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

[IDV #WHS-790529]: Issue with bottom velocity vectors



Hi Rich,

I have been revisiting some old issues over the last few days; things that were
left dangling over the last while (months, even years) but did not have time to
get back to.

At any rate, I played around with this specific issue over the weekend. The core
problem is the IDV does not handle the vertical coordinate system (bathymetry,
s_rho) properly which leads to those visualization artifacts of missing swaths
of data. The data appear to get smushed into the earth surface. You notice this
when doing a side view of the data. It does not have anything to do with the
data (why would it as Python demonstrates), nor is it a VisAD issue.

Please see the link below where I give a fake and exaggerated vertical
coordinate system in 'fixed.ncml', and 'fixed.xidv'. The flow vectors now plot
correctly. This kludge can be used as a workaround.

(See ncml.diff for the changes I had to make to the NCML.)

http://motherlode.ucar.edu/repository/entry/show?entryid=d93cf4b3-b868-49a5-b49c-f5358e426346
 
If looking at the xidv bundles, you may have to alter paths to point to the
correct NCML path.

This problem had been bugging me for a while and I wanted to get back to it.

Best,

Unidata IDV Support



> > Yuan,
> > Mine still looks strange (different, but strange).
> >
> > Can you please send me your bundle that works?
> >
> > Thanks,
> > RIch
> 
> 
> Rich,
> The attached bundle using the Match Display Region in the region subset 
> panel, I also try used a selected a region option and worked.
> 
> 
> Yuan
> >
> > On Tue, Jul 7, 2015 at 2:37 PM, Unidata IDV Support
> > <address@hidden> wrote:
> > >> Yuan,
> > >> Just to clarify:
> > >>
> > >> You are saying that the velocity vectors WILL PLOT CORRECTLY if we DO
> > >> NOT restrict to a bounding box?
> > >>
> > >> -Rich
> > >
> > > Rich,
> > >         With this dataset, if you want to get the "correctly" result, you 
> > > need to spatial subset. I don't know why. In your previous bundle, you 
> > > did not
> > > do any subset in the region and you were getting bad result.
> > >
> > >
> > > Yuan
> > >>
> > >> On Tue, Jul 7, 2015 at 2:29 PM, Unidata IDV Support
> > >> <address@hidden> wrote:
> > >> >> By spatial subset, do you mean the subsetting like plotting every
> > >> >> other vector, or do you mean subsetting like restricting the region to
> > >> >> a specified bounding box?
> > >> >>
> > >> >
> > >> > I mean do not restrict to a bounding box, if you did spatial subset in 
> > >> > the IDV (by bounding box or match display region), the result would be 
> > >> > good.
> > >> >
> > >> >
> > >> > Yuan
> > >> >> On Tue, Jul 7, 2015 at 11:33 AM, Unidata IDV Support
> > >> >> <address@hidden> wrote:
> > >> >> >> Rich,
> > >> >> >>
> > >> >> >> When I create my own bundle for these data, I see essentially no
> > >> >> >> difference between IDV and Python. See attached bundle and images.
> > >> >> >>
> > >> >> >> Best,
> > >> >> >>
> > >> >> >> Unidata IDV Support
> > >> >> >>
> > >> >> >
> > >> >> > Rich,
> > >> >> >
> > >> >> >      Somehow I noticed that the results in the IDV were different 
> > >> >> > with and without the spatial subset, even if the spatial subset 
> > >> >> > covers the whole dataset.
> > >> >> > Would you be able to verify this in the Python?
> > >> >> >
> > >> >> >
> > >> >> >
> > >> >> > Yuan
> > >> >> >>
> > >> >> >> > > Yuan,
> > >> >> >> > > So when you run my bundle, do you get the same result?
> > >> >> >> > > -Rich
> > >> >> >> > >
> > >> >> >> >
> > >> >> >> > When I ran the bundle, I got even worst result. My feeling is 
> > >> >> >> > that those missing vectors were likely associated with very 
> > >> >> >> > small values. If you draw wind barb, you will get every points.
> > >> >> >> >
> > >> >> >> >
> > >> >> >> > Yuan
> > >> >> >> > > On Thu, Jul 2, 2015 at 11:54 AM, Unidata IDV Support
> > >> >> >> > > <address@hidden> wrote:
> > >> >> >> > > >> We are running 5.1u2.   Did you look at the bundle?  All 
> > >> >> >> > > >> your
> > >> >> >> > > >> questions are answered by the bundle, no?
> > >> >> >> > > >
> > >> >> >> > > >
> > >> >> >> > > > Rich,
> > >> >> >> > > >         I have no problem getting the result similar to the 
> > >> >> >> > > > python output. See the attached. However, your bundle does 
> > >> >> >> > > > look very strange, almost like some kind of
> > >> >> >> > > > filter working there. I will check and let you know.
> > >> >> >> > > >
> > >> >> >> > > >
> > >> >> >> > > > Yuan
> > >> >> >> > > >>
> > >> >> >> > > >> On Wed, Jul 1, 2015 at 5:34 PM, Unidata IDV Support
> > >> >> >> > > >> <address@hidden> wrote:
> > >> >> >> > > >> > Rich,
> > >> >> >> > > >> >
> > >> >> >> > > >> > Here are a few comments:
> > >> >> >> > > >> >
> > >> >> >> > > >> > - Update your IDV to the latest version (5.1u2, this will 
> > >> >> >> > > >> > yield no improvement
> > >> >> >> > > >> > probably, but it is usually best to work with the latest 
> > >> >> >> > > >> > version of the IDV).
> > >> >> >> > > >> >
> > >> >> >> > > >> > - For clarification, the flow vectors are supposed to be 
> > >> >> >> > > >> > derived from what? The
> > >> >> >> > > >> > data contain two u and v fields (momentum versus stokes). 
> > >> >> >> > > >> > Which one are you
> > >> >> >> > > >> > interested in? (I wonder if there is confusion/mismatches 
> > >> >> >> > > >> > going on here.)
> > >> >> >> > > >> >
> > >> >> >> > > >> > - Are you sure you are looking at the same geographic 
> > >> >> >> > > >> > extents? The IDV is
> > >> >> >> > > >> > displayed roughly over NY harbor. I cannot tell where the 
> > >> >> >> > > >> > Python plot is
> > >> >> >> > > >> > located.
> > >> >> >> > > >> >
> > >> >> >> > > >> > - I wonder if there are numerical errors going on. One of 
> > >> >> >> > > >> > the Derived Fields in
> > >> >> >> > > >> > the IDV is "Speed (from U & V)".  When you do a "Value 
> > >> >> >> > > >> > Plot" of these data, do
> > >> >> >> > > >> > you get numbers you expect?
> > >> >> >> > > >> >
> > >> >> >> > > >> > - Experiment with the Flow Vector Control in the IDV 
> > >> >> >> > > >> > Dashboard, Displays Tab.
> > >> >> >> > > >> >
> > >> >> >> > > >> > Keep us up-to-date on your progress.
> > >> >> >> > > >> >
> > >> >> >> > > >> > Best,
> > >> >> >> > > >> >
> > >> >> >> > > >> > Unidata IDV Support
> > >> >> >> > > >> >
> > >> >> >> > > >> >
> > >> >> >> > > >> >
> > >> >> >> > > >> >
> > >> >> >> > > >> >> IDV Gurus,
> > >> >> >> > > >> >>
> > >> >> >> > > >> >> We are struggling to get bottom velocity vectors 
> > >> >> >> > > >> >> displayed properly in
> > >> >> >> > > >> >> IDV (5.0u2) for a Hurricane Sandy ocean simulation.
> > >> >> >> > > >> >>
> > >> >> >> > > >> >> The bottom velocity for 2012-10-30 00:00:00 should look 
> > >> >> >> > > >> >> like the 1st
> > >> >> >> > > >> >> attached image (plotted in python using this notebook):
> > >> >> >> > > >> >> http://nbviewer.ipython.org/gist/rsignell-usgs/a85e68a71933910a171a
> > >> >> >> > > >> >>
> > >> >> >> > > >> >> but instead it looks like the 2nd attached image.
> > >> >> >> > > >> >>
> > >> >> >> > > >> >> In the IDV image you can see that the flow at the coast 
> > >> >> >> > > >> >> is the wrong
> > >> >> >> > > >> >> direction, and there are those strange black areas with 
> > >> >> >> > > >> >> no vectors.
> > >> >> >> > > >> >>
> > >> >> >> > > >> >> What is going on here?
> > >> >> >> > > >> >>
> > >> >> >> > > >> >> I'm attaching the bundle so you can recreate this 
> > >> >> >> > > >> >> yourself.
> > >> >> >> > > >> >> --
> > >> >> >> > > >> >> Dr. Richard P. Signell   (508) 457-2229
> > >> >> >> > > >> >> USGS, 384 Woods Hole Rd.
> > >> >> >> > > >> >> Woods Hole, MA 02543-1598
> > >> >> >> > > >> >>
> > >> >> >> > > >> >>
> > >> >> >> > > >> >
> > >> >> >> > > >> >
> > >> >> >> > > >> > Ticket Details
> > >> >> >> > > >> > ===================
> > >> >> >> > > >> > Ticket ID: WHS-790529
> > >> >> >> > > >> > Department: Support IDV
> > >> >> >> > > >> > Priority: Normal
> > >> >> >> > > >> > Status: Closed
> > >> >> >> > > >> >
> > >> >> >> > > >>
> > >> >> >> > > >>
> > >> >> >> > > >>
> > >> >> >> > > >> --
> > >> >> >> > > >> Dr. Richard P. Signell   (508) 457-2229
> > >> >> >> > > >> USGS, 384 Woods Hole Rd.
> > >> >> >> > > >> Woods Hole, MA 02543-1598
> > >> >> >> > > >>
> > >> >> >> > > >>
> > >> >> >> > > >
> > >> >> >> > > >
> > >> >> >> > > > Ticket Details
> > >> >> >> > > > ===================
> > >> >> >> > > > Ticket ID: WHS-790529
> > >> >> >> > > > Department: Support IDV
> > >> >> >> > > > Priority: Normal
> > >> >> >> > > > Status: Open
> > >> >> >> > >
> > >> >> >> > >
> > >> >> >> > >
> > >> >> >> > > --
> > >> >> >> > > Dr. Richard P. Signell   (508) 457-2229
> > >> >> >> > > USGS, 384 Woods Hole Rd.
> > >> >> >> > > Woods Hole, MA 02543-1598
> > >> >> >> > >
> > >> >> >> > >
> > >> >> >>
> > >> >> >
> > >> >> >
> > >> >> > Ticket Details
> > >> >> > ===================
> > >> >> > Ticket ID: WHS-790529
> > >> >> > Department: Support IDV
> > >> >> > Priority: Normal
> > >> >> > Status: Closed
> > >> >> >
> > >> >>
> > >> >>
> > >> >>
> > >> >> --
> > >> >> Dr. Richard P. Signell   (508) 457-2229
> > >> >> USGS, 384 Woods Hole Rd.
> > >> >> Woods Hole, MA 02543-1598
> > >> >>
> > >> >>
> > >> >
> > >> >
> > >> > Ticket Details
> > >> > ===================
> > >> > Ticket ID: WHS-790529
> > >> > Department: Support IDV
> > >> > Priority: Normal
> > >> > Status: Open
> > >> >
> > >>
> > >>
> > >>
> > >> --
> > >> Dr. Richard P. Signell   (508) 457-2229
> > >> USGS, 384 Woods Hole Rd.
> > >> Woods Hole, MA 02543-1598
> > >>
> > >>
> > >
> > >
> > > Ticket Details
> > > ===================
> > > Ticket ID: WHS-790529
> > > Department: Support IDV
> > > Priority: Normal
> > > Status: Open
> > >
> >
> >
> >
> > --
> > Dr. Richard P. Signell   (508) 457-2229
> > USGS, 384 Woods Hole Rd.
> > Woods Hole, MA 02543-1598
> >
> >
> 


Ticket Details
===================
Ticket ID: WHS-790529
Department: Support IDV
Priority: Normal
Status: Closed


NOTE: All email exchanges with Unidata User Support are recorded in the Unidata inquiry tracking system and then made publicly available through the web. If you do not want to have your interactions made available in this way, you must let us know in each email you send to us.