Re: writing time array as string to netcdf file?

[1]
To my knowledge, the only way to write strings to a netCDF file is to
use netCDF4. Strings are not allowed in netCDF3.

     setfileoption("nc", "Format", "NetCDF4Classic")

[2]
One of the main reasons for the *_n functions is to avoid
having to reorder the data. Assuming
    t(Time,bottom_top,south_north,west_east)
and
    z(Time,bottom_top,south_north,west_east)

Replace
xlvl =
int2p_n_Wrap(t(Time|:,bottom_top|:,south_north|:,west_east|:),z(Time|:,bottom_top|:,south_north|:,west_east|:),

> isolevs, 1, 1)

with

xlvl = int2p_n_Wrap(t,z,isolevs,1,1)

Good luck

On 3/11/11 1:41 PM, Kelly Mahoney wrote:
> Hi,
> I found a similar question from 2006
> <http://www.ncl.ucar.edu/Support/talk_archives/2006/0808.html> in the
> help archives, but I was wondering if this is since possible?
> Like the 2006 message, I also am trying to write my time array (defined
> as type "string") to an output netcdf file, but when it gets to the
> writing stage, my error says:
>> fatal:Assignment type mismatch, right hand side can't be coerced to
>> type of left hand side
>> fatal:Execute: Error occurred at or near line 126 in file
>> Freezing_Level_Height_Basic.ncl
>
> A snippet of my code is below -- if this can be done, or if there is a
> viable workaround, I'd be very appreciative to learn more!
> Thanks,
> Kelly Mahoney
>
>
> ****************************************************************************************************************************************
>
> load "$NCARG_ROOT/lib/ncarg/nclscripts/csm/gsn_code.ncl"
> load "$NCARG_ROOT/lib/ncarg/nclscripts/csm/gsn_csm.ncl"
> load "$NCARG_ROOT/lib/ncarg/nclscripts/csm/contributed.ncl"
> load "$NCARG_ROOT/lib/ncarg/nclscripts/csm/shea_util.ncl"
> load "$NCARG_ROOT/lib/ncarg/nclscripts/wrf/WRFUserARW.ncl"
>
> begin
> setvalues NhlGetWorkspaceObjectId()
> "wsMaximumSize" : 100000000000
> end setvalues
> setfileoption("nc", "SuppressClose", False)
>
> ;**********************************************************
> ; open files, make outfile, and read in required data
> ;**********************************************************
> outname = "FZLH_GFDLts_COMP_WSM6_PAST.nc"
> outfile=addfile(outname,"c")
>
> diri_fut = "WSM6_fut/"
> fils = systemfunc("ls "+diri_fut+"wrfout_d02_2040*")+".nc"
> nfil = dimsizes(fils)
>
> ;km Also make array for cumulative field of freezing level at all times
> frz_lev_alltimes = new((/nfil,1,600,573/),float)
> alltimes = new((/nfil/),string)
>
> ;km do nf=0,nfil-1
> do nf=1,1
>
> isolevs=0.
> isolevs!0="level"
> isolevs&level=isolevs
> isolevs@units="degree_C"
> isolevs@positive="up"
>
> f=addfile(fils(nf),"r")
> times = wrf_user_list_times(f) ; get time(s) in the file
> ntimes = dimsizes(times)
>
> lat=f->XLAT(0,:,:)
> lon=f->XLONG(0,:,:)
> hgt=f->HGT(0,:,:)
> ...
> t = wrf_user_getvar(f,"tc",-1) ; read in temp in celsius
> p = wrf_user_getvar(f, "pressure",-1) ; pressure is our vertical coordinate
> z = wrf_user_getvar(f, "z",-1) ; grid point height
> ...
> alllevel=t
> ...
> xlvl =
> int2p_n_Wrap(t(Time|:,bottom_top|:,south_north|:,west_east|:),z(Time|:,bottom_top|:,south_north|:,west_east|:),
> isolevs, 1, 1)
> ...
> ;check to see if interpolation failed due to inversion
> slice=ndtooned(xlvl) ;; convert depths to 1D slice in order to find
> missing values
> retry=(ind(ismissing(slice))) ;; store locations that have missing values
> ;; loop trough model levels to get past inversions that may be causing
> interpolation to fail
> do ilev=1,27
> ;km print("At level "+ilev+", points undefined = "+dimsizes(retry))
> mytmp =
> int2p_n_Wrap(t(Time|:,bottom_top|:,south_north|:,west_east|:),z(Time|:,bottom_top|:,south_north|:,west_east|:),
> isolevs, 1, 1)
> mytmp@missing_value=-999.
> mytmp@_FillValue=-999.
>
> slice2=ndtooned(mytmp) ;; convert new depths to 1D slice
> slice(retry)=slice2(retry) ;; replace old depths with new depths in slice
> delete(retry) ;delete locations of missing values in previous depth
> computation
> retry=(ind(ismissing(slice))) ;; check for values that are still missing
> and compute depths one layer deeper
> end do ; levels
> ;; after looping through model levels the 1D variable "slice" contains
> the updated depths
> ;; and must be converted back to it's multidimensional state.
> xlvl=onedtond(slice,dimsizes(xlvl))
>
> ;km: Find and print avg frz lev ht for whole domain:
> avg_frz_lev_ht = dim_avg_Wrap(dim_avg_Wrap(xlvl))
> print("For File #"+nf+", avg FRZLEV ht is "+avg_frz_lev_ht+"m")
>
> ;km: Write to cumulative field,
>
> frz_lev_alltimes(nf,:,:,:)=xlvl
>
> alltimes(nf)=times
>
> ;km: End file loop
> delete(t)
> delete(xlvl)
> delete(retry)
> delete(slice)
> delete(f)
> end do ; files
>
> ;km: Write cumulative field to output netcdf file:
>
> alltimes!0="time"
> alltimes&time=alltimes
>
> frz_lev_alltimes!0 = "time"
> frz_lev_alltimes!1 = "level"
> frz_lev_alltimes!2 = "south_north"
> frz_lev_alltimes!3 = "west_east"
>
> frz_lev_alltimes@units = "m above sea level"
> frz_lev_alltimes@long_name = "Frz Lev Height (m)"
> frz_lev_alltimes&time = alltimes
> print(alltimes)
>
> dimNames=(/"time","level","south_north","west_east"/) ;predefine
> coordinate vars
> dimSizes=(/nfil,1,600,573/)
> dimUnlim=(/True,False,False,False/)
> filedimdef(outfile,dimNames,dimSizes,dimUnlim)
>
> outfile->FZLH = frz_lev_alltimes
>
> delete(frz_lev_alltimes)
>
> end
>
> exit
>
>
>
>
>
>
> _______________________________________________
> ncl-talk mailing list
> List instructions, subscriber options, unsubscribe:
> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
_______________________________________________
ncl-talk mailing list
List instructions, subscriber options, unsubscribe:
http://mailman.ucar.edu/mailman/listinfo/ncl-talk
Received on Fri Mar 11 14:16:45 2011