[ncl-talk] Contour Plot Help

Adam Phillips asphilli at ucar.edu
Wed Sep 16 11:11:49 MDT 2015


No. Note that _FillValue denotes a specific value, so if you set _FillValue
to 300 only those grid cells whose value exactly equals 300. would be
masked.

_FillValue tells NCL that points with that value are to be ignored when
doing calculations or plotting. You should have _FillValue set for your var
array. If your advisor does not want to see values less than 300 I would
use the where function.

So, try this:
var at _FillValue = -9.99999e11
opt = True

opt at PrintStat = True

stat_dispersion(var,opt)   ; check to see that % missing is now not 0

var = where(var.lt.300,var at _FillValue,var)

Please review the _FillValue documentation as previously suggested

and take a look at the where documentation here:

https://www.ncl.ucar.edu/Document/Functions/Built-in/where.shtml


Hope that helps. If not, please continue responding to ncl-talk.

Adam



On Wed, Sep 16, 2015 at 10:15 AM, Mariama Barr - Dallas <bama4 at umbc.edu>
wrote:

> My PI (Private Investigator) only wants to see values of a certain range.
> He doesnt want to see values less than 300. So Would setting fillValue to
> say 300 help in any way? I will also look at the documentation on this
> FillValue function.
>
> On Wed, Sep 16, 2015 at 12:09 PM, Adam Phillips <asphilli at ucar.edu> wrote:
>
>> Yes I believe it does. From your output I see that you do not have a
>> _FillValue attribute set. Are you supposed to have values less than -1e5? I
>> am guessing that your _FillValue should be set to  -9.99999e11 based on the
>> minimum value. So set this:
>> var at _FillValue = -9.99999e11
>>
>> and resend the output from stat_dispersion. I am guessing this will solve
>> your issue..
>> Adam
>>
>>
>> On Wed, Sep 16, 2015 at 9:59 AM, Mariama Barr - Dallas <bama4 at umbc.edu>
>> wrote:
>>
>>> My mistake, I see now that the data ranges significantly more, does this
>>> affect the contour even though I set a min and max for contouring?
>>>
>>> On Wed, Sep 16, 2015 at 11:57 AM, Mariama Barr - Dallas <bama4 at umbc.edu>
>>> wrote:
>>>
>>>> I have a software feature that displays all data values that I used
>>>> before, but for specifics, here is the printout:
>>>>
>>>>
>>>>>>>>
>>>> On Wed, Sep 16, 2015 at 11:47 AM, Dennis Shea <shea at ucar.edu> wrote:
>>>>
>>>>> To my knowledge, you have not done what I requested. Right before you
>>>>> use the plot function
>>>>>
>>>>>    printVarSummary(var)
>>>>>    printMinMax(var,0)
>>>>>
>>>>> The output has not been sent to ncl-talk
>>>>> ---
>>>>>
>>>>>
>>>>> I'll also add that the following might help. Add this before the plot
>>>>> function
>>>>>
>>>>>
>>>>> https://www.ncl.ucar.edu/Document/Functions/Contributed/stat_dispersion.shtml
>>>>>
>>>>>
>>>>> ==
>>>>>
>>>>>   opt = True
>>>>>   opt at PrintStat = True
>>>>>   statb = *stat_dispersion*(var, opt )
>>>>>
>>>>>     printVarSummary(var)
>>>>>     printMinMax(var,0)
>>>>>
>>>>>   plot = gsn_csm_contour_map(wks, var, res)
>>>>>
>>>>> ===
>>>>>
>>>>>
>>>>> Also, as an experiment, you might try a run with
>>>>>
>>>>>
>>>>> *res at cnFillMode <https://www.ncl.ucar.edu/Document/Graphics/Resources/cn.shtml#cnFillMode>* = "RasterFill"
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> On Wed, Sep 16, 2015 at 8:37 AM, Mariama Barr - Dallas <bama4 at umbc.edu
>>>>> > wrote:
>>>>>
>>>>>> Well I have contour levels set at a min of 375 and a max of 425 and
>>>>>> the spacing is 10.0 and it still gives an algorithm failure error. I will
>>>>>> try and increase the spacing to see if it works but I really want as small
>>>>>> spacing as possible so that I can see smaller variations of the data in the
>>>>>> map.
>>>>>>
>>>>>> On Wed, Sep 16, 2015 at 10:32 AM, Rick Brownrigg <brownrig at ucar.edu>
>>>>>> wrote:
>>>>>>
>>>>>>> Hi,
>>>>>>>
>>>>>>> There would be a practical limit based upon your machine. Did you
>>>>>>> try intermediate values between 100M and 10G?  How about Adam's suggestion
>>>>>>> of explicitly setting contour levels -- trying something fairly coarse and
>>>>>>> then dialing it back in until things fail.  There's obviously something
>>>>>>> about the data that's requiring extreme amounts of workspace memory -- but
>>>>>>> its hard to say what without some sleuthing...
>>>>>>>
>>>>>>> Rick
>>>>>>>
>>>>>>>
>>>>>>> On Wed, Sep 16, 2015 at 8:17 AM, Mariama Barr - Dallas <
>>>>>>> bama4 at umbc.edu> wrote:
>>>>>>>
>>>>>>>> Yes, I also tried adjusting the space allocated in the .hluresfile
>>>>>>>> file from  100000000 to 10000000000. It gives an algorithm error. any
>>>>>>>> extreme allocatons gives a segmentation dump so I assume that there is a
>>>>>>>> reasonable limit to the amount that can be allocated to the ncarg graphics?
>>>>>>>>
>>>>>>>> On Tue, Sep 15, 2015 at 7:22 PM, Mary Haley <haley at ucar.edu> wrote:
>>>>>>>>
>>>>>>>>> Adam's suggestions were good ones.  If you are trying to plot the
>>>>>>>>> entire globe, then try cutting back back on the number of contour levels
>>>>>>>>> too, if you are explicitly setting them.
>>>>>>>>>
>>>>>>>>> --Mary
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> On Mon, Sep 14, 2015 at 12:47 PM, Mariama Barr - Dallas <
>>>>>>>>> bama4 at umbc.edu> wrote:
>>>>>>>>>
>>>>>>>>>> I am using NCL version 6.2.1. I commented those out because I was
>>>>>>>>>> using them previously to zoom in on a certain longitude and latitude but
>>>>>>>>>> right now I want to display the entire globe.
>>>>>>>>>>
>>>>>>>>>> On Mon, Sep 14, 2015 at 1:44 PM, Dennis Shea <shea at ucar.edu>
>>>>>>>>>> wrote:
>>>>>>>>>>
>>>>>>>>>>> Before the plot = ..
>>>>>>>>>>>
>>>>>>>>>>>    printVarSummary(var)
>>>>>>>>>>>    printMinMax(var,0)
>>>>>>>>>>>
>>>>>>>>>>> Why did you comment out the mpMinLatF, etc resources?
>>>>>>>>>>>
>>>>>>>>>>> Please always include this information when sending questions to
>>>>>>>>>>> ncl-talk.
>>>>>>>>>>>
>>>>>>>>>>> Also ... the version of NCL you are using.
>>>>>>>>>>>
>>>>>>>>>>> ncl -V
>>>>>>>>>>>
>>>>>>>>>>> ==========
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> On Mon, Sep 14, 2015 at 9:15 AM, Adam Phillips <
>>>>>>>>>>> asphilli at ucar.edu> wrote:
>>>>>>>>>>>
>>>>>>>>>>>> Hi Mariama,
>>>>>>>>>>>> My guess is that you have either a very high resolution grid
>>>>>>>>>>>> and/or a very noisy data array, and NCL is warning you that the output
>>>>>>>>>>>> graphic is exceeding the maximum size that is set.
>>>>>>>>>>>>
>>>>>>>>>>>> See the FAQ page here:
>>>>>>>>>>>> http://www.ncl.ucar.edu/FAQ/#err_msgs_012
>>>>>>>>>>>>
>>>>>>>>>>>> The solutions are:
>>>>>>>>>>>> 1) Increase your work station size as the FAQ describes.
>>>>>>>>>>>> 2) Change your contour settings. If it would be better to set
>>>>>>>>>>>> non-regular levels you can set cnLevelSelectionMode = "ExplicitLevels" and
>>>>>>>>>>>> set cnLevels to   (/200,300,340,380,420,460,500/)  (Those levels are a
>>>>>>>>>>>> complete guess.)
>>>>>>>>>>>> 3) If it is a very high resolution grid you can try plotting
>>>>>>>>>>>> every other point or every third point using ::2 or ::3 syntax.
>>>>>>>>>>>>
>>>>>>>>>>>> If none of those solutions help please let the ncl-talk email
>>>>>>>>>>>> list know.
>>>>>>>>>>>> Adam
>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>>> On Mon, Sep 14, 2015 at 8:46 AM, Mariama Barr - Dallas <
>>>>>>>>>>>> bama4 at umbc.edu> wrote:
>>>>>>>>>>>>
>>>>>>>>>>>>> Hello,
>>>>>>>>>>>>> Everytime I try to manually set the contour spacing to a value
>>>>>>>>>>>>>  40.0 and below, I get a
>>>>>>>>>>>>> "fatal:ContourPlotDraw: Workspace reallocation would exceed
>>>>>>>>>>>>> maximum size 100000000" error. Below is my code:
>>>>>>>>>>>>>
>>>>>>>>>>>>>
>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>>>>>>>>>> --
>>>>>>>>>>>>> Mariama Barr-Dallas
>>>>>>>>>>>>> Meyerhoff Scholar/ NSA Scholar M23
>>>>>>>>>>>>> Computer Science Major, History Major
>>>>>>>>>>>>> University of Maryland Baltimore County
>>>>>>>>>>>>> Class of 2015
>>>>>>>>>>>>>
>>>>>>>>>>>>> _______________________________________________
>>>>>>>>>>>>> ncl-talk mailing list
>>>>>>>>>>>>> ncl-talk at ucar.edu
>>>>>>>>>>>>> List instructions, subscriber options, unsubscribe:
>>>>>>>>>>>>> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
>>>>>>>>>>>>>
>>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>>> --
>>>>>>>>>>>> Adam Phillips
>>>>>>>>>>>> Associate Scientist,  Climate and Global Dynamics Laboratory,
>>>>>>>>>>>> NCAR
>>>>>>>>>>>> www.cgd.ucar.edu/staff/asphilli/   303-497-1726
>>>>>>>>>>>>
>>>>>>>>>>>> <http://www.cgd.ucar.edu/staff/asphilli>
>>>>>>>>>>>>
>>>>>>>>>>>> _______________________________________________
>>>>>>>>>>>> ncl-talk mailing list
>>>>>>>>>>>> ncl-talk at ucar.edu
>>>>>>>>>>>> List instructions, subscriber options, unsubscribe:
>>>>>>>>>>>> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> --
>>>>>>>>>> Mariama Barr-Dallas
>>>>>>>>>> Meyerhoff Scholar/ NSA Scholar M23
>>>>>>>>>> Computer Science Major, History Major
>>>>>>>>>> University of Maryland Baltimore County
>>>>>>>>>> Class of 2015
>>>>>>>>>>
>>>>>>>>>> _______________________________________________
>>>>>>>>>> ncl-talk mailing list
>>>>>>>>>> ncl-talk at ucar.edu
>>>>>>>>>> List instructions, subscriber options, unsubscribe:
>>>>>>>>>> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>> --
>>>>>>>> Mariama Barr-Dallas
>>>>>>>> Meyerhoff Scholar/ NSA Scholar M23
>>>>>>>> Computer Science Major, History Major
>>>>>>>> University of Maryland Baltimore County
>>>>>>>> Class of 2015
>>>>>>>>
>>>>>>>> _______________________________________________
>>>>>>>> ncl-talk mailing list
>>>>>>>> ncl-talk at ucar.edu
>>>>>>>> List instructions, subscriber options, unsubscribe:
>>>>>>>> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>>
>>>>>>
>>>>>> --
>>>>>> Mariama Barr-Dallas
>>>>>> Meyerhoff Scholar/ NSA Scholar M23
>>>>>> Computer Science Major, History Major
>>>>>> University of Maryland Baltimore County
>>>>>> Class of 2015
>>>>>>
>>>>>> _______________________________________________
>>>>>> ncl-talk mailing list
>>>>>> ncl-talk at ucar.edu
>>>>>> List instructions, subscriber options, unsubscribe:
>>>>>> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
>>>>>>
>>>>>>
>>>>>
>>>>
>>>>
>>>> --
>>>> Mariama Barr-Dallas
>>>> Meyerhoff Scholar/ NSA Scholar M23
>>>> Computer Science Major, History Major
>>>> University of Maryland Baltimore County
>>>> Class of 2015
>>>>
>>>
>>>
>>>
>>> --
>>> Mariama Barr-Dallas
>>> Meyerhoff Scholar/ NSA Scholar M23
>>> Computer Science Major, History Major
>>> University of Maryland Baltimore County
>>> Class of 2015
>>>
>>> _______________________________________________
>>> ncl-talk mailing list
>>> ncl-talk at ucar.edu
>>> List instructions, subscriber options, unsubscribe:
>>> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
>>>
>>>
>>
>>
>> --
>> Adam Phillips
>> Associate Scientist,  Climate and Global Dynamics Laboratory, NCAR
>> www.cgd.ucar.edu/staff/asphilli/   303-497-1726
>>
>> <http://www.cgd.ucar.edu/staff/asphilli>
>>
>
>
>
> --
> Mariama Barr-Dallas
> Meyerhoff Scholar/ NSA Scholar M23
> Computer Science Major, History Major
> University of Maryland Baltimore County
> Class of 2015
>



-- 
Adam Phillips
Associate Scientist,  Climate and Global Dynamics Laboratory, NCAR
www.cgd.ucar.edu/staff/asphilli/   303-497-1726

<http://www.cgd.ucar.edu/staff/asphilli>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20150916/cd97f493/attachment.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: code.PNG
Type: image/png
Size: 22215 bytes
Desc: not available
Url : http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20150916/cd97f493/attachment.png 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: code.PNG
Type: image/png
Size: 33933 bytes
Desc: not available
Url : http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20150916/cd97f493/attachment-0001.png 


More information about the ncl-talk mailing list