Re: Setting contour levels using a min/max contour level and a spacing

On Sep 16, 2010, at 3:38 PM, ziying-li@hdfgroup.org wrote:

> Hi Mary,
> Yes, only 2 values(0 and 39) were contained in this data field. I followed
> that example and generate reasonable plot.
>
> But I may have another question that If I want more values in label in
> this case. What I should do? I mean only 2 valid values in data (such as 0
> and 1), but I not only want to show 0 and 1 represent what, but also 2, 3
> and 4 represent what specifically in the label.

Hi Lizzie,

If you want more contour levels, and hence more boxes in your labelbar, then
you need to indicate this when you set the contour levels. The labelbar is
generated based on the contour levels that you have.

Right now, you are setting your contour levels with:

>>> res@cnLevelSelectionMode="ManualLevels"
>>> res@cnMinLevelValF=0
>>> res@cnMaxLevelValF=39
>>> res@cnLevelSpacingF=39
>>

This gives you two values, 0 and 39.

If you want more values, then you can change cnLevelSpacingF to something smaller and the cnMaxLevelValF to 40:

res@cnLevelSpacingF=8
res@cnMaxLevelValF=40

Now you have five contour levels, and you will get six boxes in your labelbar.

The first box will represent all values in your data < 0.

The second box will represent all values >= 0 and < 8
The third box will be values >= 8 and < 16.
The final box will be for values >= 40.

You can use res@lbLabelStrings to change the actual labels in your labelbar. By default you will get labels
that equal the values in your contour levels.

See the "vegland_1.ncl" example at http://www.ncl.ucar.edu/Applications/classification.shtml#ex1
for examples of changing your labelbar.

--Mary

>
> Thanks for your prompt reply every time.
>
> Lizzie
>
>> Hi,
>>
>> Without knowing what your data values are, I'm not sure what the reason is
>> behind the lack of red in your plot.
>>
>> You should print out the min/max of your data to make sure the values are
>> what you expect:
>>
>> print("min/max data = " + min(data) + "/" + max(data))
>>
>> When you select contour levels of 0 and 39, with the colors red and green
>> for these two levels, then what this means is:
>>
>> values < 0 will be red
>> values >= 0 and < 39 will be green
>>
>> Since you have a third box that is green, then this means anything >= 39
>> will be green.
>>
>> Is this what you intended?
>>
>> You might want to see example "vegland_1.ncl" at
>> http://www.ncl.ucar.edu/Applications/classification.shtml
>> for a sample of setting a color bar with just two values.
>>
>> --Mary
>>
>> On Sep 16, 2010, at 12:51 PM, ziying-li@hdfgroup.org wrote:
>>
>>> Mary,
>>>
>>> I follow your instruction for below case and got corresponding plot as
>>> attached.
>>>
>>> http://www.ncl.ucar.edu/Support/talk_archives/2010/1903.html
>>>
>>> I checked the latitude for this file and the location should be right.
>>> But
>>> I'm not sure why '0' can not be plotted.
>>>
>>> Do you know the reason behind it?
>>>
>>> Code:
>>> ----------------------------------------------------------------------------
>>>
>>> load "$NCARG_ROOT/lib/ncarg/nclex/gsun/gsn_code.ncl"
>>> load "$NCARG_ROOT/lib/ncarg/nclscripts/csm/gsn_csm.ncl"
>>> load "$NCARG_ROOT/lib/ncarg/nclscripts/csm/contributed.ncl"
>>>
>>> begin
>>> eos_file=addfile("MOD10_L2.A2000065.0040.005.2008235221207.hdf.he2",
>>> "r")
>>> ; Read file.
>>> ;To read HDF-EOS2 files, .he2 is appended to the argument.
>>> ;For more information, consult section 4.3.2 of
>>> http://hdfeos.org/software/ncl.php.
>>>
>>> data_byte=eos_file->Snow_Cover_MOD_Swath_Snow
>>> ; read data field, getting the clues about
>>> eos_file->Fractional_Snow_Cover_MOD_Swath_Snow
>>> ; from print(eos_file)
>>>
>>> data=byte2flt(data_byte)
>>> data@_FillValue = 255b
>>>
>>> nlon=4060
>>> nlat=2708
>>>
>>> lat=asciiread("lat_MOD10_L2.A2000065.0040.005.2008235221207.output",(/nlon,nlat/),"float")
>>> lon=asciiread("lon_MOD10_L2.A2000065.0040.005.2008235221207.output",(/nlon,nlat/),"float")
>>>
>>> data@lat2d=lat
>>> data@lon2d=lon
>>>
>>> xwks=gsn_open_wks("pdf","MOD10_L2.A2000065.0040.005.2008235221207_Snow_Cover.ncl")
>>> colors=(/"white","black","red","green"/)
>>> gsn_define_colormap(xwks,colors)
>>>
>>> setvalues NhlGetWorkspaceObjectId() ; make maximum filesize larger
>>> "wsMaximumSize" : 200000000
>>> end setvalues
>>>
>>> res=True
>>> res@cnFillOn=True
>>> res@gsnMaximize=True
>>> res@gsnPaperOrientation = "portrait"
>>> res@cnLinesOn=False
>>> res@cnLineLabelsOn = False
>>> res@gsnSpreadColors=True
>>> res@cnFillMode="RasterFill"
>>> res@cnMissingValFillPattern = 0 ; missing value pattern is set to
>>> "SolidFill"
>>> res@cnMissingValFillColor=0; white color for missing values
>>>
>>> res@cnLevelSelectionMode="ManualLevels"
>>> res@cnMinLevelValF=0
>>> res@cnMaxLevelValF=39
>>> res@cnLevelSpacingF=39
>>>
>>> res@lbLabelStrings = ispan(0,39,39)
>>> res@lbLabelPosition = "Center"
>>> res@lbLabelAlignment = "BoxCenters"
>>>
>>> plot=gsn_csm_contour_map_ce(xwks,data,res)
>>>
>>> delete(plot) ; cleaning up resources used
>>> delete(data)
>>>
>>> delete(res)
>>>
>>> delete(eos_file)
>>> end<MOD10_L2.A2000065.0040.005.2008235221207Snow_Cover.JPG>_______________________________________________
>>> 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 Sep 17 10:55:46 2010