[ncl-talk] numerical differences from NCL

[Xi Chang]

Thanks guys,

solved it !

*(1/p)*dTheta/d(lnp) (approximate)*

(0) min=-0.0184797   max=0.00307985


*dTheta/dp (direct)*

(0) min=-0.0149857   max=0.00320144


They are similar! thanks

Xi Chang

On Wed, Aug 17, 2016 at 8:06 PM, Dennis Shea <shea at ucar.edu> wrote:

>          y = ln(p)
>   dy/dp = 1/p
>
> so .... maybe
>
> *p*dTheta/d(lnp),*
>
> On Wed, Aug 17, 2016 at 8:32 AM, Xi Chang <xi.chang01 at gmail.com> wrote:
>
>> Hallo,
>>
>> I tried to calculate dTheta/dp using center finite difference with the
>> folllowing methods:
>>
>> 1. direct calculation of *dTheta/dp* using center finite difference
>>
>> 2. using this approximation  * (1/p)*dTheta/d(lnp), *also using center
>> finite difference.
>>
>> BUT surprisingly this gives me different results, for instance the method
>> 1 give ma value 100 K/Pa but the method 2 gives me much smaller values
>> 0.0xxx K/Pa.
>>
>> Thanks for suggestion.
>>
>> Chang
>>
>> _______________________________________________
>> 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/20160817/cc97a689/attachment.html