# Re: Calculating moisture transport

From: Guilherme Martins <jgmsantos_at_nyahnyahspammersnyahnyah>
Date: Mon Nov 25 2013 - 11:05:37 MST

Hi Dennis,

I want the vertical integral, but I can't understand when you talk about
the *weight* by the layer thickness.

I calculated the dp:

ptop = 300.
linlog = 1
*dp = dpres_plevel_Wrap(p, psfc, ptop, 0)* ; [time | 12] x [level | 7]
x [lat | 29] x [lon | 97]. *Here, it's ok*

I don't know how divide *uqW_total (t) *by *dp* (t,z,y,x). They have
different dimensions, I used the conform function, but it return error.

*uqW_total = dim_sum_n_Wrap(uqW,(/1,2/)) * dxdy* ; var(t)

Thanks,

Guilherme.

2013/11/25 Dennis Shea <shea@ucar.edu>

> What was sent (eg:
> uqW = dim_sum_n_Wrap(uq(:,:,{latS:latN},{lonW:lonW}),3)
>
> The {latS:latN} corresponds to eqn 3 without the 'dx'
>
> dimensions are (time,lev,lat) [0,1,2]
>
> Then you use
>
> uqW_total = dim_sum_n_Wrap(uqW,(/1,2/)) * dxdy
>
> If you want a vertical integral, then you *must* weight by the
> layer thickness. Not a simple sum. The denominator would
> be the total layer thickness.
>
> http://www.ncl.ucar.edu/Document/Functions/Built-in/dpres_plevel.shtml
> or
> http://www.ncl.ucar.edu/Document/Functions/Contributed/dpres_plevel_Wrap.
> shtml
>
>
>
>
>
>
>
>
> On 11/25/13 3:10 AM, Marcos Tonelli wrote:
>
>> Dear Guilherme,
>>
>> again, your results will depend on the extent of each border (which could
>> have been shown in your email, rather than just in your .ncl).
>> How big are your results?
>>
>>
>>
>> On Fri, Nov 22, 2013 at 5:54 PM, Guilherme Martins <jgmsantos@gmail.com
>> >wrote:
>>
>> Hi all,
>>>
>>> I'm trying calculate the moisture transport using NCEP Reanalysis1 data
>>> using the equations 3 and 4 below. I know this data aren't good to
>>> calculate this flux.
>>>
>>> Levels used: 1000,925,850,700,600,500,400 and 300hPa. The integration is
>>> done between 1000 and 300hPa.
>>>
>>> Dennis gave me some tips but I'm doing something wrong.
>>>
>>> The results normally are: 15-20 10^8 kg/s.
>>>
>>> I send my script and my dataset.
>>>
>>> My ncl is 6.1.2.
>>>
>>> Thanks,
>>>
>>> Guilherme.
>>>
>>> [image: Imagem inline 1]
>>>
>>>
>>>
>>> --
>>> *Guilherme Martins*
>>>
>>>
>>> _______________________________________________
>>> 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
>>
>>

```--
*Guilherme Martins*