Re: [ferret_users] Missing vector arrows

["William S. Kessler" <William.S.Kessler@xxxxxxxx>]

I have revised Ned Cokelet's poly_vector script that some people  
might like better (attached file polyvectors_scaled_arrows_5.jnl). A  
major difference is that the original script leaves the arrowheads  
the same length and just reduces the length of the vector tail for  
small magnitudes, which can result in hard-to-read triangular  
"vectors"; my revision scales the AREA of the entire vector, keeping  
the shape the same. The operation of the script is identical.

Another version (polyvectors_scaled_arrows_5.jnl) operates similarly  
except that it makes the vector tails just a line. This is something  
like the scattered_vectors script except, um, except, (not sure what  
the functional difference is, but it has the same operation as  
polyvectors_scaled_arrows_5.jnl, so you can switch between them on  
the fly).

Either version can be color-filled by some other quantity on the same  
grid or "transparent" on top of the background field.

For example of the output of these scripts, see:

version 5 for thick vectors:
http://www.pmel.noaa.gov/people/kessler/noumea/ORCA/orca-transport- 
harm-varellipse-plotvec-5c5-2060-fig-aaa.gif
http://www.pmel.noaa.gov/people/kessler/noumea/gliders/ 
mw9304_adcp_vectors1.gif

version 6 for line vectors:
http://www.pmel.noaa.gov/people/kessler/noumea/gliders/ 
mw9304_adcp_vectors1b.gif

If anyone improves these scripts or finds errors please post it to  
the list or let me know what you did.

Billy K

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
William S. Kessler
NOAA / Pacific Marine Environmental Laboratory
7600 Sand Point Way NE
Seattle WA 98115 USA

william.s.kessler@noaa.gov
Tel: 206-526-6221
Fax: 206-526-6744
Home page: http://www.pmel.noaa.gov/people/kessler/

Attachment: poly_vectors_scaled_arrows_5.jnl
Description: Binary data

Attachment: poly_vectors_scaled_arrows_6.jnl
Description: Binary data

On Nov 15, 2006, at 10:05 AM, Ansley Manke wrote:

> Hi Jerome,
> It looks like the algorithm for drawing the arrow part of the  
> vectors breaks down, making the arrow lie along the vector "shaft"  
> for some vectors. Why it seems to happen for some of the vectors in  
> your plots is not clear. We'll have to take another look at that.
>
> One workaround for you might be the polygon vector scripts, which  
> are available for both rectilinear and curvilinear data. If you  
> like the look of these, they use an entirely different method of  
> drawing, and should not have the trouble you're seeing.  They also  
> have the advantage of the option to color the vectors according to  
> some variable field if one wishes. Try running the script
> yes? go poly_vector_demo
> and see if you like the look. The demo uses a map projection to  
> create curvilinear data, but you can use the poly_vectors.jnl  
> script directly to plot curvilinear data; the script wants inputs  
> containing the x and y coordinates of the vector tails in 2  
> dimensions which is exactly what we have in a curvilinear data set.
>     ! say we have data with uwnd, vwnd, and curvilinear
>     ! coordinate variables lon_array and lat_array
>
> yes? use curvilinear_dat.nc
>
>     ! make a vector plot as the "underlay" for the polygon command
>     ! This sets up the plot scaling
> yes? vector/nokey uwnd, vwnd, lon_array, lat_array
>
> yes? go poly_vectors.jnl lon_array lat_array uwnd vwnd 5 "arrow"
>
>    ! These lines are written by the poly_vectors.jnl script:
> *** POLY_VECTORS: Issue commands such as follow to plot the vectors  
> ***
> *** POLYGON/OVER/NOLABEL/KEY/NOAXES/LINE PLT_X_ARROW, PLT_Y_ARROW,  
> MY_VALUES[J=1:27118] ***
> *** SET REGION SAVE ***
>
>    ! Create a constant variable to color the plots with a single color
> yes? let my_values = 1 + 0*ysequence(uwnd)
>
> yes? poly/pal=blue/over/nolab/key/noax/line  PLT_X_ARROW,  
> PLT_Y_ARROW, my_values
> Note that the polygon vector plots don't automatically skip some  
> vectors for plot readability; the polygon script discusses sub- 
> sampling if needed.
>
>
>
> Jerome Fiechter wrote:
> > Hello Ferreters,
> >
> > I just ran into a problem plotting vectors on a curvilinear grid  
> > where some, but not all, arrows at the end of the vectors are  
> > missing. Has anybody seen this before or knows how to fix it? Any  
> > help would be appreciated.
> >
> > Thank you,
> >
> > Jerome
> >
> >
> >
> > <mime-attachment.gif>
> -- ~> ~> ~> ~> ~> ~> ~> ~> ~> ~> ~> Ansley Manke, NOAA/PMEL  
> ansley.b.manke@noaa.gov 7600 Sand Point Way NE, Seattle WA Phone  
> 206-526-6246, FAX 206-526-6744