Re: Vector plotting on vertical sections

[Ansley Manke <ansley.b.manke@xxxxxxxx>]

Hi all,
This looks like a bug in the 4-argument vector plotting, where the
arrowheads aren't properly scaled when the aspect ratio correction
is large.  We're looking into it.

Ansley Manke

john.hunter@utas.edu.au wrote:

> Hello All,
>
> I have a problem plotting current vectors on a vertical section. 
> (There is a minor complication in that the model (POM) uses 
> sigma-coordinates -- however I calculate the true vertical coordinate in 
> Ferret, such that scaling using "shade" or "contour" works fine; so I do 
> not think my problem is related to the use of sigma-coordinates.)
>
> For a typical vertical section, the horizontal extent is about 800000
> metres and the vertical extent is about 1000 metres, so a vertical
> expansion of around 800 is entailed in the plotting.
>
> This is what I do:
>
> 1. I use the default aspect ratio (axis length ratio = 0.75)
> by calling:
>
>   set window <window number>
>
> 2. I plot "shade" or "contour" first
>
> 3. I then do a vector overlay using EITHER:
>
>   vector/overlay/aspect=`ydx` .......
>     (where ydx ~ 800, but calculated from the exact horizontal and
>     vertical extents)
>
>   OR:
>
>   vector/overlay .......
>
>   Both of these produce vectors which point in the correct direction
>   relative to the upper and lower boundaries of the model (i.e. the
>   surface and sea bed), and so I assume they are correct. The results
>   of both these options also appear to be identical. Any other 
>   selection of "ydx" in the first of the above options causes the 
>   vectors to point in what is clearly the wrong direction.
>
>   HOWEVER in both of the above ("correct") options the arrow heads
>   appear to have been distorted by the vertical expansion, to form
>   vertical lines (like "hammer heads" on the end of each vector)
>   -- it appears as if the vectors AND their arrow heads have been
>   "drawn" on a 1:1 vertical:horizontal scaling, and THEN the whole
>   plot has been expanded in the vertical by a factor of ~800.
>
> How do I fix this? I am sure I have seen Ferret plots of vertical
> sections where this works O.K., so I am sure there is a simple
> solution.
>
> Cheers,
>
> John
>
> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
> + John Hunter,                                Tel. (work) 03 6226 7849 +
> + Antarctic Climate & Ecosystems              Tel. (home) 03 6234 8391 +
> + Cooperative Research Centre,                Mobile      0427 098 831 +
> + University of Tasmania,                            Fax. 03 6226 2973 +
> + Private Bag 80,                 email (work) john.hunter@utas.edu.au +
> + Hobart, Tasmania 7001,          email (private) jhunter@trump.net.au +
> + Australia                    http://www.antcrc.utas.edu.au/~johunter +
> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
>
>