Date: Tue Nov 12 2013 - 13:19:42 MST
Try the attached. It uses an alternative approach to computing
the vertical integral. [I am still not sure why the script gets
a fatal message using vibeta.]
=======
Variable: fint
Type: float
Total Size: 47040 bytes
11760 values
Number of Dimensions: 3
Dimensions and sizes: [time | 12] x [lat | 28] x [lon | 35]
Coordinates:
time: [36515.5..36849.5]
lat: [-33.75..33.75]
lon: [ -80.. 5]
Number Of Attributes: 3
units : 10^7 kg s-1
long_name : vertically integrated divergent moisture flux
_FillValue : 1e+20
(0) fint: min=-2.67179 max=2.67318
========
Please note that there are numerous grid points that have _FillValue.
(0) p=100000 nmsg_q=355
(0) p=92500 nmsg_q=99
(0) p=85000 nmsg_q=47
(0) p=70000 nmsg_q=22
(0) p=60000 nmsg_q=0
(0) p=50000 nmsg_q=0
(0) p=40000 nmsg_q=0
(0) p=30000 nmsg_q=0
These occur in the Andes Mts.
On 11/12/13 11:16 AM, Guilherme Martins wrote:
> Hi Dennis,
>
> My domain is: Lon: -80 5 and lat: -35 35
>
> There are missing values in my dataset (u,v,q, and p).
>
> In attach I send a piece of my dataset and my script. I was searching for
> but I don't find anything.
>
> Thanks,
>
> Guilherme
>
>
> 2013/11/12 Dennis Shea <shea@ucar.edu>
>
>> Your pressure levels are:
>>
>> 100000,92500,85000,70000,60000,50000,40000,30000
>>
>>
>> I speculate that over the Himalaya region, certain
>> grid points have missing (_FillValue) data at 50000
>> and below (100000).
>>
>> The messages are 'warning' messages. I don't understand the 'fatal'
>> message.
>>
>>
>> fatal:vibeta: ier = -999
>>> fatal:["Execute.c":8128]:Execute: Error occurred at or near line 44 in
>> file
>>
>> The underlying code has:
>>
>> c make sure there are at least several levels.
>>
>> if (nlmax.le.3) then
>> ier = -999
>> vint = xmsg
>> return
>> endif
>>
>> This is the only place the -999 occurs.
>>
>>
>>
>>
>>
>> On 11/12/13 7:00 AM, Guilherme Martins wrote:
>>
>>> Hi Dennis,
>>>
>>> Now, when I run the script appear:
>>>
>>> warning:vibeta: there must be at least three levels with data above the
>>> surface
>>> warning:vibeta: there must be at least three levels with data above the
>>> surface
>>> warning:vibeta: there must be at least three levels with data above the
>>> surface
>>> warning:vibeta: there must be at least three levels with data above the
>>> surface
>>> fatal:vibeta: ier = -999
>>> fatal:["Execute.c":8128]:Execute: Error occurred at or near line 44 in
>>> file
>>> fluxo.q.ncl
>>>
>>> Guilherme.
>>>
>>>
>>> 2013/11/12 Dennis Shea <shea@ucar.edu>
>>>
>>>
>>>> Assuming pc(time,lat,lon) and divc(time,lev,lat,lon)
>>>>
>>>> then ....
>>>>
>>>> fint = vibeta(p,divc(time|:,lat|:,lon|:,lev|:),linlog,pc,pbot,ptop)
>>>>
>>>>
>>>> On 11/11/13 10:28 PM, Guilherme Martins wrote:
>>>>
>>>> Hi all,
>>>>>
>>>>> I'm using the vibeta function, but when I run my script I get the error:
>>>>>
>>>>> fatal:vibeta: If psfc is not a constant, then its dimension sizes must
>>>>> be
>>>>> equal to all but the last (rightmost) dimension sizes of x
>>>>> fatal:["Execute.c":8128]:Execute: Error occurred at or near line 43 in
>>>>> file
>>>>> fluxo.q.ncl
>>>>>
>>>>> The NCL told me the error, but I can't understand it. Can anyone help
>>>>> me?
>>>>>
>>>>> My script is in attach.
>>>>>
>>>>> My ncl is 6.1.2.
>>>>>
>>>>> Thanks,
>>>>>
>>>>> Guilherme.
>>>>>
>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> 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
- text/plain attachment: fluxo.q.ncl