ETA Data support

[Kevin W. Thomas]

>Dear ARPS support,
>
>Recently I inquired about using ETA data in EXT2ARPS.  I retrieved the
>ETA files from ftpprd.ncep.noaa.gov, and mirrored what you did at
>http://www.caps.ou.edu/wx/p/r/conus/wrf, yet was unable to create my own
>initial condition.  The error I recieved in my ext2arps output file
>was "No 3-D variable was found in the GRIB file." However, there is data
>in the file, and the file was found.  What might be causing this?
>
>Also, with my previous inquiry, I recieved two replys.  One from Kevin
>Thomas and one from Yunheng Wang.  Yunheng Wang said I needed to have ARPS5.1.3
>(of which I am running version 5.0.0) and to read the comments for
>subroutine getnmceta218 for brief instructions on using ETA.  Kevin Thomas,
>however, did not say I needed version 5.1.3.  Before, when using the ARPS
>sample data, I was able to get ETA data read in correctly using my current
>version of ARPS. Is this still the case, or should I upgrade?
>
>Thanks in advance,
>Jim Perkins
>Coastal Carolina Univsersity

Jim...

If you are using ETA 40km data, then you don't have to have 5.1.3 or later.

If you are using ETA 12km data, then you do.  No guarantees here, as the
12km code has only been lightly tested.  It just takes too long to ftp the
required files from NCEP.  Local people are working on getting the data from
other sources.

Are you sure you are retrieving the right files?  The correct ones look like:

        eta.t##zawip3d$$.tm00

where "##" is the model run hour and "$$" is the valid hour into the forecast.
For example, eta.t18z.awip3d51.tm00.

For ETA 12km. the files look like:

        eta.t##z.awip218$$.@@

where "@@" is the tile number.

If you have the right files, make sure your copies are complete.  Truncated
files can give you strange errors.

        Kevin W. Thomas
        Center for Analysis and Prediction of Storms
        University of Oklahoma
        Norman, Oklahoma
        Email:  kwthomas at ou.edu