Re: Streamline Density

• To: James Kelly <J.Kelly@xxxxxxxxxx>
• Subject: Re: Streamline Density
• From: Tom Rink <rink@xxxxxxxxxxxxx>
• Date: Mon, 17 Apr 2006 07:43:17 -0500

James,

better.  Otherwise, if someone is willing to port the code to
Java, I'll integrate it into VisAD core.

TomR

James Kelly wrote:

Thanks Tom,

I was hoping that the McIDAS streamlines and the VisAD ones were the same, and the latter just required a bit of tweaking.

The McIDAS streamlines certainly look better as the density improves (and if one forces the density of the VisAD streamlines to be greater than 2.0, then strange things happen such as streamlines crossing each other... probably why the density is limited to 2.0 in the code!).

So in the longer term, we'd be interested discussing getting the your (McIDAS) routine into VisAD.

Cheers,

James

Tom Whittaker wrote:

I cannot speak for VisAD, however in McIDAS the algorithm for drawing
streamlines is the one described in:  Monthly Weather Review: Vol.
105, No. 6, pp. 786–788.

The "density" value is used to determine in what grid "boxes"
streamlines can start.  Highest density says that a streamline will
start in every grid box and be drawn to its termination (rules for
stopping drawing are described in the paper).  Streamlines are always
started in the center of a grid box, thus no more than one will ever
be started in a box -- that does not, of course, prevent more than one
streamline from passing through a grid box.

Hope that helps.

tom

On 3/29/06, James Kelly <J.Kelly@xxxxxxxxxx> wrote:

Hi,

I've been looking at the quality of streamline analyses (of atmospheric
wind fields) lately.

I was experimenting with streamline displays in the McIDAS system, and

thought (IMHO) that the streamline analyses looked better as the density

of the streamlines increased.

I then tried to increase the density of the streamlines in a visad
display, and but wasn't able to generate the same density as the McIDAS
display.
Should I be able to do this? If so how?

Tried changing "streamlineDensity" from the default of 1.0 to higher
values of 10.0, 100.0, 1000.0 and 1000.0, but didn't notice much change
beyond 100.0.

Further details:
1) In McIDAS I did this:
GRDDISP <modelname> FHOUR=0 PARAM=STREAML LEV=SFC  LAT=-20 20 LON=-160
-100 CINT=0 COLOR=15

Image is attached as "mcidas_streamline.png"

2) In Visad/jython I did this:

maps = (xMap, yMap, rngUMap, rngVMap)

disp = subs.makeDisplay(maps)
flowcontrol = rngUMap.getControl()
flowcontrol.enableStreamlines(TRUE)
strmDensity = flowcontrol.getStreamlineDensity()
stepFactor = flowcontrol.getStepFactor()
flowcontrol.setStreamlineDensity(10000.0)
print " streamline density " , strmDensity, " stepFactor ", stepFactor

drLAPSuv=subs.addData("windbarbs", uv, disp)

Image is attached as "visad_streamline.png"

Cheers,

James


--
James Kelly
Australian Bureau of Meteorology

--
Tom Whittaker
University of Wisconsin-Madison
SSEC/CIMSS
1225 W. Dayton Street
Madison, WI  53706  USA
ph: 608-262-2759

===============================================================================

To unsubscribe visad, visit:
http://www.unidata.ucar.edu/mailing-list-delete-form.html

===============================================================================

==============================================================================
To unsubscribe visad, visit:
http://www.unidata.ucar.edu/mailing-list-delete-form.html
==============================================================================

• References:
  • Streamline Density
    • From: James Kelly
  • Re: Streamline Density
    • From: Tom Whittaker
  • Re: Streamline Density
    • From: James Kelly