[ncl-talk] getind_latlon2d with rectilinear grid
Jiaxin Zhang
jiaxinzhang3 at gmail.com
Thu Sep 18 16:37:14 MDT 2014
Dear NCL users,
I am trying to extract variable values in some grids in specific locations
for different times.
I am attaching the NCL script that Dennis kindly provided to extract the
values for 3 grids and 1 time only. The attached NCL script works well with
the attached file "cVeg_CanESM2_1time.nc" that only contains 1 time. But I
am having problems when I try to run the same script for the attached file
"cVeg_CanESM2_12months.nc" that contains variable values for 12 times.
After modifying the script to include time as dimension when reading the
input file, I am getting error messages. I guess I am missing something but
I can not figure out what is wrong.
Could you please modify the attached script so that it works with the file
"cVeg_CanESM2_12months.nc"?
Your help is very appreciated!
Many thanks in advance!
Jiaxin
2014-09-18 13:22 GMT-07:00 Dennis Shea <shea at ucar.edu>:
> I'll respond here but I am quite busy and just can not answer multiple
> questions.
>
> Perhaps someone else can chime in?
>
> [1] The original file you provided had no explicit time dimension.
> The 'time' was includes as an attribute of the variable.
> Note that there are no attributes (eg: hours since ) associated with
> the time attribute.
> Actually, I believe that netCDF does not allow an attribute (here
> 'time') to have
> have an attribute.
>
> [2] In this case it is *user* responsibility to manually create a time
> dimension.
> It is user responsibility to manually provide this information. It is
> not on the file.
>
> See attached
>
> **Please respond to ncl-talk only. Do not include a salutation to any
> individual. Thank You**
>
>
> On Wed, Sep 17, 2014 at 5:50 PM, Jiaxin Zhang <jiaxinzhang3 at gmail.com>
> wrote:
>
>> Thank you so much, Dennis!
>> This works perfectly!
>>
>> I just have a couple of questions. I am trying to do this for different
>> times using the attached script and data file (for 12 months). For
>> simplicity, in the attached script I am testing for only one site (instead
>> of 3 sites). Although the script apparently extracts the variable values
>> for that grid in all 12 times, I guess there are some errors in the script
>> because the new netcdf created with the extracted variable values does not
>> have the time variable. What is necessary to modify in the attached script
>> so that the original time information is copied in the new created netcdf?
>>
>> Another issue I noticed is that when extracting variable values from 3
>> sites (3 grids) at the same time and after that creating a new netcdf,
>> apparently the new netcdf has only the lat and lon for the last site but
>> not for the rest of the extracted sites. Would it be possible to keep the
>> lat and lon information for all the sites from the original file in the new
>> created netcdf?
>>
>> Thank you!
>>
>> Jiaxin
>>
>>
>> 2014-09-17 5:57 GMT-07:00 Dennis Shea <shea at ucar.edu>:
>>
>>> Using getind_latlon2d should have failed. The 1st two arguments are
>>> prototyped as 2-dimensional. You are inputting one-dimensional arrays.
>>>
>>>
>>> function getind_latlon2d (
>>> lat2d [*][*] : numeric, <=== you are inputting lat[*]
>>> lon2d [*][*] : numeric, <=== lon[*]
>>> lat [*] : numeric,
>>> lon [*] : numeric
>>> )
>>>
>>> ==========
>>> Let NCL do it
>>>
>>> load "$NCARG_ROOT/lib/ncarg/nclscripts/csm/contributed.ncl"
>>>
>>> diri = "/Users/shea/Downloads/"
>>> fili = "cVeg_CanESM2_1time.nc"
>>> pthi = diri+fili
>>> fi = addfile(pthi, "r")
>>> ;print(fi)
>>>
>>> latpt = (/ 42.5, 25.4, 35.0 /)
>>> lonpt = (/-72.2,-81.1,-84.3 /)
>>> lonpt = where(lonpt.lt.0, lonpt+360, lonpt)
>>> npt = dimsizes(latpt)
>>>
>>> cVegpt= new ( npt, getfilevartypes(fi, "cVeg"), 1e20)
>>>
>>> do np=0,npt-1
>>> cVegpt(np) = fi->cVeg({latpt(np)},{lonpt(np)})
>>> end do
>>> print(cVegpt)
>>>
>>>
>>> On Tue, Sep 16, 2014 at 1:29 PM, Jiaxin Zhang <jiaxinzhang3 at gmail.com>
>>> wrote:
>>>
>>>> Dear NCL users,
>>>>
>>>> I am trying to follow the example provided for getind_latlon2d in:
>>>>
>>>>
>>>> http://www.ncl.ucar.edu/Document/Functions/Contributed/getind_latlon2d.shtml
>>>>
>>>> My data is in a rectilinear grid that can be completely described by
>>>> one dimensional lat and lon arrays, whereas getind_latlon2d requires 2d
>>>> arrays for lat and lon. How can I solve this problem? Is it better to use a
>>>> different function from getind_latlon2d for 1d lat and lon arrays?
>>>>
>>>> I am attaching my NCL script and my file. I have 3 different sites
>>>> (with given coordinates in attached script), that I would need to locate in
>>>> the data grid. After that I would need to extract the value in that
>>>> particular location (extract only the value of the 3 grids that contain the
>>>> 3 sites). I was using getind_latlon2d to locate the 3 points in the data
>>>> grid, but maybe there is better option to do that for rectilinear grids.
>>>>
>>>> Your help is very appreciated.
>>>> Thank you!
>>>>
>>>> Jiaxin
>>>>
>>>>
>>>> _______________________________________________
>>>> ncl-talk mailing list
>>>> 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/20140918/b772b55e/attachment-0001.html
-------------- next part --------------
A non-text attachment was scrubbed...
Name: extract.ncl
Type: application/octet-stream
Size: 2240 bytes
Desc: not available
Url : http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20140918/b772b55e/attachment-0001.obj
-------------- next part --------------
A non-text attachment was scrubbed...
Name: cVeg_CanESM2_1time.nc
Type: application/x-netcdf
Size: 35544 bytes
Desc: not available
Url : http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20140918/b772b55e/attachment-0002.nc
-------------- next part --------------
A non-text attachment was scrubbed...
Name: cVeg_CanESM2_12months.nc
Type: application/x-netcdf
Size: 396292 bytes
Desc: not available
Url : http://mailman.ucar.edu/pipermail/ncl-talk/attachments/20140918/b772b55e/attachment-0003.nc
More information about the ncl-talk
mailing list