[ncl-talk] NDVI_mask problem
Dennis Shea
shea at ucar.edu
Fri May 11 07:59:23 MDT 2018
[1] *setfileoption* is placed *BEFORE *opening the nc file
[2] the *missing_value* attribute is associated with 'ndvi' not 'NDVI'
===================================================
;---Read NDVI data to contour and mask
dir = "./"
filename = "ndvi3g_geo_v1_1985_0106.nc4"
setfileoption("nc","MissingToFillValue", False) ; eliminate *warning*
message
f = addfile(dir+filename,"r")
ndvi = f->ndvi(0,:,:)
ndvi&lat at units = "degrees_north" ; Add units
attribute
ndvi&lon at units = "degrees_east"
ndvi at _FillValue = toshort(ndvi at missing_value) ; match typeof 'ndvi'
(netCDF rule)
printVarSummary(ndvi)
printMinMax(ndvi,0)
NDVI = floor(ndvi/10)/1000 ; no
meta data
copy_VarCoords(ndvi, NDVI) ; add
coordinate meta data
NDVI at long_name = "Normalized Difference Vegetation Index" ; add a
long_name
printVarSummary(NDVI)
printMinMax(NDVI,0)
On Fri, May 11, 2018 at 7:02 AM, Md. Jalal Uddin <dmjalal90 at gmail.com>
wrote:
> I am still getting following warnings though I put
> setfileoption("nc","MissingToFillValue", False)
> NDVI at _FillValue = toshort(NDVI at missing_value) ; match typeof 'ndvi'
> (netCDF rule).
>
>
> warning:NetOpenFile: MissingToFillValue option set True, but missing_value
> attribute and data variable (ndvi) types differ: not adding virtual
> _FillValue attribute
> warning:Attempt to reference attribute (missing_value) which is undefined
> warning:toshort: there are 1 int less than SHRT_MIN, which has been
> flagged missing.
>
> Please find the script herewith. Please help me to fix these problems.
>
>
>
> On Thu, May 10, 2018 at 10:02 PM, Dennis Shea <shea at ucar.edu> wrote:
>
>> ncl 1> ndvi = f->ndvi
>> ncl 2> print("typeof(ndvi)="+typeof(ndvi)) ; *"short"*
>> ncl 3> print("typeof(ndvi at missing_value)="+typeof(ndvi at missing_value))
>> ; *"double"*
>>
>> The variable (ndvi) does not have an *_FillValue* attribute associated
>> with it. By default, NCL tries to be 'smart' and set the variable's (ndvi)
>> *missing_value* to* _FillValue*. However, by netCDF (hence, NCL) rule
>> the type of the variable (ndvi) and the type of _FillValue attribute should
>> match. They do not. Again, one variable is type "short" the other is type
>> "double".
>>
>> The issue is explicitly mentioned in the error message.
>>
>> MissingToFillValue option set True, but missing_value attribute and data
>> variable (ndvi) types differ: not adding virtual _FillValue attribute
>>
>> Please read the text associated with:
>> https://www.ncl.ucar.edu/Document/Functions/Built-in/setfileoption.shtml
>>
>>
>> * MissingToFillValue*
>> You can explicitly set this option to False"
>>
>> * setfileoption*("nc","*MissingToFillValue*", False)
>>
>> This will eliminate the warning message.
>>
>> Still, you are left with no way to identify _FillValue. This is the
>> attribute NCL recognizes.
>> The following script on the NDVI page explicitly converts the type double
>> missing_value (=5000.0d)
>> to type short.
>>
>> ndvi_2.ncl
>>
>> NDVI = f->ndvi ; original unpacked values
>> NDVI at _FillValue = toshort(NDVI at missing_value) ; match typeof 'ndvi' (netCDF rule)
>>
>> You need to do this for masking.
>>
>> Good Luck
>>
>>
>> On Thu, May 10, 2018 at 6:00 AM, Md. Jalal Uddin <dmjalal90 at gmail.com>
>> wrote:
>>
>>> Thank you for your help. Mary's suggestions are working now. However, I
>>> am getting following errors.
>>>
>>> MissingToFillValue option set True, but missing_value attribute and data
>>> variable (ndvi) types differ: not adding virtual _FillValue attribute
>>>
>>> I also went to the NCL NDVI examples page (https://www.ncl.ucar.edu/Appl
>>> ications/ndvi.shtml) and tried to follow Example 2 that suggested by Dennis
>>> Shea. I got output but received the same problem.
>>>
>>> How could I solve the problem?
>>>
>>>
>>> On Wed, May 9, 2018 at 11:54 PM, Dennis Shea <shea at ucar.edu> wrote:
>>>
>>>> If you go to NCL's Examples page:
>>>>
>>>> https://www.ncl.ucar.edu/Applications/
>>>>
>>>> Under 'Datasets', click 'NDVI'
>>>>
>>>> There are three examples of differently sourced NDVI data sets.
>>>>
>>>> The 'GIMMS' version is Example 2.
>>>>
>>>> Please read the associated text description and the actual script.
>>>> As noted, the netCDF file is poorly written.
>>>>
>>>>
>>>>
>>>> On Wed, May 9, 2018 at 8:14 AM, Mary Haley <haley at ucar.edu> wrote:
>>>>
>>>>> Hi Jalal,
>>>>>
>>>>> I took a look at one of these NDVI data files, and noticed that the
>>>>> lat/lon coordinate arrays on the file have no attributes, which is not good.
>>>>>
>>>>> These arrays should at least have "units" and "long_name" attributes
>>>>> and units should be something like "degrees_north" and "degrees_east". For
>>>>> rectilinear data (data with lat/lon coordinate arrays), NCL needs these
>>>>> units to be something that indicates they are in degrees, or otherwise it
>>>>> can't be sure it is plotting the data correctly.
>>>>>
>>>>> It's always important to look at your data when you get errors like
>>>>> this. You do have:
>>>>>
>>>>> printVarSummary(ndvi)
>>>>>
>>>>> but I also added:
>>>>>
>>>>> printVarSummary(ndvi&lat)
>>>>> printVarSummary(ndvi&lon)
>>>>>
>>>>> in order to look at the lat/lon arrays to see what the units were and
>>>>> you can see there are none:
>>>>>
>>>>> Variable: lat (coordinate)
>>>>> Type: double
>>>>> Total Size: 17280 bytes
>>>>> 2160 values
>>>>> Number of Dimensions: 1
>>>>> Dimensions and sizes: [lat | 2160]
>>>>> Coordinates:
>>>>> Number Of Attributes: 0
>>>>>
>>>>>
>>>>> To fix this, I did this after reading in ndvi:
>>>>>
>>>>> ndvi = f->ndvi(0,:,:)
>>>>> ndvi&lat at units = "degrees_north" ; Fix the units
>>>>> ndvi&lon at units = "degrees_east"
>>>>> printVarSummary(ndvi)
>>>>>
>>>>> I also had to remove the reference to "ndvi at long_name" in a later
>>>>> title string, because ndvi doesn't have a long_name attribute.
>>>>>
>>>>> --Mary
>>>>>
>>>>> On Wed, May 9, 2018 at 3:44 AM, Md. Jalal Uddin <dmjalal90 at gmail.com>
>>>>> wrote:
>>>>>
>>>>>> Hi all,
>>>>>>
>>>>>> I tried to plot gimms ndvi (https://ecocast.ar
>>>>>> c.nasa.gov/data/pub/gimms/3g.v1/) data by masking. I am getting
>>>>>> following errors. I used copy_VarCoords(ndvi,ndvi_mask) command to
>>>>>> fix it but failed. Please help to solve this problem. Please find my script
>>>>>> in the attachment.
>>>>>>
>>>>>> (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'
>>>>>> (0) check_for_lon_coord: Warning: Data either does not contain a
>>>>>> valid longitude coordinate array or doesn't contain one at all.
>>>>>> (0) A valid longitude coordinate array should have a 'units'
>>>>>> attribute equal to one of the following values:
>>>>>> (0) 'degrees_east' 'degrees-east' 'degree_east' 'degrees east'
>>>>>> 'degrees_E' 'Degrees_east' 'degree_E' 'degreeE' 'degreesE' 'deg east'
>>>>>> (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'
>>>>>> (0) check_for_lon_coord: Warning: Data either does not contain a
>>>>>> valid longitude coordinate array or doesn't contain one at all.
>>>>>> (0) A valid longitude coordinate array should have a 'units'
>>>>>> attribute equal to one of the following values:
>>>>>> (0) 'degrees_east' 'degrees-east' 'degree_east' 'degrees east'
>>>>>> 'degrees_E' 'Degrees_east' 'degree_E' 'degreeE' 'degreesE' 'deg east'
>>>>>> warning:Attempt to reference attribute (long_name) which is undefined
>>>>>>
>>>>>> Cheers,
>>>>>> Jalal
>>>>>> --
>>>>>> *Md. Jalal Uddin*
>>>>>> MSc in Applied Meteorology (English Language)
>>>>>> Nanjing University of Information, Science and Technology, China
>>>>>> Jasmine Jiangsu Government Scholar
>>>>>> Cell: +8613260859092
>>>>>> Office: Beichen Building, School of Geography and Remote Sensing,
>>>>>> Room No. 406
>>>>>>
>>>>>> *AND *
>>>>>> B.Sc. in Disaster Management (Hons.)
>>>>>> Patuakhali Science and Technology University, Bangladesh.
>>>>>> Cell: +8801792052662, +8801838613203
>>>>>> Email: dmjalal90 at gmail.com
>>>>>> Web: www.dmjalal90.weebly.com
>>>>>> Facebook: jalal.hossen.39
>>>>>> LinkedIn: https://bd.linkedin.com/in/md-jalal-uddin-80a026b0
>>>>>> Twitter: dmjalal90
>>>>>> Skype: dmjalal90
>>>>>>
>>>>>> _______________________________________________
>>>>>> ncl-talk mailing list
>>>>>> ncl-talk at ucar.edu
>>>>>> List instructions, subscriber options, unsubscribe:
>>>>>> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
>>>>>>
>>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> ncl-talk mailing list
>>>>> ncl-talk at ucar.edu
>>>>> List instructions, subscriber options, unsubscribe:
>>>>> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
>>>>>
>>>>>
>>>>
>>>
>>>
>>> --
>>> *Md. Jalal Uddin*
>>> MSc in Applied Meteorology (English Language)
>>> Nanjing University of Information, Science and Technology, China
>>> Jasmine Jiangsu Government Scholar
>>> Cell: +8613260859092
>>> Office: Beichen Building, School of Geography and Remote Sensing, Room
>>> No. 406
>>>
>>> *AND *
>>> B.Sc. in Disaster Management (Hons.)
>>> Patuakhali Science and Technology University, Bangladesh.
>>> Cell: +8801792052662, +8801838613203
>>> Email: dmjalal90 at gmail.com
>>> Web: www.dmjalal90.weebly.com
>>> Facebook: jalal.hossen.39
>>> LinkedIn: https://bd.linkedin.com/in/md-jalal-uddin-80a026b0
>>> Twitter: dmjalal90
>>> Skype: dmjalal90
>>>
>>
>>
>
>
> --
> *Md. Jalal Uddin*
> MSc in Applied Meteorology (English Language)
> Nanjing University of Information, Science and Technology, China
> Jasmine Jiangsu Government Scholar
> Cell: +8613260859092
> Office: Beichen Building, School of Geography and Remote Sensing, Room No.
> 406
>
> *AND *
> B.Sc. in Disaster Management (Hons.)
> Patuakhali Science and Technology University, Bangladesh.
> Cell: +8801792052662, +8801838613203
> Email: dmjalal90 at gmail.com
> Web: www.dmjalal90.weebly.com
> Facebook: jalal.hossen.39
> LinkedIn: https://bd.linkedin.com/in/md-jalal-uddin-80a026b0
> Twitter: dmjalal90
> Skype: dmjalal90
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20180511/2ca51125/attachment.html>
More information about the ncl-talk
mailing list