[ncl-talk] Issues With EP Flux Calculation

[Joshua Walston]

Thank you both for taking the time to respond to my issue. I greatly
appreciate your feedback. I will keep plugging away and hopefully find a
solution.

Josh




On Thu, Dec 8, 2016 at 5:59 PM, Dennis Shea <shea at ucar.edu> wrote:

> You are correct, there are issues for the ERA_Interim data. Another user
> had issues with the JRA data set. Even though the original code was written
> by a NOAA person,  I see no reason why the function should only work with
> NCEP reanalysis data.
>
> I cleaned up your code and used an updated 'epflux' function. Nothing
> substantive was changed BUT I enabled much more printing. This enabled me
> to do a 'side-by-side' with the NCEP results. Basically, all the variables
> look somewhat consistent *except* the
>
> Vertical divergence : min=-252582   max=249489
>
> which was *MUCH* larger than the NCEP values.
>
> That is the current status.
>
>
>
> On Wed, Dec 7, 2016 at 7:24 AM, Rick Brownrigg <brownrig at ucar.edu> wrote:
>
>> Hi Josh,
>>
>> I've not had a chance to look at your script, but that "Assignment type
>> mismatch..." error message always means one thing:  the variable on the
>> lefthand side (lhs) has been defined elsewhere prior to the assignment, and
>> it has a certain type, shape, and dimensions. The righthand side variable
>> or expression must match in all regards -- type, shape, dimension. In your
>> case, it doesn't.
>>
>> If you simply want to reuse a variable name on the lhs, you can use the
>> reassignment operator ":=", which is an explicit indication by you the
>> programmer that you don't care that the lhs gets stomped on; eg.:
>>
>>      foo = (/ 1, 2, 3, 4 /)
>>      foo = "A string, not an 1D integer array"     ; assignment YIELDS AN
>> ERROR, not same type, shape, or dimension
>>      foo := "A string, not an 1D integer array"    ; RE-assignment is OK.
>>
>> If simple reuse is not what you want, try putting in some print
>> statements to make sure you know what lhs and rhs are/should-be.
>>
>> Hope that helps...
>> Rick
>>
>> On Tue, Dec 6, 2016 at 1:13 PM, Joshua Walston <joshwalston at gmail.com>
>> wrote:
>>
>>> Hi NCL Users,
>>>
>>> I am attempting to use the Eliassen-Palm flux function via the
>>> ncl_contributed functions for the upcoming ncl_6.4 version. There seems to
>>> be an error in my calculations as the output figures seem odd. My script is
>>> running fine with no warnings, however the divergence and flux vectors seem
>>> incorrect. Can anyone see what I am doing wrong?
>>>
>>> Additionally, I have a 6 hourly data set that I would like to calculate
>>> the flux for (i.e. would prefer to not have a time average). Is it possible
>>> to loop through the data and get a return value of type "list" for each
>>> time, so I can plot the flux at each 6 hour time step. If so, how? When I
>>> attempt I receive a warning that states:
>>>
>>> "fatal:Assignment type mismatch, right hand side can't be coerced to
>>> type of left hand side"
>>>
>>> I have placed 3 figures (00z, 06z and the average over all times), my
>>> script, and the data (ERA40 Interim) here:
>>>
>>> <http://goog_1241068196>
>>> ftp://pubfiles.dri.edu/pub/jwalston/
>>>
>>> Thank you in advance for your help,
>>> Josh
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>> _______________________________________________
>>> 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/20161209/161a9b9d/attachment.html