[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

20000920: 20000909: Zonal Wind Plot

Daryl,

It occurs to me that you should be using xav(urel) 
not xav(uwnd). UREL is the field that you will find in the
grid file- which is the grid relative u component of the wind.
That grid component will always be along the grid row
and pointing in the same relative direction to your plot
(whereas UWND will have point the other way on the other side of 
the earth thinking sperically not flat). 

Steve Chiswell
Unidata User Support


 
>From: Daryl Herzmann <address@hidden>
>Organization: UCAR/Unidata
>Keywords: 200009202212.e8KMCPb11942

>Dr Chiswell,
>
>I must be jinxed on this end, because I can not make this work.  I did
>everything your email specified.
>
>1. Changed      YAXIS = 0:1000
>2. Changed      PTYPE = log
>3. Edited pqact.conf to use  1.25x1.25 degrees resolution
>
>gdinfo 
> GRID FILE: /data/gempak/hrs/00092000_a.gem                                   
>       
>
> GRID NAVIGATION: 
>     PROJECTION:          CED                 
>     GRID SIZE:          289 145
>     LL CORNER:             -90.00    -30.00
>     UR CORNER:              90.00    -30.00
>
>
>So I created plots that show tbe differences.  Maybe these differences are
>not major??
>
>Here is a plot using CXSTNS  = -80.;100.>80.;100.
>http://www.meteor.iastate.edu/wx/data/current/100zonalWind.gif
>
>Here is a plot using CXSTNS  = -80.;300.>80.;300.
>http://www.meteor.iastate.edu/wx/data/current/300zonalWind.gif
>
>Here is my updated script.
>
>$GEMEXE/gdcross << EOF >> $LOGFILE
>       DEVICE  = $device
>       GDFILE  = $grid
>       WIND    = bm5
>       SCALE   = 0
>       GVECT   = 
>       GDATTIM = f00
>       GFUNC   = XAV(UWND)
>       GVCORD   = PRES
>       YAXIS   = 0:1000
>       CINT    = 1
>       CTYPE   = c
>       CXSTNS  = -80.;100.>80.;100.
>       TITLE   = 31/-2/100 Zonal Wind ~
>       PTYPE   = log
>       LINE    = 3/1/1
>       BORDER  = 1/1/3
>       CONTUR  = 1
>       FINT    = 
>       FLINE   = 
>
>       list
>       run
>
>       exit
>EOF
>
>
>Any ideas?
>       Daryl
>
>
>
>On Fri, 15 Sep 2000, Unidata Support wrote:
>
>~|
>~|Daryl,
>~|
>~|Several things to clarify here. First, your YAXIS value is not
>~|correct. Since you are using GVCORD=PRESS, YAXIS=0:10000 is not
>~|reasonable. YAXIS=  will default to 1020/100 which is what
>~|you appear to have in your gif.
>~|
>~|Second, Since your plot is PTYPE=lin instead of PTYPE=log (which
>~|is more conventional for looking at GVCORD=PRES), your surface
>~|elevation border appears very exagerated. One thing that may be 
>~|affecting your plot is that even though GFUNC should be the same at all poin
> ts
>~|along the longiture, the surface terrain will be different along your
>~|CXSTNS line. As a result, the gridding of the values into the X/Z plane
>~|will be different.
>~|
>~|One thing I notice is that your grid is very coarse. I used the
>~|AVN thinned data stored at 1.25x1.25 degrees resolution:
>~| GRID FILE: $MODEL/2000091500_thin.gem                                      
>                                                                 
>~|
>~| GRID NAVIGATION: 
>~|     PROJECTION:          CED                 
>~|     GRID SIZE:          289 145
>~|     LL CORNER:             -90.00    -30.00
>~|     UR CORNER:              90.00    -30.00
>~|
>~|
>~|If you are thickening the grid to a coarser resolution, then that
>~|combined with the varying surface topography probably accounts for 
>~|the discrepancy you are showing, since both will affect the resulting
>~|contour through the data.
>~|
>~|In your pqact.conf for decoding the AVN thinned grids, the -q parameter
>~|determines the output grid resolution. For the highest resolution,
>~|I have: -q "lin,dlat=1.25,dlon=1.25"
>~|
>~|Dr. Steve Chiswell
>~|Unidata User Support
>~|
>~|
>~|
>~|
>~|>From: Daryl Herzmann <address@hidden>
>~|>Organization: UCAR/Unidata
>~|>Keywords: 200009152140.e8FLebb17438
>~|
>~|>Mr Chiswell,
>~|>    Thanks for your repsonse. I believe that I am using a CED
>~|>grid.  gdinfo for my file reports..
>~|>
>~|> GEMPAK-GDINFO>r
>~|> GRID FILE: /data/gempak/hrs/00091500_a.gem  
>~|>
>~|> GRID NAVIGATION: 
>~|>     PROJECTION:          CED                 
>~|>
>~|>
>~|>    My script looks like this:
>~|>
>~|>set grid=/data/gempak/hrs/00091500_a.gem
>~|>set device="GF|zonalWind.gif"
>~|>
>~|>$GEMEXE/gdcross << EOF >> $LOGFILE
>~|>    DEVICE  = $device
>~|>    GDFILE  = $grid
>~|>    WIND    = 
>~|>    SCALE   = 0
>~|>    GVECT   = 
>~|>    GDATTIM = f00
>~|>    GFUNC   = xav(uwnd)
>~|>    GVCORD   = PRES
>~|>    YAXIS   = 0:10000
>~|>    CINT    = 2
>~|>    CTYPE   = c
>~|>    CXSTNS  = -80.;150.>80.;150.
>~|>    TITLE   = 31/-2/Zonal Wind AT 150
>~|>    PTYPE   = lin
>~|>    LINE    = 3/1/1
>~|>    BORDER  = 1/1/3
>~|>    CONTUR  = 1
>~|>    FINT    = 
>~|>    FLINE   = 
>~|>
>~|>    list
>~|>    run
>~|>
>~|>    exit
>~|>EOF
>~|>
>~|>
>~|>Here is a plot using CXSTNS  = -80.;120.>80.;120.
>~|>http://www.meteor.iastate.edu/wx/data/current/120zonalWind.gif
>~|>
>~|>Here is a plot using CXSTNS  = -80.;150.>80.;150.
>~|>http://www.meteor.iastate.edu/wx/data/current/150zonalWind.gif
>~|>
>~|>
>~|>    I am new to gempak, so I appologize if I am doing anything very
>~|>wrong.
>~|>
>~|>Thanks again,
>~|>    Daryl
>~|>
>~|>
>~|>On Fri, 15 Sep 2000, Unidata Support wrote:
>~|>
>~|>~|
>~|>~|Daryl,
>~|>~|
>~|>~|I have tried to duplicate your problem, but basically can't duplicate you
> r 
>~|>~|problem. First, let me make sure you are using a CED grid, otherwise
>~|>~|XAV which is the average along a grid row will not be along a constant
>~|>~|latitude (and therefore will change with your cxstns longitude position).
>~|>~|
>~|>~|I tested using the global AVN thinned grid. And ran for:
>~|>~|cxstns = -80;150>80;150
>~|>~|cxstns = -80;-150>80;-150
>~|>~|cxstns = -80;-30>80;-30
>~|>~|
>~|>~|Basically, all the contours aligned- except for a few small variations
>~|>~|for wiggles in the 0 line between east and west zonal winds. I presume 
>~|>~|that you are seeing much greater differences. Or is this the type of 
>~|>~|discrepancy you are seeing?
>~|>~|
>~|>~|I believe the wiggles I am seeing around 0 are roundoff errors in interpo
> lat
>~|> ing 
>~|>~|from the CXSTNS lat/lon values to the projected grid point locations amd
>~|>~|then gridding these for the x-z display.
>~|>~|
>~|>~|If you are seeing greater difference in the complete pattern, then it
>~|>~|would help me to know what the grid projection you are using is, the
>~|>~|source of the data, and if possible, a gif image of the variations
>~|>~|overlaid.
>~|>~|
>~|>~|Steve Chiswell
>~|>~|Unidata User Support
>~|>~|
>~|>~|
>~|>~|
>~|>~|
>~|>~|>From: Daryl Herzmann <address@hidden>
>~|>~|>Organization: UCAR/Unidata
>~|>~|>Keywords: 200009090542.e895g1b07934
>~|>~|
>~|>~|>Hello All,
>~|>~|> I have been trying to create a cross-section plot of the zonal
>~|>~|>wind averaged over all longitudes.  The plot would look like the classic
>~|>~|>zonal wind plots in textbooks, showing the jets.  Anyway, the problem is
>~|>~|>that I can not seem to get an average using gdcross.  The variable CXSTN
> S
>~|>~|>seems to be causing trouble, since I have to specify a longitude value,
>~|>~|>when I dont want to.
>~|>~|>
>~|>~|> Here is the beef of what I am trying now.
>~|>~|>
>~|>~|>$GEMEXE/gdcross
>~|>~|> CXSTNS  = -80.;150.>80.;150.
>~|>~|> GFUNC   = xav(uwnd)
>~|>~|>
>~|>~|>
>~|>~|> This will produce a nice plot, but if I chose a different value
>~|>~|>for CXSTNS, then the plot changes, which makes me believe that xav is
>~|>~|>indeed not working.
>~|>~|>
>~|>~|> Any ideas?
>~|>~|>
>~|>~|>Thanks much,
>~|>~|> Daryl
>~|>~|>                                                    __   ,,,,
>~|>~|>                                                   /  \_|$$$$|__,
>~|>~|>                                                   |     ==========}
>~|>~|>,,,,,, Daryl Herzmann  [ SR - MTEOR ]   ,,,,,,,,   +--+ JOHN    |_
>~|>~|>!!!!!! address@hidden              !!!!!!!! ___  |  DEERE ,_/
>~|>~|>$$$$$$ www.public.iastate.edu/~akrherz  $$$$$$$$/__|==-__-----_|
>~|>~|>|||||| ICQ: 25560887 AOL: "Daryl Herz"  ||||||||_______{}_____o_____
>~|>~|>
>~|>~|>
>~|>~|>
>~|>~|
>~|>~|*************************************************************************
> ***
>~|>~|Unidata User Support                                    UCAR Unidata Prog
> ram
>~|>~|(303)497-8644                                                  P.O. Box 3
> 000
>~|>~|address@hidden                                   Boulder, CO 80
> 307
>~|>~|-------------------------------------------------------------------------
> ---
>~|>~|Unidata WWW Service                        http://www.unidata.ucar.edu/  
>    
>~|>~|*************************************************************************
> ***
>~|>
>~|>
>
>