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

[IDV #GAI-830760]: Is there a 'turnkey' application that displays trajectory plots?



Jim,

Great! We are glad to be of service. Thanks to you for helping us improve the
IDV by finding problems.  You've done a lot of that lately.

Ohh and yes, I'll be teaching part of that Python workshop. See you then.

Best,

Unidata IDV Support

> Thank-you! That made everything work much better. You guys rock!
> 
> By the way, I'm going to be at Unidata in Boulder on Oct 21, 22 for the
> netCDF Python workshop. I'd love to take a moment to say hi in person.
> 
> Grace and peace,
> 
> Jim
> 
> On 9/23/14, 6:22 PM, Unidata IDV Support wrote:
> > Jim,
> >
> > More answers below.
> >
> >> Hi.
> >>
> >> It works!
> >>
> >> I'm far from familiar with IDV, but I did notice two things when I was
> >> displaying the trajectories.
> >>
> >> * When I added a second trajectory to the display, IDV seemed to
> >> become confused about how to handle the map projection, focus, etc.
> >> When the new trajectory is added, the display changes focus and zoom
> >> to show only the new trajectory. If I remove the second trajectory,
> >> the focus point stays at the location of the removed trajectory. As
> >> I played with IDV trying to get the focus back to the first
> >> trajectory, the behavior seemed somewhat random, one time moving to
> >> the North Pole, another time massively zooming out the North-South
> >> direction of the view. I was never able to get IDV back under control.
> > In the IDV Map View, go to the Projections menu and uncheck "Auto-set
> > Projection".  In the same menu select Predefined > Africa.
> >
> > When displaying individual trajectories, in the Legend of the Map View
> > right-click and select View > "Use Data Projection" for the trajectory of 
> > your
> > choice.
> >
> >> * All of the trajectories seem to be displayed using the same single
> >> timestamp, which is the baseline time from the 'units' attribute.
> >> These trajectories are time series, but clicking the animation
> >> 'play' button has no visible effect.
> > In the Legend of the Map view left-click on the display. This action will 
> > take
> > you the the Displays tab of the IDV Dashboard.  Select the Times tab and for
> > "Times to use" select "Track Times".
> >
> > Best,
> >
> > Unidata IDV Support
> >
> >> Both of these things may be operator error.
> >>
> >> Grace and peace,
> >>
> >> Jim
> >>
> >> On 9/19/14, 5:44 PM, Unidata IDV Support wrote:
> >>> Jim,
> >>>
> >>> Please try the new nightly build (tomorrow). We should now be able to 
> >>> handle the
> >>> "incomplete multidimensional array representation".  In fact, please use 
> >>> this
> >>> data format instead. It is far better for IDV performance as you will see.
> >>>
> >>> Best,
> >>>
> >>> Unidata IDV Support
> >>>
> >>>> Hi.
> >>>>
> >>>> The trajectories plot! Thank-you so much!
> >>>>
> >>>> Grace and peace,
> >>>>
> >>>> Jim
> >>>>
> >>>> On 9/17/14, 3:40 PM, Unidata IDV Support wrote:
> >>>>> Jim,
> >>>>>
> >>>>> Apologies for the long delay. The problem with the example file is it 
> >>>>> lacks
> >>>>> altitude data. However, the IDV should be able to handle this problem 
> >>>>> more
> >>>>> gracefully and we made a fix to do just that. Please try the IDV 
> >>>>> nightly build
> >>>>> to visualize your trajectories. Also, loading in your sample file is 
> >>>>> quite slow
> >>>>> and we are looking into why that is.
> >>>>>
> >>>>> Concerning the the CF "incomplete multidimensional array 
> >>>>> representation", would
> >>>>> it be possible to have a sample file uploaded to
> >>>>> http://motherlode.ucar.edu/repository/alias/idvupload ?
> >>>>>
> >>>>> Best,
> >>>>>
> >>>>> Unidata IDV Support
> >>>>>
> >>>>>> Hi.
> >>>>>>
> >>>>>> So, the latest build of IDV installed flawlessly on my Mac (yay!), and 
> >>>>>> I
> >>>>>> was able to successfully open the trajectory file that I sent to you.
> >>>>>> This may be entirely a function of my lack of familiarity with IDV, but
> >>>>>> I was unable to get it to produce displays of the trajectories as lines
> >>>>>> on the display map view. I could get it to show a marker moving across
> >>>>>> the map when pressing the animation 'play' button, but that is all I 
> >>>>>> was
> >>>>>> ever able to get. I could view each trajectory as a table, so it 
> >>>>>> clearly
> >>>>>> has loaded the data.
> >>>>>>
> >>>>>> On another front, if I try to load a trajectory file that has the same
> >>>>>> contents as the one I sent you, except that it uses the CF "incomplete
> >>>>>> multidimensional array representation" (section 9.3.2 in the CF 
> >>>>>> Metadata
> >>>>>> Conventions document
> >>>>>> <http://cfconventions.org/Data/cf-conventions/cf-conventions-1.6/build/cf-conventions.html#idp6424656>),
> >>>>>> it doesn't load correctly. It doesn't fail with an error, but if you
> >>>>>> view each individual trajectory as a table, you always see the data for
> >>>>>> the first trajectory.
> >>>>>>
> >>>>>> Grace and peace,
> >>>>>>
> >>>>>> Jim
> >>>>>>
> >>>>>> On 8/26/14, 5:21 PM, Unidata IDV Support wrote:
> >>>>>>> Jim,
> >>>>>>>
> >>>>>>> Just for clarification there was a bit of a miscommunication here at 
> >>>>>>> Unidata. As
> >>>>>>> we mention below, we are ready for you to try the nightly IDV at your
> >>>>>>> convenience.
> >>>>>>>
> >>>>>>> Best,
> >>>>>>>
> >>>>>>> Unidata IDV Support
> >>>>>>>
> >>>>>>>> Hi all:
> >>>>>>>>
> >>>>>>>> Jim is right that coordinates do not have to be any particular type. 
> >>>>>>>> Time coordinates must be a numeric type with udunit compatible 
> >>>>>>>> units. We are working here to fix the IDV code that assumes doubles.
> >>>>>>>>
> >>>>>>>> John
> >>>>>>>>
> >>>>>>>>
> >>>>>>>>> Jim,
> >>>>>>>>>
> >>>>>>>>> Thanks for the example data set. This file was helpful in finding 
> >>>>>>>>> several
> >>>>>>>>> problems.  For this next iteration, please try the IDV nightly 
> >>>>>>>>> build (tomorrow,
> >>>>>>>>> 8/27/2014).  (The example file takes a while to load, just FYI.) 
> >>>>>>>>> Let us know if
> >>>>>>>>> you run into problems.
> >>>>>>>>>
> >>>>>>>>> http://www.unidata.ucar.edu/downloads/idv/nightly/index.jsp
> >>>>>>>>>
> >>>>>>>>> Best,
> >>>>>>>>>
> >>>>>>>>> Unidata IDV Support
> >>>>>>>>>
> >>>>>>>>>> Hi.
> >>>>>>>>>>
> >>>>>>>>>> There is no CF requirement that a time variable be a double. In 
> >>>>>>>>>> fact, it
> >>>>>>>>>> could even validly be a short or byte. This is also true for any 
> >>>>>>>>>> other
> >>>>>>>>>> variable. The guidance is from the NODC template, not from CF.
> >>>>>>>>>>
> >>>>>>>>>> We have many datasets that we produce and consume where the time is
> >>>>>>>>>> stored as a 4-byte integer.
> >>>>>>>>>>
> >>>>>>>>>> Grace and peace,
> >>>>>>>>>>
> >>>>>>>>>> Jim
> >>>>>>>>>>
> >>>>>>>>>> On 8/21/14, 5:16 PM, Unidata IDV Support wrote:
> >>>>>>>>>>> Jim,
> >>>>>>>>>>>
> >>>>>>>>>>> I have started digging into this topic. From what I can tell, the
> >>>>>>>>>>> time field should be a double, but in the file you sent us, it is 
> >>>>>>>>>>> a float.
> >>>>>>>>>>>
> >>>>>>>>>>> We are following the CF guidance on trajectory data found here:
> >>>>>>>>>>> https://geo-ide.noaa.gov/wiki/index.php?title=NODC_NetCDF_Trajectory_Template
> >>>>>>>>>>>
> >>>>>>>>>>> Again, the blog entry here may be of some help:
> >>>>>>>>>>> http://www.unidata.ucar.edu/blogs/developer/en/entry/ipynb_cookbooks_to_create_cf
> >>>>>>>>>>>
> >>>>>>>>>>> Would it be possible to correct this float versus double problem 
> >>>>>>>>>>> at the source?
> >>>>>>>>>>>
> >>>>>>>>>>> Alternatively, NCML may be able to help us here.
> >>>>>>>>>>>
> >>>>>>>>>>> Your thoughts?
> >>>>>>>>>>>
> >>>>>>>>>>> Best,
> >>>>>>>>>>>
> >>>>>>>>>>> Unidata IDV Support
> >>>>>>>>>>>
> >>>>>>>>>>>> Jim,
> >>>>>>>>>>>>
> >>>>>>>>>>>> Thanks for the information. I'll start digging into this but 
> >>>>>>>>>>>> (un)fortunately I
> >>>>>>>>>>>> am out for a couple weeks so it will be a little while before I 
> >>>>>>>>>>>> get back to you.
> >>>>>>>>>>>> Thanks for your patience.
> >>>>>>>>>>>>
> >>>>>>>>>>>> Best,
> >>>>>>>>>>>>
> >>>>>>>>>>>> Unidata IDV Support
> >>>>>>>>>>>>
> >>>>>>>>>>>>> Hi.
> >>>>>>>>>>>>>
> >>>>>>>>>>>>> I was using the nightly release IDV 5.0u1 from August 5. I've 
> >>>>>>>>>>>>> uploaded
> >>>>>>>>>>>>> one version of the file I was trying to view, which is using the
> >>>>>>>>>>>>> 'continuous ragged array' form. The name of the file is
> >>>>>>>>>>>>> cfs_ew_850hpa_1993_AFR.nc. I have the same data stored using the
> >>>>>>>>>>>>> 'multidimensional array' form, but it was too big to upload 
> >>>>>>>>>>>>> (the upload
> >>>>>>>>>>>>> folder quota was exceeded).
> >>>>>>>>>>>>>
> >>>>>>>>>>>>> I hope this is helpful. I'm actually trying to validate the 
> >>>>>>>>>>>>> file by
> >>>>>>>>>>>>> verifying that some neutral 3rd party application properly 
> >>>>>>>>>>>>> handles it,
> >>>>>>>>>>>>> so there are no guarantees that what I have is correct!
> >>>>>>>>>>>>>
> >>>>>>>>>>>>> Grace and peace,
> >>>>>>>>>>>>>
> >>>>>>>>>>>>> Jim
> >>>>>>>>>>>>>
> >>>>>>>>>>>>> On 8/6/14, 6:22 PM, Unidata IDV Support wrote:
> >>>>>>>>>>>>>> Jim,
> >>>>>>>>>>>>>>
> >>>>>>>>>>>>>> First off, what version of the IDV are you looking at? Only 
> >>>>>>>>>>>>>> the nightly release
> >>>>>>>>>>>>>> or IDV 5.0u1 (which has not been released yet, but very soon) 
> >>>>>>>>>>>>>> has this new
> >>>>>>>>>>>>>> trajectory feature. Secondly, can you upload the file in 
> >>>>>>>>>>>>>> question to
> >>>>>>>>>>>>>> http://motherlode.ucar.edu/repository/alias/idvupload and let 
> >>>>>>>>>>>>>> us know when it is
> >>>>>>>>>>>>>> there?  I am eager to see some real trajectory data since 
> >>>>>>>>>>>>>> everything I have
> >>>>>>>>>>>>>> experimented with thus far has been canned.  Moreover, there 
> >>>>>>>>>>>>>> are other
> >>>>>>>>>>>>>> researchers wanting this feature so real data will be useful 
> >>>>>>>>>>>>>> for me.
> >>>>>>>>>>>>>>
> >>>>>>>>>>>>>> Thanks for you help.
> >>>>>>>>>>>>>>
> >>>>>>>>>>>>>> Best,
> >>>>>>>>>>>>>>
> >>>>>>>>>>>>>> Unidata IDV Support
> >>>>>>>>>>>>>>
> >>>>>>>>>>>>>>> Hi.
> >>>>>>>>>>>>>>>
> >>>>>>>>>>>>>>> That is good timing! I tried opening my file and received 
> >>>>>>>>>>>>>>> this error:
> >>>>>>>>>>>>>>>
> >>>>>>>>>>>>>>> Problem opening file:/Users/jbiard/Desktop/For
> >>>>>>>>>>>>>>> Hilawe/cfs_ew_850hpa_1993_AFR.nc
> >>>>>>>>>>>>>>> java.lang.IllegalArgumentException: Type is float,
> >>>>>>>>>>>>>>> must be double
> >>>>>>>>>>>>>>> Type is float, must be double
> >>>>>>>>>>>>>>>
> >>>>>>>>>>>>>>> This is interesting, since there is no constraint on variable 
> >>>>>>>>>>>>>>> type in CF.
> >>>>>>>>>>>>>>>
> >>>>>>>>>>>>>>> Grace and peace,
> >>>>>>>>>>>>>>>
> >>>>>>>>>>>>>>> Jim
> >>>>>>>>>>>>>>>
> >>>>>>>>>>>>>>> On 8/6/14, 2:53 PM, Unidata IDV Support wrote:
> >>>>>>>>>>>>>>>> Jim,
> >>>>>>>>>>>>>>>>
> >>>>>>>>>>>>>>>> Good timing.
> >>>>>>>>>>>>>>>>
> >>>>>>>>>>>>>>>> As of IDV 5u1 which will be release this week, the IDV can 
> >>>>>>>>>>>>>>>> now handle multiple
> >>>>>>>>>>>>>>>> trajectories in one CF-compliant file. When wanting to view 
> >>>>>>>>>>>>>>>> these data, select a
> >>>>>>>>>>>>>>>> Data Source Type of "Trajectory Sounding Files".  You will 
> >>>>>>>>>>>>>>>> have to build the
> >>>>>>>>>>>>>>>> trajectory file in a CF compliant manner as described here:
> >>>>>>>>>>>>>>>> http://www.unidata.ucar.edu/blogs/developer/en/entry/ipynb_cookbooks_to_create_cf.
> >>>>>>>>>>>>>>>>
> >>>>>>>>>>>>>>>> This new IDV functionality should still be considered beta 
> >>>>>>>>>>>>>>>> quality and there
> >>>>>>>>>>>>>>>> could very well still be problems. Please keep us up-to-date 
> >>>>>>>>>>>>>>>> on your progress.
> >>>>>>>>>>>>>>>>
> >>>>>>>>>>>>>>>> Best,
> >>>>>>>>>>>>>>>>
> >>>>>>>>>>>>>>>> Unidata IDV Support
> >>>>>>>>>>>>>>>>
> >>>>>>>>>>>>>>>>
> >>>>>>>>>>>>>>>>> Jim
> >>>>>>>>>>>>>>>>>
> >>>>>>>>>>>>>>>>> We plot trajectories here.
> >>>>>>>>>>>>>>>>>
> >>>>>>>>>>>>>>>>> http://www.esrl.noaa.gov/psd/cgi-bin/data/trajtool/traj.pl
> >>>>>>>>>>>>>>>>>
> >>>>>>>>>>>>>>>>> We use Fortran code to calculate the trajectories, stored 
> >>>>>>>>>>>>>>>>> the output in
> >>>>>>>>>>>>>>>>> nertCDF and then use a NCL script to plot them. The script 
> >>>>>>>>>>>>>>>>> works pretty
> >>>>>>>>>>>>>>>>> well (except where trajectories pass over 0E). We looked 
> >>>>>>>>>>>>>>>>> into CF complaint
> >>>>>>>>>>>>>>>>> netCDF trajectories standard a while back but didn't 
> >>>>>>>>>>>>>>>>> implement them. Part
> >>>>>>>>>>>>>>>>> of the reason was that the code output into it's own netCDF 
> >>>>>>>>>>>>>>>>> format. Still,
> >>>>>>>>>>>>>>>>> we could have reread and stored the data. I think we would 
> >>>>>>>>>>>>>>>>> have needed to
> >>>>>>>>>>>>>>>>> change how time was handled. Also, times went backward for 
> >>>>>>>>>>>>>>>>> back
> >>>>>>>>>>>>>>>>> trajectories and we weren't clear if that was okay.
> >>>>>>>>>>>>>>>>>
> >>>>>>>>>>>>>>>>> We may try to transition to CF at some point.
> >>>>>>>>>>>>>>>>>
> >>>>>>>>>>>>>>>>> Cathy Smith
> >>>>>>>>>>>>>>>>>
> >>>>>>>>>>>>>>>>>
> >>>>>>>>>>>>>>>>> On 8/6/14 8:42 AM, Jim Biard wrote:
> >>>>>>>>>>>>>>>>>
> >>>>>>>>>>>>>>>>> Hi.
> >>>>>>>>>>>>>>>>>
> >>>>>>>>>>>>>>>>> We are working on building a CF-compliant trajectory 
> >>>>>>>>>>>>>>>>> dataset. I'd like to
> >>>>>>>>>>>>>>>>> be able to verify correctness by using an existing 
> >>>>>>>>>>>>>>>>> application to plot the
> >>>>>>>>>>>>>>>>> trajectories, but I can't seem to find any applications 
> >>>>>>>>>>>>>>>>> that are
> >>>>>>>>>>>>>>>>> "trajectory aware".  Any suggestions?
> >>>>>>>>>>>>>>>>>
> >>>>>>>>>>>>>>>>> Grace and peace,
> >>>>>>>>>>>>>>>>>
> >>>>>>>>>>>>>>>>> Jim
> >>>>>>>>>>>>>>>>> --
> >>>>>>>>>>>>>>>>> CF-metadata mailing
> >>>>>>>>>>>>>>>>> address@hidden.eduhttp://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata
> >>>>>>>>>>>>>>>>>
> >>>>>>>>>>>>>>>>> --
> >>>>>>>>>>>>>>>>> ----------------------------------------------
> >>>>>>>>>>>>>>>>>
> >>>>>>>>>>>>>>>>> Emails about data/webpages may get quicker responses from 
> >>>>>>>>>>>>>>>>> emailing
> >>>>>>>>>>>>>>>>> address@hidden
> >>>>>>>>>>>>>>>>>
> >>>>>>>>>>>>>>>>> _______________________________________________
> >>>>>>>>>>>>>>>>> CF-metadata mailing list
> >>>>>>>>>>>>>>>>> address@hidden
> >>>>>>>>>>>>>>>>> http://mailman.cgd.ucar.edu/mailman/listinfo/cf-metadata
> >>>>>>>>>>>>>>>>>
> >>>>>>>>>>>>>>>> Ticket Details
> >>>>>>>>>>>>>>>> ===================
> >>>>>>>>>>>>>>>> Ticket ID: GAI-830760
> >>>>>>>>>>>>>>>> Department: Support IDV
> >>>>>>>>>>>>>>>> Priority: Normal
> >>>>>>>>>>>>>>>> Status: Closed
> >>>>>>>>>>>>>> Ticket Details
> >>>>>>>>>>>>>> ===================
> >>>>>>>>>>>>>> Ticket ID: GAI-830760
> >>>>>>>>>>>>>> Department: Support IDV
> >>>>>>>>>>>>>> Priority: Normal
> >>>>>>>>>>>>>> Status: Closed
> >>>>>>>>>>> Ticket Details
> >>>>>>>>>>> ===================
> >>>>>>>>>>> Ticket ID: GAI-830760
> >>>>>>>>>>> Department: Support IDV
> >>>>>>>>>>> Priority: Normal
> >>>>>>>>>>> Status: Closed
> >>>>>>>>>>>
> >>>>>>>>>> --
> >>>>>>>>>> CICS-NC <http://www.cicsnc.org/> Visit us on
> >>>>>>>>>> Facebook <http://www.facebook.com/cicsnc>  *Jim Biard*
> >>>>>>>>>> *Research Scholar*
> >>>>>>>>>> Cooperative Institute for Climate and Satellites NC 
> >>>>>>>>>> <http://cicsnc.org/>
> >>>>>>>>>> North Carolina State University <http://ncsu.edu/>
> >>>>>>>>>> NOAA's National Climatic Data Center <http://ncdc.noaa.gov/>
> >>>>>>>>>> 151 Patton Ave, Asheville, NC 28801
> >>>>>>>>>> e: address@hidden
> >>>>>>>>>> o: +1 828 271 4900
> >>>>>>>>>>
> >>>>>>>>>>
> >>>>>>>>>>
> >>>>>>>>>>
> >>>>>>>>>>
> >>>>>>>>>>
> >>>>>>> Ticket Details
> >>>>>>> ===================
> >>>>>>> Ticket ID: GAI-830760
> >>>>>>> Department: Support IDV
> >>>>>>> Priority: Normal
> >>>>>>> Status: Closed
> >>>>>>>
> >>>>>> --
> >>>>>> CICS-NC <http://www.cicsnc.org/> Visit us on
> >>>>>> Facebook <http://www.facebook.com/cicsnc>      *Jim Biard*
> >>>>>> *Research Scholar*
> >>>>>> Cooperative Institute for Climate and Satellites NC 
> >>>>>> <http://cicsnc.org/>
> >>>>>> North Carolina State University <http://ncsu.edu/>
> >>>>>> NOAA's National Climatic Data Center <http://ncdc.noaa.gov/>
> >>>>>> 151 Patton Ave, Asheville, NC 28801
> >>>>>> e: address@hidden
> >>>>>> o: +1 828 271 4900
> >>>>>>
> >>>>>>
> >>>>>>
> >>>>>>
> >>>>>>
> >>>>>>
> >>>>> Ticket Details
> >>>>> ===================
> >>>>> Ticket ID: GAI-830760
> >>>>> Department: Support IDV
> >>>>> Priority: Critical
> >>>>> Status: Open
> >>>>>
> >>>> --
> >>>> CICS-NC <http://www.cicsnc.org/> Visit us on
> >>>> Facebook <http://www.facebook.com/cicsnc>        *Jim Biard*
> >>>> *Research Scholar*
> >>>> Cooperative Institute for Climate and Satellites NC <http://cicsnc.org/>
> >>>> North Carolina State University <http://ncsu.edu/>
> >>>> NOAA's National Climatic Data Center <http://ncdc.noaa.gov/>
> >>>> 151 Patton Ave, Asheville, NC 28801
> >>>> e: address@hidden
> >>>> o: +1 828 271 4900
> >>>>
> >>>>
> >>>>
> >>>>
> >>>>
> >>>>
> >>> Ticket Details
> >>> ===================
> >>> Ticket ID: GAI-830760
> >>> Department: Support IDV
> >>> Priority: Critical
> >>> Status: Closed
> >>>
> >> --
> >> CICS-NC <http://www.cicsnc.org/> Visit us on
> >> Facebook <http://www.facebook.com/cicsnc>  *Jim Biard*
> >> *Research Scholar*
> >> Cooperative Institute for Climate and Satellites NC <http://cicsnc.org/>
> >> North Carolina State University <http://ncsu.edu/>
> >> NOAA's National Climatic Data Center <http://ncdc.noaa.gov/>
> >> 151 Patton Ave, Asheville, NC 28801
> >> e: address@hidden
> >> o: +1 828 271 4900
> >>
> >>
> >>
> >>
> >>
> >>
> >
> > Ticket Details
> > ===================
> > Ticket ID: GAI-830760
> > Department: Support IDV
> > Priority: Critical
> > Status: Closed
> >
> 
> --
> CICS-NC <http://www.cicsnc.org/> Visit us on
> Facebook <http://www.facebook.com/cicsnc>     *Jim Biard*
> *Research Scholar*
> Cooperative Institute for Climate and Satellites NC <http://cicsnc.org/>
> North Carolina State University <http://ncsu.edu/>
> NOAA's National Climatic Data Center <http://ncdc.noaa.gov/>
> 151 Patton Ave, Asheville, NC 28801
> e: address@hidden
> o: +1 828 271 4900
> 
> 
> 
> 
> 
> 


Ticket Details
===================
Ticket ID: GAI-830760
Department: Support IDV
Priority: Critical
Status: Closed