<div dir="ltr"><div class="gmail_default" style="font-size:small">Hi Cho,</div><div class="gmail_default" style="font-size:small"><br></div><div class="gmail_default" style="font-size:small">Guido is correct about the do loop screwing up the workstation.</div><div class="gmail_default" style="font-size:small"><br></div><div class="gmail_default" style="font-size:small">Each plot has a "parent" workstation, and if you change workstations for each plot, then by the time you get to the gsn_panel call, the only plot associated with "wks" at that point is the last plot.</div><div class="gmail_default" style="font-size:small"><br></div><div class="gmail_default" style="font-size:small">In general, you want to keep "do" loops as clean as possible, because they can be slow. </div><div class="gmail_default" style="font-size:small"><br></div><div class="gmail_default" style="font-size:small">One other thing to consider is using "addfiles" to read all the files at once, instead of reading one file at a time with "addfile". I don't know if that will work in your case, however, because I'm not familiar with your data. Also, if you have a lot of files, or really large arrays on each file, then "addfile" may still be your best bet.</div><div class="gmail_default" style="font-size:small"><br></div><div class="gmail_default" style="font-size:small">A shout-out to Guido for catching the workstation issue! </div><div class="gmail_default" style="font-size:small"><br></div><div class="gmail_default" style="font-size:small">--Mary</div><div class="gmail_default" style="font-size:small"><br></div></div><div class="gmail_extra"><br><div class="gmail_quote">On Wed, Mar 8, 2017 at 2:46 AM, Guido Cioni <span dir="ltr"><<a href="mailto:guidocioni@gmail.com" target="_blank">guidocioni@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div dir="auto"><div>Hey, <div dir="auto">In order to debug I would suggest you to move everything that does not have to change in every interaction (like the plot resources and the call to the workstation) out of the do loop. That can cause problems given that you're overwriting the same variable over and over. </div><div dir="auto">Second, use the := operator when reading files and defining variables so that you're sure that you're overwriting the old ones. </div><div dir="auto">I'm pretty sure it's the call to the workstation in every loop that is screwing up everything ☺️</div><div><div class="h5"><br><div class="gmail_extra"><br><div class="gmail_quote">Il 8 mar 2017 5:42 AM, 조익현 <<a href="mailto:ehcho@kma.go.kr" target="_blank">ehcho@kma.go.kr</a>> ha scritto:<br type="attribution"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Dear NCL users,<br>
<br>
<br>
I have a question about multi panel plots.<br>
<br>
I'm trying to get multi panel plots(3x3) using loop of 9 GRIB files,<br>
<br>
but it displays only last one output(plot(8)) on 3x3 panel.<br>
<br>
<br>
The basic structure is the following<br>
;;;<br>
do i1 = 0, 8<br>
grib file read and define resource<br>
plot(i1) = gsn_csm_contour_map<br>
end do<br>
<br>
resP = True<br>
gsn_panel(wks,plot,(/3,3/),res<wbr>P)<br>
;;;;<br>
<br>
<br>
Any help will be appreciated.<br>
<br>
Cho<br>
<br>
<br>
<br>
==============================<wbr>============ here is the script<br>
<br>
load "$NCARG_ROOT/lib/ncarg/nclscri<wbr>pts/csm/gsn_code.ncl"<br>
load "$NCARG_ROOT/lib/ncarg/nclscri<wbr>pts/csm/gsn_csm.ncl"<br>
load "$NCARG_ROOT/lib/ncarg/nclscri<wbr>pts/csm/contributed.ncl"<br>
<br>
begin<br>
<br>
<br>
<br>
ANLTIM = "<a href="tel:(201)%20703-0700" value="+12017030700" target="_blank">2017030700</a>"<br>
<br>
plot=new(9,graphic)<br>
<br>
do i1 = 0,8<br>
i=i1*12<br>
timestep = sprinti("%0.3i", i)<br>
<br>
dir = "/op/nwp/post/EXTJ/ECMW/OPER/D<wbr>AIN/201703/0700/"<br>
pr = "e025_v025_nhem_h"+timestep+".<wbr>"+ANLTIM+"00.gb1"<br>
a = addfile(dir+pr+".grb1", "r")<br>
<br>
lat = a->g0_lat_0({30:50})<br>
lon = a->g0_lon_1({110:140})<br>
<br>
slp = a->MSL_GDS0_SFC({30:50},{110:1<wbr>40})*0.01<br>
<br>
slp!0 = "lat"<br>
slp!1 = "lon"<br>
slp&lon = lon<br>
slp&lat = lat<br>
<br>
wks = gsn_open_wks("x11","output")<br>
gsn_define_colormap(wks, "WhiteYellowOrangeRed")<br>
<br>
;;============================<wbr>============================<br>
;; PLOT<br>
;;============================<wbr>============================<br>
res = True<br>
res@gsnDraw = False<br>
res@gsnFrame = False<br>
<br>
res@mpLimitMode = "Corners"<br>
res@mpLeftCornerLatF = 32.0<br>
res@mpRightCornerLatF = 43.0<br>
res@mpLeftCornerLonF = 119.5<br>
res@mpRightCornerLonF = 134.0<br>
<br>
res@mpProjection ="LambertConformal"<br>
res@mpLambertParallel1F = 30.0<br>
res@mpLambertParallel2F = 60.0<br>
res@mpLambertMeridianF = 126.0<br>
<br>
res@tfDoNDCOverlay = False<br>
res@gsnAddCyclic = False<br>
<br>
res@cnInfoLabelOn = False<br>
res@cnLineLabelInterval = 2<br>
; res@cnMonoLineThickness = False<br>
res@cnFillOn = False<br>
<br>
wrf_smooth_2d(slp,9)<br>
plot(i1) = gsn_csm_contour_map(wks,slp,re<wbr>s)<br>
<br>
end do<br>
<br>
;;;;;;;;;;;;;;;;;;;;;<br>
;;;; Panel Plot ;;;;;<br>
;;;;;;;;;;;;;;;;;;;;;<br>
resP = True<br>
gsn_panel(wks,plot,(/3,3/),res<wbr>P)<br>
end<br>
______________________________<wbr>_________________<br>
ncl-talk mailing list<br>
<a href="mailto:ncl-talk@ucar.edu" target="_blank">ncl-talk@ucar.edu</a><br>
List instructions, subscriber options, unsubscribe:<br>
<a href="http://mailman.ucar.edu/mailman/listinfo/ncl-talk" rel="noreferrer" target="_blank">http://mailman.ucar.edu/mailma<wbr>n/listinfo/ncl-talk</a><br>
</blockquote></div></div>
</div></div></div></div>
<br>______________________________<wbr>_________________<br>
ncl-talk mailing list<br>
<a href="mailto:ncl-talk@ucar.edu">ncl-talk@ucar.edu</a><br>
List instructions, subscriber options, unsubscribe:<br>
<a href="http://mailman.ucar.edu/mailman/listinfo/ncl-talk" rel="noreferrer" target="_blank">http://mailman.ucar.edu/<wbr>mailman/listinfo/ncl-talk</a><br>
<br></blockquote></div><br></div>