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

[IDV #LCA-261543]: GrADS .ctl files

Hi Stefan-

Sorry for the delay in responding.  My real job takes priority.

> Such a legend doesn't exist in the version I am using(of I have it
> turned off and I don't know how to turn it on). I am starting IDV just
> by clicking the IDV application icon.

Sorry, if you click on the clock in the lower left corner of the main window, 
the clock will switch to a memory monitor.
 
> Is there any documentation through Unidata that describes how each
> derived quantity is calculated? This is especially important for
> Equiv. Potential Temp, in which many different methods are used and
> some are better than others.

Just the code (Mostly in ucar.unidata.data.grid.DerivedGridFactory). ;-)  In 
most cases, IDV uses the brute force approach which does not account for 
conservation of various parameters like GEMPAK or WRF post-processing code.  I 
would opt for having ARWPost calculate the variables you need most and use 
those.  For Theta-E, the code looks like:

        return Util.clone(PotentialTemperature.create(pressure,
                AdiabaticEquivalentTemperature.create(temperature,
                    mixingRatio)), getRealType());

which might give you a sense of how it is doing this.

Don

> On Apr 25, 2011, at 3:55 PM, Unidata IDV Support wrote:
> 
> > Hi Stefan-
> >
> >> Been testing everything and all seems well so far. Only issue I am
> >> getting is when editing derived quantities using the formula editor.
> >> When I try to go plot a skew-t or another derived variable, I get a
> >> Java
> >> heap space error: java.lang.OutOfMemoryError: Java heap space. I am
> >> only
> >> trying to plot one time step using this derived quantity with edited
> >> formula, so Im sure my machine isn't running out of memory(IDV is
> >> allowed to use >1GB). I was able to successfully derived theta-e
> >> using
> >> the formula editor and plot one time step.
> >
> > I didn't have that issue on my system.  How much memory are you
> > running the IDV with? (e.g. what does it show in the memory legend
> > at the bottom of the screen.)  Also, how are you starting the IDV?
> >
> > Question: are all the
> >> temperatures in the formula editors based off of temperature in
> >> Celsius
> >> or Kelvin? Thanks!
> >
> > Theoretically, it shouldn't matter.  The IDV will convert to Kelvin
> > for the calculations.  If you find that the answers are not what you
> > expect, let us know.
> >
> > Don
> >
> >> On Apr 21, 2011, at 2:45 PM, Unidata IDV Support wrote:
> >>
> >>> Hi Stefan-
> >>>
> >>> On 4/21/11 11:23 AM, Stefan Cecelski wrote:
> >>>> I'll respond to both of your emails here.
> >>>
> >>> that's fine.  I'm using the official Unidata tracking system to
> >>> document the solutions.
> >>>
> >>>> In response to the two concurrent open files:
> >>>> The .ctl names are different, and access two totally different .dat
> >>>> files with different file names. The multi-timestep .dat file is
> >>>> 14.55GB. I also have other 1.33km horizontal resolution files at
> >>>> 37GB
> >>>> which I will be testing on. I will test to see if I select single
> >>>> time
> >>>> steps if it works. Maybe it has to do with opening too much data?
> >>>
> >>> Okay. The problem is that the pointers to the data that were beyong
> >>> 2GB are larger than an integer can hold so it would end up as a
> >>> negative number.  I made a change to use longs instead of ints.
> >>>
> >>> I updated the ncIdv.jar on RAMADDA, so please try that one.
> >>>
> >>>> In response to the derived variables:
> >>>> Thanks very much, this clears it up. The attractiveness in using
> >>>> IDV for
> >>>> visualization is the fact that I don't have to script the code to
> >>>> calculate such things as IPV, Skew-Ts, ect. In fact, I had to write
> >>>> to
> >>>> FORTRAN code to make ARWpost output for Equivalent Potential
> >>>> Temperature. It is a SERIOUS pain in the butt, and would much
> >>>> rather
> >>>> just have IDV do it! I presume I can manually create functions that
> >>>> will
> >>>> select the right variables for derived quantities correct?
> >>>
> >>> Here's something you can do (and I didn't know you could do this
> >>> until I tried ;-)).  From the IDV main menu, select Edit->Formulas-
> >>>> Edit Formulas->Derived Quantities and then the formula you want to
> >>> use (e.g. Sounding Data (%N1% & %N2% only)).  In the Formula Editor
> >>> dialog, click on the Derived Tab. In the Parameters section, add in
> >>> the parameters that you want on the line under TEMP, DEWPOINT.  I
> >>> used tc,td.  Click the Change Formula button.  Then, if you have the
> >>> data source already loaded, you can right click on it and select
> >>> Reload Data from the pop-up menu, or you can load in a new data
> >>> source if you didn't have one.  Now, under derived, you should see
> >>> the old one and a new one that says Sounding Data (tc,td only) in
> >>> the list (near the bottom).  If you want it near the top, just
> >>> switch the order of the Parameter lines in the Formula Editor.
> >>>
> >>> You'll notice that the next time you go through the Edit->Formulas
> >>> menu, the edited formula will be tagged with <local>.  You haven't
> >>> overwritten the system one, just created a local version.  If you
> >>> mess up, you can always delete that one (Remove Formula button in
> >>> the Formula Editor) and start again.
> >>>
> >>> Don
> >>>
> >>>
> >>> Ticket Details
> >>> ===================
> >>> Ticket ID: LCA-261543
> >>> Department: Support IDV
> >>> Priority: Normal
> >>> Status: Open
> >>>
> >>
> >>
> >>
> >
> >
> > Ticket Details
> > ===================
> > Ticket ID: LCA-261543
> > Department: Support IDV
> > Priority: Normal
> > Status: Open
> >
> 
> 


Ticket Details
===================
Ticket ID: LCA-261543
Department: Support IDV
Priority: Normal
Status: Open