Re: Subtracting pan-Arctic MM5 analysis from ERA40 global reanalysis

Yes, you must regrid so the array sizes "conform".
This is required in any array language.

I speculate that the MM5 grid is much higher resolution
than the 2.5x2.5 ERA40 grid.

Justin Glisan wrote:
> Hello,
>
> I'm trying to subtract a monthly RMSD field (for the pan-Arctic MM5 from an ERA40 monthly RMSD field.
> The MM5 data is only for the polar region, but the ERA40 reanalysis covers the whole globe. When I run
> the program, I receive the error below. It may not make sense, since I'm subtracting a smaller field from
> a larger one. However, I am only interested in the polar region (i.e. the smaller spatial scale). It just so
> happens the ERA40 data is global and my MM5 data is polar.
>
> I think my question would revolve around regridding the ERA40 data to the MM5 grid, so the arrays are
> the same size.
>
> fatal:Dimension sizes of left hand side and right hand side of assignment do not match
> fatal:Execute: Error occurred at or near line 184
>
> Below, I've printed the variable attributes,
> ========================================
> Variable: rmsdTime_MM5
> Type: double
> Total Size: 95504 bytes
> 11938 values
> Number of Dimensions: 2
> Dimensions and sizes: [icross | 94] x [jcross | 127]
> Coordinates:
> Number Of Attributes: 4
> units : Pa
> standard_name : MSLP
> long_name : RMSD: Sea Level Pressure
> rmsd_op_ncl : dim_rmsd applied to dimension: time
> ========================================
> Variable: rmsdTime_ERA40
> Type: double
> Total Size: 84096 bytes
> 10512 values
> Number of Dimensions: 2
> Dimensions and sizes: [latitude | 73] x [longitude | 144]
> Coordinates:
> latitude: [90..-90]
> longitude: [ 0..357.5]
> Number Of Attributes: 5
> _FillValue : -32767
> missing_value : -32767
> units : Pa
> long_name : RMSD: Mean sea level pressure
> rmsd_op_ncl : dim_rmsd applied to dimension: time
> ========================================
> Variable: diff
> Type: double
> Total Size: 95504 bytes
> 11938 values
> Number of Dimensions: 2
> Dimensions and sizes: [icross | 94] x [jcross | 127]
> Coordinates:
> Number Of Attributes: 6
> rmsd_op_ncl : dim_rmsd applied to dimension: time
> long_name : RMS field differences
> standard_name : MSLP
> units : Pa
> lat2d : <ARRAY of 11938 elements>
> lon2d : <ARRAY of 11938 elements>
> (0) min=0 max=0
> ========================================
>
> I assume this is occurring because the actual spatial expanse of each field is different? Am I right to
> assume this? Each RMSD plot comes out fine for the respective input data (i.e. MM5 & ERA40), so I
> pretty sure the error is in the actual subtraction.
>
> ; Now, take the difference of the MM5 and the ERA40.
>
> diff = rmsdTime_ERA40 ; trick to create array with coordinate info
> diff = rmsdTime_ERA40 - rmsdTime_MM5 ; over write values with differences
> diff_at_long_name = "RMS field differences"
>

_______________________________________________
ncl-talk mailing list
ncl-talk_at_ucar.edu
http://mailman.ucar.edu/mailman/listinfo/ncl-talk
Received on Thu May 01 2008 - 08:04:24 MDT