = 'kh'
standard_name = 'kh'
content = 'TZYX'
associate = 'time depth_w latitude_t longitude_t'
coordinates = 'time depth_w latitude_t longitude_t'
_FillValue = -9999
positive = 'up'
tken
Size: 682x712x30x1
Dimensions: ni_w,nj_w,nk_w,time
Datatype: single
Attributes:
units = '(m/s)2'
long_name = 'tken'
standard_name = 'tken'
content = 'TZYX'
associate = 'time depth_w latitude_t longitude_t'
coordinates = 'time depth_w latitude_t longitude_t'
_FillValue = -9999
positive = 'up'
w
Size: 682x712x30x1
Dimensions: ni_w,nj_w,nk_w,time
Datatype: single
Attributes:
units = 'm s-1'
long_name = 'vertical_sea_water_velocity_at_w_location'
standard_name = 'vertical_sea_water_velocity_at_w_location'
content = 'TZYX'
associate = 'time depth_w latitude_t longitude_t'
coordinates = 'time depth_w latitude_t longitude_t'
_FillValue = -9999
positive = 'up'
Now I'm interested in the 4D variable vel_u (ni_u,nj_u,nk_u,time).
I want to extract ni_u 151 to 152, nj_u 234 to 235 , nk_u everything and time everything.
This question helped me along NCO: Extract a variable from NetCDF file using NCO ncks and so did the following link http://nco.sourceforge.net/nco.html#crd
Using those, I tried the following code on my linux computer:
ncks -C -F -d vel_u,151,152,1 20091208_195356.nc test.nc
This gives me two problems:
-C as suggested in question NCO: Extract a variable from NetCDF file using NCO ncks nj_uSo how do I put those parts (ni_u 151 to 152, nj_u 234 to 235 ) of the vel_u variable in my test.nc file?
Any answer is deeply appreciated!
I think I found the answer: -d is for dimensions and -v is for variables, making the answer to my question the following:
ncks -C -F -d nj_u,234,235,1 -d ni_u,151,152,1 -v vel_u 20091208_195356.nc test.nc
-C to make sure you only copy the variable vel_u
-F because caunting should start at 1 and not 0
-d NameDimension,Min,Max,Step
-v NameVariable
input.nc
output.nc
This seems to be giving me exactly what I want.