GOES-West CONUS --fixed_grid dimensions

[jlc248]

I'm starting to work with GOES-17 (GOES-West) data and glmtools. I'm probably doing something stupid, but I was surprised to see the dimensions of the CONUS sector fixed grid as shape=(1500,3084), with dx = dy = 2.0. The GOES-West ABI CONUS fixed grid is shape=(1500,2500), as is the GOES-East ABI CONUS fixed grid. glmtools makes 1500x2500 grids for GOES-East.

So, I'm either doing something dumb, or not understanding why the two fixed grids would be different.

The command run:

(glmval) [jcintineo@fuego glmtools]$ python examples/grid/make_GLM_grids.py -o foo/{start_time:%Y/%b/%d}/{dataset_name} --fixed_grid --split_events --goes_position west --goes_sector conus --dx=2.0 --dy=2.0 *s20211891959*

Contents:

(glmval) [jcintineo@fuego glmtools]$ ncdump -h foo/2021/Jul/08/OR_GLM-L2-GLMC-M3_G17_s20211891959000_e20211892000000_c20211901353030.nc 
netcdf OR_GLM-L2-GLMC-M3_G17_s20211891959000_e20211892000000_c20211901353030 {
dimensions:
    x = 3084 ;
    y = 1500 ;
    dim_0 = 1 ;

What am I missing here?

[jlc248]

I assumed that GOES-17 GLM grid just extended eastward a bit more than the ABI fixed grid, but that doesn't seem to be the case. The x coordinates are slightly different between ABI and GLM, meaning some remapping has to be performed to co-locate the ABI and GLM grids. The y coords are the same.

At 2-km:

• ABI data: x[0] = -0.069972; x[2499] = 0.069972; y[0] = 0.128212; y[1499] = 0.044268005
• GLM data: x[0] = -0.0706440; x[2499] = 0.0693; y[0] = 0.128212; y[1499] = 0.044268

Is this by design? Or is this a bug? @deeplycloudy