Hi, dear roms users and developers,
I am trying to force my application directly using the atmospheric model without interpolation to the forcing fields.
on this page: http://www.myroms.org/index.php?page=forcing
it was mentioned that user can use a ascii file as forcing files, I am wondering if it is still possible with roms3.3 or the webpage is outdated?
on wikiroms:
https://www.myroms.org/wiki/index.php/F ... forcing.3F
It mentioned about "Gridded fields on a larger, usually coarser grid". But I cannot find detailed info on how to built such kind of netcdf file. The major questions are: what are the naming conventions for the coordinates (e.g., lat, lon) and what attributes are required for the variables. It will be helpful if there is a cdl template for that (coarser grid forcing). I did find cdl templates for forcing that has same grid with the model.
Thanks!
how to make a forcing file with a different grid than ROMS?
Re: how to make a forcing file with a different grid than RO
A cdl has been added to the wiki page you reference.
Re: how to make a forcing file with a different grid than RO
thanks, Kate!
In the cdl on that page:
double lat(lat) ;
lat:units = "degrees_north" ;
lat:point_spacing = "uneven" ;
lat:axis = "Y" ;
double lon(lon) ;
lon:units = "degrees_east" ;
lon:modulo = 360. ;
lon:point_spacing = "even" ;
lon:axis = "X" ;
Can I define coordinate as lon(lat,lon) and lat(lat,lon) if the grid is somehow slanted? Also are the attributes such as modulo, axis, point_spacing really used by ROMS?
In the cdl on that page:
double lat(lat) ;
lat:units = "degrees_north" ;
lat:point_spacing = "uneven" ;
lat:axis = "Y" ;
double lon(lon) ;
lon:units = "degrees_east" ;
lon:modulo = 360. ;
lon:point_spacing = "even" ;
lon:axis = "X" ;
Can I define coordinate as lon(lat,lon) and lat(lat,lon) if the grid is somehow slanted? Also are the attributes such as modulo, axis, point_spacing really used by ROMS?
kate wrote:A cdl has been added to the wiki page you reference.
Re: how to make a forcing file with a different grid than RO
The answers are all "no". I hacked some modulo code into my branch, but did it as a CPP flag instead. Checking for "modulo" would be good. Interpolating from typical WRF grids would probably be useful too, but it's tougher.