[ncl-talk] Question about the sign of shear deformation

[Dennis Shea]

I changed the documentation

               SHD = dv/dx + du/dy
               STD = du/dx - dv/dy

See: *shear_stretch_deform*
<http://www.ncl.ucar.edu/Document/Functions/Contributed/shear_stretch_deform.shtml>
and *shear_stretch_deform_cfd*
<http://www.ncl.ucar.edu/Document/Functions/Contributed/shear_stretch_deform_cfd.shtml>

*Thank You*
D


On Wed, May 8, 2019 at 1:47 AM Lyndz <olagueralyndonmark429 at gmail.com>
wrote:

> Dear Sir Dennis,
>
> Thank you so much for the help.
>
> One more point though in the documentation. I think this line should be
> corrected?
>
>                                  Less commonly used are the shear-deformation (SHD) and stretch-deformation (STD).
>
>                SHD = du/dx - dv/dy
>                STD = dv/dx + du/dy
>
> to
>
>                STD = du/dx - dv/dy
>
>                SHD = dv/dx + du/dy
>
>
> Sincerely,
>
> Lyndz
>
>
> On Wed, May 8, 2019 at 6:02 AM Dennis Shea <shea at ucar.edu> wrote:
>
>> This could well be a matter of definition.  NCL does not include any
>> explicit negative (or positive) sign or coefficient (eg:  0.5).
>>
>>
>>
>> *  shear  = dvdx + dudy  stretch= dudx - dvdy   deform = sqrt(shear^2 +
>> stretch^2)*
>>
>> where the gradients are determined by spherical harmonics.
>>
>> You can associate signs/coefficients to these terms  as you wish.
>>
>> ====
>> You have
>> ************************************************
>> ; Stretch, deform, and total
>> ;************************************************
>>   grdType = 1
>>   *cyclic  = False       ; <========  This should be True*
>>   opt_ssd = 0
>>
>>   ssdList = shear_stretch_deform(u,v,u&lat,grdType,opt_ssd)
>>
>>   stretch_sph = ssdList[0]
>>   shear_sph   = ssdList[1]
>>   deform_sph  = ssdList[2]
>>   delete(ssdList)
>>
>> ======================================================================
>> I don't think anyone wants to download a 600+MB file from the cloud.
>>
>> Use the netCDF operator the extract the exact subset
>>
>> *%>* *ncks* -O -d Time,tStrt,Tlast ncep_6h_jan.nc  NECP.deform.nc
>>
>> where tStrt and tLast are the temporal index values of the specified time
>> interval.
>>
>> ===
>> The following is 'ok'.
>>
>>   u   = f->U_GRD_2_ISBL(:,:,:,:)       ; pull u off file
>>   v   = f->V_GRD_2_ISBL(:,:,:,:)       ; pull v off file
>>
>> However, when reading an entire array, it is more efficient to use the
>> following. You can indicate dimensionalit in comment
>>
>>   u   = f->U_GRD_2_ISBL            ;         (:,:,:,:)       ; pull u off
>> file
>>   v   = f->V_GRD_2_ISBL             ;         (:,:,:,:)       ; pull v
>> off file
>>
>> ===
>> The following is superfluous
>>
>>   lat = f->lat
>>   lon = f->lon
>>
>> These coordinates are part of each variable's structure
>>
>>    u&lat     ===> lat
>>
>> ===
>> What order are the variables? N-S or S-N ? The following reverses the
>> order.
>>
>>   u   = u(:,:,::-1,:)
>>   v   = v(:,:,::-1,:)
>>
>> On Mon, May 6, 2019 at 1:34 AM Lyndz <olagueralyndonmark429 at gmail.com>
>> wrote:
>>
>>> Dear NCL experts,
>>>
>>> I am calculating the shear_stretch deformation routine in NCL (
>>> https://www.ncl.ucar.edu/Document/Functions/Contributed/shear_stretch_deform.shtml
>>> ).
>>>
>>> According to the description:
>>>
>>> *"When the stretch-deformation is negative/positive the flow is
>>> confluent/diffluent."*
>>>
>>> However, I encounter some inconsistensies in the results.
>>>
>>> Problems:
>>>
>>> [1] When I plot the stretch deformation term, I see positive values
>>> (diffluence) along a region of confluence (encircled region in red in the
>>> attached figure).
>>> This plot is for timestep 62. Shading is for the 925-hPa stretch
>>> deformation, while the streamline is for the 925-hPa winds.
>>>
>>> I am not sure if I implemented this routine correctly. I am attaching my
>>> script in this email. The file that I am using can be accessed here:
>>>
>>> https://www.dropbox.com/s/m4sy7n76piw2mwn/ncep_6h_jan.nc?dl=0
>>>
>>> I would like to ask for suggestions on how to do this correctly in NCL.
>>>
>>> [2] I would like to draw line along the regions of positive stretch or
>>> total deformation. If yes, how can I do this in NCL?
>>>
>>> Thank you in advance for the help.
>>> Sincerely,
>>>
>>> *Lyndz*
>>>
>>> _______________________________________________
>>> 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/20190508/0f551335/attachment.html>