[ncl-talk] plotting I J variables from the WRF out file

Tue May 12 08:44:23 MDT 2020

I'd seen that warning from stringtofloat before, working with your file,
and did not know where it came from. Turns out it's due to there being 1
extra newline at the end of your file. I suspect the other errors result
from subsequent code trying to use the missing value as an index.

Rick

On Tue, May 12, 2020 at 1:13 AM ali mughal <mughalali655 at gmail.com> wrote:

> Ok in continuation to my previous email I did some experiments to write 2m
> air temperature and Lu_index to a csv file using the I,j values from
> zero_skipped.csv. Don't know how good they are but in both scripts I am
> getting the following error
>
> warning:stringtofloat: a bad value was passed; input strings must contain
> numeric digits, replacing with missing value
> warning:stringtofloat: a bad value was passed; input strings must contain
> numeric digits, replacing with missing value
> warning:stringtofloat: a bad value was passed; input strings must contain
> numeric digits, replacing with missing value
> warning:toint: there are 1 floats larger than INT_MAX, which has been
> flagged missing.
> fatal:["NclFile.c":2099]:Subscript out of range, error in subscript #1
> fatal:Memory allocation failure:[errno=12]
> fatal:["Execute.c":8578]:Execute: Error occurred at or near line 76 in
> file 2m_T.ncl
>
> In addition the LCZ.csv script is not printing the lu_indexes for these
> locations.
>
> Could anyone suggest any changes to the code attached ?
>
> Regards
>
> On Tue, May 12, 2020 at 11:53 AM Rick Brownrigg <brownrig at ucar.edu> wrote:
>
>> Hi,
>>
>> Please always respond back to the mail-list -- there's a much broader
>> pool of expertise that can contribute.
>>
>> I don't know what you mean by "one extra value at the end".  Is that one
>> too many lat/lon pairs?  A statement like loc=loc-1 would effectively
>> subtract 1 degree from all lat/lon pairs, thus translating your data. Do
>> you see anything plotted if you don't do this?
>>
>> I don't know anything about the 2m air temperature, and this is where the
>> broader community comes into play.
>>
>> Rick
>>
>>
>>
>> On Mon, May 11, 2020 at 9:42 PM ali mughal <mughalali655 at gmail.com>
>> wrote:
>>
>>> Hi
>>> Thanks a lot for the help.
>>>
>>> Please see the attached corrected file which shows the data from
>>> zero_skipped.csv. However, there is one  extra value at the end of both lat
>>> and lons. When I use loc=loc-1 I don't see any data plotted. Please let me
>>> know if my procedure is correct.
>>>
>>> Second question is how can I extract the 2m air temperature at the same
>>> grid points as the values in zero_skipped.csv.
>>>
>>> Will be really grateful for the help
>>>
>>> On Tue, May 12, 2020 at 9:19 AM Rick Brownrigg <brownrig at ucar.edu>
>>> wrote:
>>>
>>>> Hi,
>>>>
>>>> I'm not surprised that no data is plotted. Again, you are treating the
>>>> 1st and 2nd columns of your zero_skipped file as lon/lat values. They look
>>>> like i,j indices to me, but conceivably could be lon/lats at 1 degree
>>>> boundaries. Regardless, they are values that are well outside the narrow
>>>> mpMin/Max/Lat/Lon values you've specified.
>>>>
>>>> I don't understand the lat_lons.ncl script, and can't run it because I
>>>> don't have the WRF file. However, if I understand what you are trying to
>>>> do, I think you need to open that WRF file in test_1.ncl, like you did in
>>>> lat_lons.ncl, and after you've read in and converted the variables "lon"
>>>> "lat", pass those into wrf_user_ij_ll, and use the results of that to set
>>>> the sfXArray/sfYArray resources. For clarity, I personally would rename the
>>>> variables lat/lon as read in from the ascii file to something more like "i"
>>>> "j" or "row" "col"
>>>>
>>>> Rick
>>>>
>>>> On Mon, May 11, 2020 at 5:29 PM ali mughal via ncl-talk <
>>>> ncl-talk at ucar.edu> wrote:
>>>>
>>>>> Dear NCL Team
>>>>>
>>>>> For the first question regarding modification to the code I did the
>>>>> attached changes (i.e. test1.ncl) though it shows the map where I intend to
>>>>> plot the data but still there is no data plotted. The I, j values in the
>>>>> csv I shared earlier and repeated here for convenience (zero_skipped.csv)
>>>>> correspond to the lat longs in the file (WRF_latlongs.xlsx).
>>>>>
>>>>> For the second question regarding the usage of wrf_user_ij_ll when I
>>>>> try to use this function like the way in the attached file (lat_lons.ncl) I
>>>>> get a segmentation fault (core dumped) message. However, if I don't use the
>>>>> script and copy these values directly into the ncl interface instead of
>>>>> submitting it in the form a script I get the values printed out. Please
>>>>> note that the I,j values printed here correspond to the same lat longs as
>>>>> in the file (WRF_lotlongs.xlsx).
>>>>>
>>>>> Regards
>>>>>
>>>>> On Tue, May 12, 2020 at 3:15 AM Dennis Shea <shea at ucar.edu> wrote:
>>>>>
>>>>>> Actually, it is not quite clearer what you are looking for:
>>>>>>
>>>>>> As indicated by Rick  *wrf_user_ij_to_ll*
>>>>>> <https://www.ncl.ucar.edu/Document/Functions/WRF_arw/wrf_user_ij_to_ll.shtml>
>>>>>>
>>>>>> other possibilities include the following very similar functions:
>>>>>>
>>>>>> *region_ind*
>>>>>> <https://www.ncl.ucar.edu/Document/Functions/Contributed/region_ind.shtml>
>>>>>> and  *getind_latlon2d*
>>>>>> <https://www.ncl.ucar.edu/Document/Functions/Contributed/getind_latlon2d.shtml>
>>>>>>
>>>>>> *You* must experiment. Please use print statements like *printVarSummary
>>>>>> *while debugging.
>>>>>> The output from *printVarSummary *is for you to look at to examine
>>>>>> variable type and shape, etc.
>>>>>>
>>>>>> On Sun, May 10, 2020 at 7:45 PM Rick Brownrigg via ncl-talk <
>>>>>> ncl-talk at ucar.edu> wrote:
>>>>>>
>>>>>>> Hi,
>>>>>>>
>>>>>>> Please do not reply directly back to me, but rather to the group as
>>>>>>> a whole -- they provide a broader pool of expertise to draw upon.
>>>>>>>
>>>>>>> I *think* wrf_user_ij_to_ll work() will do what you want, but it
>>>>>>> will likely require some experimentation. Liberal use of printVarSummary()
>>>>>>> is your best friend ;-)
>>>>>>>
>>>>>>> Rick
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> On Sun, May 10, 2020 at 3:03 PM ali mughal <mughalali655 at gmail.com>
>>>>>>> wrote:
>>>>>>>
>>>>>>>> Dear Rick
>>>>>>>>
>>>>>>>> Thanks for the reply.
>>>>>>>>
>>>>>>>> In continuation to my previous email, I asked that in the shared
>>>>>>>> the first two columns represent I,j values of the WRF grid. If i have to
>>>>>>>> evaluate 2m air temperature on the same I,j values will wrf_user_ij_to_ll
>>>>>>>> work or do I need to use some other method.
>>>>>>>>
>>>>>>>> Regards
>>>>>>>>
>>>>>>>> On Sunday, May 10, 2020, Rick Brownrigg <brownrig at ucar.edu> wrote:
>>>>>>>>
>>>>>>>>> Hi,
>>>>>>>>>
>>>>>>>>> There are at least two problems. One is that str_get_field()
>>>>>>>>> returns strings, and so "data" in the call to gsn_csm_contour_map() is a
>>>>>>>>> string, and this is the source of the error message. Try adding these 3
>>>>>>>>> lines:
>>>>>>>>>
>>>>>>>>> lat := stringtofloat(lat)
>>>>>>>>> lon := stringtofloat(lon)
>>>>>>>>> data := stringtofloat(data)
>>>>>>>>>
>>>>>>>>> The other problem is that the min/max bounds are inappropriate for
>>>>>>>>> the data -- I get a blank plot. If I used the true min/max I get a mostly
>>>>>>>>> centered over Siberia:
>>>>>>>>>
>>>>>>>>> res at mpMinLatF = min(lat) ;-6.080154 ; range to zoom in on
>>>>>>>>> res at mpMaxLatF = max(lat) ;10.04061
>>>>>>>>> res at mpMinLonF = min(lon) ;95.908
>>>>>>>>> res at mpMaxLonF = max(lon) ;112.092
>>>>>>>>>
>>>>>>>>> Finally, the values for lat/lon in the file look suspect -- is
>>>>>>>>> this really a 1-degree grid? I might have guessed they are row/column
>>>>>>>>> values.
>>>>>>>>>
>>>>>>>>> Good luck,
>>>>>>>>> Rick
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> On Sun, May 10, 2020 at 2:23 AM ali mughal via ncl-talk <
>>>>>>>>> ncl-talk at ucar.edu> wrote:
>>>>>>>>>
>>>>>>>>>> Dear Users
>>>>>>>>>>
>>>>>>>>>> I have csv file attached with the data available at specific
>>>>>>>>>> grids of the WRF outfile.Could any one help how to correct my attached code
>>>>>>>>>> in NCL?
>>>>>>>>>>
>>>>>>>>>> When I try to run the code, I get the following error message
>>>>>>>>>>
>>>>>>>>>> (0)     This file has 4825 rows and 3 columns.
>>>>>>>>>> fatal:Argument type mismatch on argument (1) of
>>>>>>>>>> (gsn_csm_contour_map) can not coerce
>>>>>>>>>> fatal:["Execute.c":8578]:Execute: Error occurred at or near line
>>>>>>>>>> 64 in file test_1.ncl
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> Later I want to use the same code to be modified to calculate 2m
>>>>>>>>>> air temperature at the same I J grids.
>>>>>>>>>>
>>>>>>>>>> Regards
>>>>>>>>>> _______________________________________________
>>>>>>>>>> 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
>>>>>>
>>>>>> _______________________________________________
>>>>> 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/20200512/be312744/attachment.html>