[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[IDV #LQF-598273]: FW: track display

Hi Bill-

> It was really good to see you at the Storm Conference. You provided lots
> of information to
> the participants, especially me.

I thought it was the best conference in terms of my participation and
interaction with people.

> I was able to access the topographic data on the GEON web site,
> http://geon.unavco.org/unavco/IDV_for_GEON.html ,
> and it looks great. Accessing the file via a URL in the IDV did not work
> well, but downloading the file
> to my hard drive and accessing it through the file system was fast and
> smooth.

Did you get and error or was it just slow?

> I made a bundle file,
> http://apollo.lsc.vsc.edu/~fingerhutb/IDV_bundles/Menu.xidv , to access
> my index.html file.
> And used it in class. Works better.

Okay.

> In class, I had my Dynamics students use the  Geostrophic: Wind,
> Relative Vorticity, & Vorticity Advection
> <http://apollo.lsc.vsc.edu/%7Efingerhutb/IDV_bundles/GeoWind.xidv>
> link
> since we were talking about geostrophic vorticity. That reminded me that
> we were going to add more vector math
> capabilities to the IDV. I am still interested in using this if you can
> get to it.

I know I keep saying it, but it's next on the list.  It's just
finding time to do it that is the problem.

> This semester I am teaching Numerical Weather Prediction, and felt the
> urge to use Lorenz's model, which is
> described nicely at http://en.wikipedia.org/wiki/Lorenz_attractor . The
> output is essentially a 3-d trajectory. When
> I looked for an application to graph the output, all I could find was
> (drum roll, please) the IDV. So, I coded the model
> in a spreadsheet, saved the data in an ascii file, created a cdl file,
> created a netcdf file, read the data into the IDV
> and created a display. Attached is a zipped bundle file, and I created a
> neat animation which is at
> http://apollo.lsc.vsc.edu/~fingerhutb/IDV_bundles/lorenz.mov .
> 
> I would have never figured out how to do this without the GEON web site,
> so I am appreciative of their work. However,
> they could have done a better job of explaining the steps in this
> process. In addition, the process of getting my data
> into an IDV compatible format is 'trying'. Does it make sense to add a
> page to the IDV users manual on how to
> get ascii data into the IDV ? (BTW, once before Tom Whitaker helped me
> read point data into the IDV from an
> ascii file). 

yes.  We've added a section on getting point data in, but I should
add Tom's notes from the steering committee forum as well for
other data.

> And does it make sense to just add code that reads ascii
> files directly ? 

Yes.  The problem is what format do you support?  Jeff has recently
added a widget that allows you to read in point data and save
the configurations so you can apply it to other files.  For grids,
there are several "standard" formats - ArcGrid ASCII, Surfer, etc.
But, someone has to write the code and right now, there's no one
to do that work.

Wasn't one of the goals to
> overcome problems associated with different data formats ?

Yes.  But, every time we add support for one new format, 5 others
crop up.
 
> Getting back to the attached bundle and the display it creates, I have a
> few comments.
> o The default track was 2 pixels wide, and did not display properly.
> Changing it to 1 pixel fixed the problem. You might
> try changing it back to 2 pixels and see what happens.
> o This was the second situation where I would have liked to have had
> an auto-rotate option, as in the globe
> display, to make a nice 3-d animation in a map display window.
> Any chance such an option could be added ?
> (Thus, I used the globe display to make a movie.)

it's there now.

> o I would have liked to color my track by the time, but that option
> is not available. So, I used longitude, which is available.
> Any chance that you could add time ?

We'll look into this.

> o It also occured to me that it must be possible to do all of this
> directly. I am pretty confident that I could write a
> jython method that runs the model. This would create arrays of
> time, latitude, longitude, and altitude. And I would guess
> that there is an already a method that creates a track data
> object, similar to DerivedGridFactory.createTrueWindVectors .
> If I knew its name and how to call it, I could create the same
> results much mor efficiently. Can you tell me how
> do this or where to look for information ?

There are hooks in the track stuff already to have time (that's
how we do the subsetting by time).  It's just a matter of bubbling
it up to the user.

> Of course, if you (or Jeff or Tom or ???) have any ideas/suggestions
> with regards to these endeavors please pass them along.

I'm always amazed at what you come up with that the IDV can do so
don't have any suggestions at this point.

Don

Ticket Details
===================
Ticket ID: LQF-598273
Department: Support IDV
Priority: Normal
Status: Open