[ncl-talk] time_lat plot errors

[Kiswendsida Hyacinthe GUIGMA]

Many thanks for your help. Actually I was using gsn_csm_lat_time while in my mind I was using gsn_csm_time_lat. And also my graphics were not nice because I had some missing values coming from the running average and the linear interpolation. I removed them and all is fine now.Thank you!
 -----------------------------
GUIGMA 

    On Friday, 15 December 2017, 16:48, Mary Haley <haley at ucar.edu> wrote:
 

 Guigma,

You are calling gsn_csm_lat_time, which expects your variable to be ordered lat x time. ​Your variable is ordered time x lat, so you need to call gsn_csm_time_lat.
You should *always* look at your variables with printVarSummary before you try to plot them, to make sure they have the correct dimensions.  Here's what printVarSummary(x_regrid_sj)​ produces:​
Variable: x_regrid_SJType: floatTotal Size: 8096 bytes            2024 valuesNumber of Dimensions: 2Dimensions and sizes: [time | 92] x [lat | 22]Coordinates:             time: [1996.748657226562..1996.997314453125]            lat: [21..0]Number Of Attributes: 8  _FillValue : -99999  standard_name : precipitation  long_name : precipitation  units : mm/day  dataset : GPCP 1DD V1.2 Combined Data Sets  runave_op_ncl : runave_n: nave=10  average_op_ncl : dim_avg_n over dimension(s): lon  missing_value : -99999(0) Latitude (degrees_north) : min=0   max=21

If you want lat on the Y axis, then you need to transpose the array before you call gsn_csm_lat_time:

plot=gsn_csm_lat_time(wks, transpose(x_regrid_SJ), res)

​Second, the following two lines should be removed, beca​use both gsn_csm_lat_time and gsn_csm_time_lat will automatically check x_regrid_SJ for coordinate arrays:
res at sfXArray =x_regrid_SJ&lat

res at sfYArray =x_regrid_SJ&time


​--Mary​


On Fri, Dec 15, 2017 at 5:17 AM, Kiswendsida Hyacinthe GUIGMA <karongseba at gmail.com> wrote:
>
> Hi!
> I am trying to make a time-latitude plot of rainfall using NCL. From a NetCDF that covers the region comprise between 0-20N and 10W-10E, I averaged over the longitudes and so my data became 2D (time,lat). As my original data are from GPCP where coordinates are n.5,(n+1).5 ... while me I want my data on n, n+1 ... I interpolated grids using linint1_n_Wrap function. But the problem is when I use the gsn_csm_lat_time function to plot I first got this error:
> 1-
> (0)     check_for_y_lat_coord: Warning: Data either does not contain a valid latitude coordinate array or doesn't contain one at all.
> (0)     A valid latitude coordinate array should have a 'units' attribute equal
> to one of the following values:
> (0)         'degrees_north' 'degrees-north' 'degree_north' 'degrees north' 'degrees_N' 'Degrees_north' 'degree_N' 'degreeN' 'degreesN' 'deg north'
> warning:cnFillPalette is not a valid resource in time_lat1996_contour at this time
> (0)     get_lat_values: Warning: Your latitude values do not fall between -90 and 90 inclusive.
> (0)     You will not get 'nice' latitude labels.
>
> Then after reading answers to similar problems in NCL-TALK I modified my code to include sfXArray and sfYArray and also to have exactly the same MetaData as in the original file for the latitude coordinate. Then the error changed a bit and become:
> 2-
> (0)     get_lat_values: Warning: Your latitude values do not fall between -90 and 90 inclusive.
> (0)     You will not get 'nice' latitude labels.
>
> In addition to that, the plot doesn't fit completely with the window (see attached).
>
> Attached below: my code, the plot I got and the data I used.
>
> Your help is well appreciated.
>
>  
> -----------------------------
> GUIGMA
>
> _______________________________________________
> ncl-talk mailing list
> ncl-talk at ucar.edu
> List instructions, subscriber options, unsubscribe:
> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
>


   
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20171218/ed642b1b/attachment.html>