Afdust_adj NCVGT Error (NetCDF dimension)

SMOKE
1

Hello,

I’m running NEI 2017 platform and faced a problem when running afdust_adj. I already run other sectors (e.g. onroad and nonpoint) including afdust with no error.
The error is:

NCVGT: : NetCDF: Index exceeds dimension bound
ERROR: Running precip adjustments

After searching in this forum, I’ve made sure that the time period is the same between met files and smoke runs. However, when I checked the dimensions, I noticed that METDOT3D file has one extra row and column. Is this normal?
Here are dimensions of the met files I’m using:

netcdf METCRO2D_171201:
TSTEP = UNLIMITED ; // (24 currently)
DATE-TIME = 2 ;
LAY = 1 ;
VAR = 35 ;
ROW = 299 ;
COL = 459 ;

netcdf METCRO3D_171201:
TSTEP = UNLIMITED ; // (24 currently)
DATE-TIME = 2 ;
LAY = 29 ;
VAR = 16 ;
ROW = 299 ;
COL = 459 ;

netcdf METDOT3D_171201:
TSTEP = UNLIMITED ; // (24 currently)
DATE-TIME = 2 ;
LAY = 29 ;
VAR = 6 ;
ROW = 300 ;
COL = 460 ;

Also, this is the dimension of merged smoke output (with no afdust):

emis_mole_all_20171201_12US1_nobeis_norwc_2017gb_17j:
TSTEP = UNLIMITED ; // (25 currently)
DATE-TIME = 2 ;
LAY = 1 ;
VAR = 63 ;
ROW = 299 ;
COL = 459 ;

Does anyone know what is wrong?

Thanks,
Soroush

2

There is no problem on MCIP files. METDOT is designed to stores the vector values from all corners (dots) of grid cell while METCRO files are for the scalar values from the center (cross) of grid cell. That is why METDOT has an extra dimension compared to METCRO.
You should use the METCRO2D file for the adjustment.

3

Hi Soroush,

Could I know how you fix this error?

Thanks

Just solved it. I need to remove wrong old files in afdust folder and rerun afdust and afdust_adj. It works now.

1 Like