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

20040129: jython methods in IDV

>From: "Fingerhut, William A" <address@hidden>
>Organization: Lyndon State College
>Keywords: 200401231403.i0NE3Dp2024942

Hi Bill-

Sorry for the delay in responding.  I wanted to be able to
devote some time to answering.

re: jython methods
>Thanks. I hope to write more methods and I would like to do them
>correctly.

Great.  We'll just have to figure out how to incorporate them
into the system so others can use them.

>The example in the IDV user guide was a good start for me. Everything
>seemed
>to make sense. Doing grid math without explicitly codeing loops is nice.
>Not having to dimension arrays confuses me some -- read on.

Glad the docs helped.  We're in the process of revamping them so
if there are any suggested improvements, let us know.

>Once I found the method to extract pressure levels from a 3-d grid, I
>was
>ready to start. With one line of code I calculated temperature at all
>grid
>points in a constant lapse rate atmosphere. Easy!

Sometimes it just amazes me how things just seem to work in the IDV.

>However, above the tropopause the lapse rate has a different value. So, I
>need to use slightly different logic for different portions of the domain.
>It appears that I need to use array elements and explicitly code loops?

Yes.  I think the approach you took is correct.

>The user guide does not address how to do this?

Okay, I'll add that to the list of improvements.  I think it might
be best to point to Tom W's tutorial.

>I found the syntax of for and if, and the len and range functions. Using range 
>in the for statement is a little strange; seems like there should be a better 
>way?

No, that's how I'd do it.

>Then, I added a loop to change the temperature values above the tropopause.
>This worked and I was pretty happy. However, I am wondering if I shouldn't 
>use something similar to:
>       for i in range(len(t)):
>       z=p[i]
>       if z < 225.: 
>                       tstd[i]=216.65
>               else:
>                       tstd=288.15*(z/1013.25)**(287.05*.0065/9.806)
>I didn't try this, but I was wondering if jython would create the tstd
>array with the correct dimensions from this code? If I preceeded this code
>with   tstd=t it might?

Actually, it will give you an error, because the result of the
math is different from the type of the actual data.  I tried:

def tempAnom2(t):
    """ temperature anomoly from U.S. Standard Atmosphere """
    # get pressure levels of temperature grids
    tmp=extractPressureFromNWPGrid(t)
    # calculate temperature for a constant lapse rate (6.5 C/km) atmosphere
    # change temperature in stratosphere to isothermal (216.65 K)
    for i in range(len(tmp)):
      p=tmp[i]
      if p < 225: 
        tmp[i]=216.65
      else:
        tmp[i]=288.15*(p/1013.25)**(287.05*.0065/9.806)
   # calculate the temperature anomoly
    tanom=t-tmp
    return tanom

but it erred out when trying to do the tmp[i].  In your formula, if
you wanted to use pressure as the deciding factor instead of temp,
you could do:

    for i in range(len(p)):
      z=p[i]
      if p < 225: 
        tstd[i]=216.65

since p and tstd have the same dimensionality.

>Anyway, here is what I came up with. Do you have any suggestions for better 
>ways to do things?
>
>def tempAnom(t):
>    """ temperature anomoly from U.S. Standard Atmosphere """
>    # get pressure levels of temperature grids
>    p=extractPressureFromNWPGrid(t)
>    # calculate temperature for a constant lapse rate (6.5 C/km)
>atmosphere
>    tstd=288.15*(p/1013.25)**(287.05*.0065/9.806)
>    # change temperature in stratosphere to isothermal (216.65 K)
>    for i in range(len(tstd)):
>      z=tstd[i]
>      if z < 216.65: 
>        tstd[i]=216.65
>    # calculate the temperature anomoly
>    tanom=t-tstd
>    return tanom

It worked, ship it!

re: bundle
>I know the technology exists and you are really good, but the attachment
>
>was not the bundle, but an image I created using the above method. The 
>confusion might be due to the fact that the file name does not have an 
>extension, like .jpg . I created the image (hope I remember this right)
>from the vertical cross section control, one of the menus, save image.
>I assumed the IDV would add the filename extension so I did not specify
>one. Should the IDV do this for the user?

When I did this in my 1.1 b2 version, it put the .jpg extension on
it.  If you can figure out exactly how you saved this and it's different
from the above, let me know.  Is this on windows or linux?

>This vertical cross section nicely shows that where the troposphere is
>colder than normal, the stratosphere is warmer than normal, etc. 

Cool.

>I just tried to open the bundle file, and it failed.
>
>Failed to open C:\Documents and Settings\waf01060.LSC\My
>Documents\IDV\Anom_Temp.xidv Data object "C:\Documents and
>Settings\waf01060.LSC\My Documents\IDV\Anom_Temp.xidv" not compatible
>with "VisADDataSource" data family
>ucar.unidata.data.BadDataException: Failed to open C:\Documents and
>Settings\waf01060.LSC\My Documents\IDV\Anom_Temp.xidv Data object
>"C:\Documents and Settings\waf01060.LSC\My Documents\IDV\Anom_Temp.xidv"
>not compatible with "VisADDataSource" data family      at
>ucar.unidata.data.DataManager.createDataSource(DataManager.java:650)
>at
>ucar.unidata.idv.IntegratedDataViewer.createDataSource(IntegratedDataVie
>wer.java:3684)
>
>Does this have something to do with my method, or ???

Did you get this error when opening from the File/URL chooser
in the Data Source selector? For bundles, you have to use the 
File->Open... menu to load them in.  But, we'll look into changing the
File chooser to allow bundles.  You can load bundles in from the
URL chooser, so you should be able to from the File chooser as well.

I'm cc'ing Tom W. on this response in case he has any thoughts
on how your procedure could work better.  I can think of things,
but it involves knowing too much about the VisAD data structure.
I think what you did is great because you don't know anything
about the VisAD structure and it worked!  Good job.

Don
NOTE: All email exchanges with Unidata User Support are recorded in the
Unidata inquiry tracking system and then made publically 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.