Re: subscript time coordinate of OLR data

Hi Carl, Nice reply. I found the following solution working well:   timeUnits = "hours since 1-1-1 00:00:0.0"   filtStart   = ut_inv_calendar( 1980, 01, 01, 00, 0, 0, timeUnits, 0 )   filtEnd     = ut_inv_calendar( 2000, 12, 31, 00, 0, 0, timeUnits, 0 )   inStart = filtStart + 192 * 24    inEnd   = filtEnd - 192 * 24      time = fin->time({inStart:inEnd})  lat  = fin->lat   lon  = fin->lon     data = short2flt(fin->$varName$({inStart:inEnd},:,:))      for new data, I calculated nt as:      maxDate =inEnd    minDate =inStart   nt = 1 + round(maxDate - minDate, 3 )/24   This works well, hopefully it will be correct.   Best regards,   JuKy    ________________________________ From: Carl Schreck <carl_at_cicsnc.org> To: juki juki <juky_emc2_at_yahoo.com> Cc: NCL Talk <ncl-talk_at_ucar.edu> Sent: Monday, June 25, 2012 9:29 PM Subject: Re: subscript time coordinate of OLR data Hi Jouky, If you run ncl_filedump on olr.day.mean.nc, it should show you the correct timeUnits: ---       double time ( time )          units :hours since 1-1-1 00:00:0.0          long_name :Time          actual_range :( 17298624, 17631720 )          delta_t :0000-00-01 00:00:00          standard_name :time          axis :T --- So in this case you will have to modify the other snippet of code to account for the times being in hours instead of days:   nt = 1 + round( 24 * ( maxDate - minDate ), 3 ) Hope this helps!     Carl On Mon, Jun 25, 2012 at 3:21 AM, juki juki <juky_emc2_at_yahoo.com> wrote: Dear All ncl users, > > >I need your help for subscripting the time coordinate of OLR data. For example, I want to subscript the following time: > > >  timeUnits = "days since 1800-01-01 00:00:00" >  inStart   = ut_inv_calendar( 1980, 01, 01, 00, 0, 0, timeUnits, 0 ) >  inEnd     = ut_inv_calendar( 2000, 12, 31, 00, 0, 0, timeUnits, 0 ) > > >  varName  = "olr" > > >  pathIn   = "olr.day.mean.nc" >  >  ;; Open the input files >  fin  = addfile( pathIn, "r" ) >  time = fin->time({inStart:inEnd}) >  lat  = fin->lat >  lon  = fin->lon >  >  data = short2flt(fin->$varName$({inStart:inEnd},:,:)) > > > ----------------- >With this subscripting technique, I got error "subcript out of range. The error is due to value of inStart and inEnd being not the same as one of value of time. I try to use another timeunits for example, timeUnits = "hours since 1-1-1 00:00:0.0", I can read the data but when making another process for example: > > >  maxDate =inEnd >  minDate =inStart > >  nx = 144 >  nt = 1 + round( maxDate - minDate, 3 ) >  nVar = 1; suppose only 1 > >  datax   = new( (/ nVar, nt, nx /), float ) >  datax!0 = "var" >  datax!1 = "time" >  datax!2 = "lon" >  >  datax(0,:,:) = dim_avg_n(data, 1 ) > > >I got error again because number of row of datax and data is not the same in which value of maxDate - minDate with timeUnits = "hours since 1-1-1 00:00:0.0" is much bigger than total number of row of OLR data. THUS, how to subscript the time coordinate of OLR data ? What time unit must be used ? > > >Thanks for help, > > >Regards, >Jouky > > >_______________________________________________ >ncl-talk mailing list >List instructions, subscriber options, unsubscribe: >http://mailman.ucar.edu/mailman/listinfo/ncl-talk > > -- Carl J. Schreck III, PhD Postdoctoral Research Associate Cooperative Institute for Climate and Satellites (CICS-NC) North Carolina State University and NOAA's National Climatic Data Center 151 Patton Avenue Asheville, NC 28801 Tel: 828-257-3140 carl_at_cicsnc.org http://monitor.cicsnc.org/mjo/ http://www.atmos.albany.edu/student/carl/

_______________________________________________
ncl-talk mailing list
List instructions, subscriber options, unsubscribe:
http://mailman.ucar.edu/mailman/listinfo/ncl-talk
Received on Mon Jun 25 19:09:26 2012