Ocean Color Forum - Not logged in
Forum Ocean Color Home Help Search Login
Previous Next Up Topic SeaDAS / SeaDAS 6.x - General Questions / Seadas 6.3 run-time enviroment problems (locked) (1480 hits)
By daz Date 2012-04-14 17:01
the run-time enviroment interface of  seadas6.3 has changed (see attachment) since i started using seadas command-line on a new machine few days ago,
although i'm not sure whether using seadas command-line on the machines causes the problem of changing interface, it results in seadas crashes and X11 quits when i was trying to
scroll the display windows.

Also, seadas command "display" didn't work when i was runing the script but other commands after it like 'bl2map' and 'out' worked properly, Im not sure if these two things are related or not.

Attachment: Untitled1.png (154.2k)
By WhiteG Date 2012-04-14 19:24
Such problems are usually limited to very specific hardware/software.  What platform are you using (OS version, VM or native) and what video hardware? 
Do you have problems with other GUI applications?  Are there any relevant messages in the terminal where seadas is running or (assuming the most common case that you are running linux and using DISPLAY=:0.0 there may be useful messages in /var/log/Xorg.0.log)?  I've had similar problems in the past, usually they were due to bugs/features in the graphics drivers.  In the features case there are lots of obscure IDL option settings that can be found in ITTVIS technotes.  I have also seen problems with the virtual appliance that were solved by upgrading the Windows graphics drivers following advice of the VMware community postings.   If you are lucky some other SeaDAS user may have the same OS/hardware and already know a solution.
By daz Date 2012-04-17 16:16 Edited 2012-04-17 17:10
Thank you George
I sorted it out by changing the name of the home directory (it was called 'admin' and seadas dont like this name),

However, command 'display' still didn't work (no display window brings up) and loaded band wasn't projected by mapimg, im not sure whether something missed out, below is my code:

for i=0, N_ELEMENTS(filenames)-1 do begin & $
   load, filenames, FTYPE='MODIS', PROD_NAME=['flh','l2_flags'] & $
   mapimg, bands=[1],iproj=8,xsize=800,ysize=800,p0lat=LAT,p0lon=LON,scale=escala,ISOTROPIC=1 & $
   loadpal, paleta_chl, /file_idl & $
   loadgp, color=5, red=landcolor, green=landcolor, blue=landcolor & $
   display, BAND_NO=1, FBUF=1, smax=0.8, smin=-0.01, STYPE='LIN' & $ ; LIN or LOG
   landmask, FBUF=1, COLOR=5 & $      ; overlay landmask
   coast, /rivers, /coasts, COLOR=6 & $
   grid, latdel=1, londel=1, grdcol=6, lblcol=6, glinestyle=1, fonttype=-1, charsize=1,latalign=0.0, lonalign=0.0 & $
   cbar, FBUF=1, orient='v', fonttype=-1, charsize=2 & $
   out, L2DIR+'flh.png' , FTYPE='png' , FBUF=1, /CBAR, bk_cbar=0 & $ ; black background
   clear_up & $
endfor

Thank you

Dan
By WhiteG Date 2012-04-18 11:58
I doubt SeaDAS would be affected by the name of the "home" directory, so it is more likely that there was some problem with file permissions that was solved by the renaming. 

I can't run your script here as it refers to variables that have not been defined, but the "load, filenames" doesn't look right as filenames is an array.  Do you have a version of the script that worked with a previous version of SeaDAS?  To find the problem, start with a minimal script such as:

filenames=['benchmarks/A2006167181000.L2', 'benchmarks/A2006167181000.L2_test']
for ifile=0, N_ELEMENTS(filenames)-1 do begin & $
   print, filenames[ifile] & $
   load, filenames[ifile], FTYPE='MODIS', PROD_NAME=['flh','l2_flags'] & $
   clear_up & $
endfor

When I run this I get:

$ seadas -em -b t3.script
IDL Version 7.0, Mac OS X (darwin i386 m32). (c) 2007, ITT Visual Information Solutions

% Embedded IDL: NASA GSFC SeaDAS Development, SeaDAS.
% Embedded IDL: NASA GSFC SeaDAS Development, SeaDAS.
SeaDAS Version 6.3 (pid = 94304)
benchmarks/A2006167181000.L2
grp_name=Geophysical Data
Getting - "flh" data from HDF file...
Getting - "l2_flags" data from HDF file...
benchmarks/A2006167181000.L2_test
grp_name=Geophysical Data
Getting - "l2_flags" data from HDF file..

(note that the second file lacks flh).  Just keep adding lines from your original script and fixing problems as you find them.
By daz Date 2012-04-19 15:27 Edited 2012-04-19 16:25
Thank you, very much appreciated.

I think command 'display' is not meant to bring up any windows presenting loaded band. But now I have other problems which I hope you can help me with.

The first problems is 'coast' command, it does work properly on projected bands, no coastline was shown up by using  'coastal, /coasts' see attachment, but projected imagery can give coastline in seadas GUI.

The second problem is colour tables, I modified a standard colour table and save it as my own, but when I load it for one band next time in seadas GUI, the seadas got frozen after using 'Function' of 'Load Colour LUT' to make any changes when using 'Apply lut' to any loaded band, i've also tried to save LUT both in ASCII and hdf format, but it has the same problem, error message shown as below

% Attempt to subscript LUTFILES (SDSXLOADCT_COM) with LUTINDEX
   (SDSXLOADCT_COM) is out of range.
% Execution halted at: SDSXLOADCT_EVENT  489
   /Users/daz/seadas6.3/idl_lib/sdsxloadct.pro
%                      WIDGET_PROCESS_EVENTS
%                      $MAIN$    

my code:

L2DIR='/MODIS_Data/GOM_2005/20041213/'
filenames = findfile(L2DIR+'A*.L2')
PRINT,filenames
SLAT=24.0
NLAT=27.0
WLON=-83.5
ELON=-81

paleta_flh = '/MODIS_Data/GOM_2005/flh_2'
for i=0, N_ELEMENTS(filenames)-1 do begin & $
   load, filenames, FTYPE='MODIS', PROD_NAME=['flh','l2_flags','chl_oc3'] & $
mband_cmd,cmd_array=['result = b1*10'], navband=1, bandname='10*flh', unit='W/m^2 um^-1 sr^-1' & $
mapimg, bands=[3,4],xsize=670,ysize=800,LIMIT=[SLAT,WLON,NLAT,ELON],ISOTROPIC=1, /CYLIN  & $ ;, p0lat=LAT,p0lon=LON
   loadpal, paleta_chl, /file_idl & $
   loadgp, color=5, red=landcolor, green=landcolor, blue=landcolor & $
   display,BAND_NO=5,FBUF=1,smax=7, smin=0.05, STYPE='LOG' & $

   ;landmask, FBUF=1, COLOR=5 & $ ; overlay landmask
   coast, FBUF=1, /rivers, /coasts, COLOR=7, /HIRES & $
   grid, latdel=1, londel=1, grdcol=6, lblcol=6, glinestyle=1, fonttype=-1, charsize=1,charthick=1.5, latalign=0.5, lonalign=0.5 & $
   cbar, FBUF=1, orient='v', fonttype=-1, charsize=1.5,charthick=1.5 & $
   ;load_graph,5,FBUF=1,ftype='BAND',range=[0.05,7], color=[7],merge=2, /GEO, /inverse  & $

   display, BAND_NO=6, FBUF=2, smax=0.8, smin=-0.09, STYPE='LIN' & $ ; LIN or LOG
   load_graph,6,FBUF=2,ftype='BAND',range=[-0.09,0.8], color=[7], merge=2, /GEO, /inverse  & $
   landmask, FBUF=2, COLOR=5 & $ ; overlay landmask
   coast, FBUF=2, /rivers, /coasts, COLOR=6, /HIRES & $
   grid, latdel=1, londel=1, grdcol=6, lblcol=6, glinestyle=1, fonttype=-1, charsize=1,charthick=1.5, latalign=0.5, lonalign=0.5 & $
   cbar, FBUF=2, orient='v', fonttype=-1, charsize=1.5,charthick=1.5 & $
  
  
   out, L2DIR+'chl_oc3.png' , FTYPE='png' , FBUF=1, /CBAR, bk_cbar=1 & $ ; white background
   out, L2DIR+'flh.png' , FTYPE='png' , FBUF=2, /CBAR, bk_cbar=1 & $ ; white background
  
      out,L2DIR+'mapped_chl_oc3.HDF', /data, band=5, ftype='MAPPED',SD='chl_oc3_mapped',/GEO & $
   out,L2DIR+'mapped_chl_oc3.HDF', /data, band=6, ftype='MAPPED',SD='FLH*10_mapped',/GEO ,/append & $
  
      clear_up & $
endfor


By WhiteG Date 2012-04-19 17:12 Edited 2012-04-19 17:15
The documentation for the coasts command is rather sketchy.  I think /coasts is a toggle -- since the default is to draw coastlines, using coasts, ... ,/coasts turns off coastlines.   You can call coasts multiple times with different options, e.g.,

coast, FBUF=1, /rivers, /coasts    
coast, FBUF=1, color=7

should give rivers in red and coasts in white if you have the default colors.

http://oceancolor.gsfc.nasa.gov/forum/oceancolor/topic_show.pl?pid=18541 has information about user-defined LUT's.
By daz Date 2012-04-20 14:20
thank you, George
your replies are really helpful.I've sorted that out.
Previous Next Up Topic SeaDAS / SeaDAS 6.x - General Questions / Seadas 6.3 run-time enviroment problems (locked) (1480 hits)



Responsible NASA Official: Gene C. Feldman
Curator: OceanColor Webmaster
Authorized by: Gene C. Feldman
Updated: 27 November 2007
Privacy Policy and Important Notices NASA logo