19990122: bogus eta levels in vis5d

>From: Owen Cooper <address@hidden>
>Organization: UVa
>Keywords: 199901221927.MAA23634 McIDAS Vis5D importer

Owen,

>I've been using vis5d to import mcidas eta data but am having a problem
>with vis5d creating two bogus data levels.  There are supposed to be 19
>data levels between 1000 and 100 mb (one every 50 mb).

Right.

>However I am
>getting two extra levels, one at 2.0 km (766.933 mb) and another
>at 10 km (252.469 mb).  When you plot vertical slices of temp. or u wind
>component you clearly get anomolous slivers of data at these levels.
>We think these levels originate from the following eta grids
>
>   126 99022      0 T    2    M  ETA    0  65  93   -PS-               
>   127 99022      0 RH   2    M  ETA    0  65  93   -PS-               
>   128 99022      0 U    10   M  ETA    0  65  93   -PS-               
>   129 99022      0 V    10   M  ETA    0  65  93   -PS-                
>
>And somehow the vis5d importer is translating these 2 M and 10 M
>heights into 2 km and 10 km.

Looks like a bug to me.

>Support for this argument comes from the
>fact that the following grids
>
>   139 99022      0 T    30   MB ETA    0  65  93   -PS-               
>   140 99022      0 RH   30   MB ETA    0  65  93   -PS-               
>   141 99022      0 U    30   MB ETA    0  65  93   -PS-               
>   142 99022      0 V    30   MB ETA    0  65  93   -PS-               
>   143 99022      0 T    60   MB ETA    0  65  93   -PS-               
>   144 99022      0 RH   60   MB ETA    0  65  93   -PS-               
>   145 99022      0 U    60   MB ETA    0  65  93   -PS-               
>   146 99022      0 V    60   MB ETA    0  65  93   -PS-               
>   147 99022      0 T    90   MB ETA    0  65  93   -PS-               
>   148 99022      0 RH   90   MB ETA    0  65  93   -PS-               
>   149 99022      0 U    90   MB ETA    0  65  93   -PS-               
>   150 99022      0 V    90   MB ETA    0  65  93   -PS-  
>
>are translated into 30, 60 and 90 km levels.

So, the importer is apparently thinking that any level less than 100 is
in km.

>I can get rid of these
>30 60 and 90 km levels because they are all at the top, and all I have to
>do is have vis5d import 3 less vertical levels and it simply chops off the
>top three.  But I haven't figured out a way for the vis5d importer to
>ignore the 2 and 10 km levels.  Is there a way for vis5d to drop these two
>levels?

I am not familiar enough with the importer to be able to answer this one.

>Please let me know if you have any ideas

You could get around this problem in another way:  copy the grids you
want to import into Vis5D into a different GRID file and then use the
importer.  For instance:

GRDCOPY RTGRIDS/ETA MYDATA/GRIDS.1234 PAR=T LEV=100 150 200 250 300 350 400 450 
500 550 600 650 700 750 800 850 900 950 1000 DAY=1999022 NUM=ALL

Since specifying all of the grid levels each time you want to run is
ugly, I would define a string to hold the list and specify it on the
command line:

TE ETALEVS "100 150 200 250 300 350 400 450 500 550 600 650 700 750 800 850 900 
950 1000
GRDCOPY RTGRIDS/ETA MYDATA/GRIDS.1234 PAR=T LEV=#ETALEVS DAY=1999022 NUM=ALL

You will, of course, have to set the list of PARameters to be
everything you want (or leave it out to get all parameters).  You may
also wish to specify more selection criteria so that grids of a
different spatial coverage are not included in the copy, etc.

I realize that this approach is less than ideal, but it is a way to get
you moving forward.

>Thanks
>Owen

Tom