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

19991129: ISENT functionality on grids

>From: "Jennie L. Moody" <address@hidden>
>Organization: UVa
>Keywords: 199911291621.JAA06834 McIDAS-X GRDxxx ISENT

Jennie,

>I have a few questions about the old routine ISENT.  We used to
>calculate isentropic potential vorticity, using the ISENT command to
>convert the vertical coordinate (doing the vertical interpolations) to
>theta.  Once you have theta as the vertical coordinate, it was
>straightforward to use the MDX command to calculate vorticity on these
>surfaces, and then delP (the finite difference of pressure variations
>on a theta surface), and then their product (actually,
>(1/delP)*absvor).

Right.  SSEC has stated intentions of adding 'math' to point source
routines (they are working on SFCPLOT, etc. right now), and they
have stated that there needs to be an ADDE version of ISENT, but
when these will be available is as yet unknown.

>Anyway, I would like to do the same thing with grids.  Currently, we
>just calculate PV on constant pressure surfaces, by calculating
>delTheta/dP (rather than delP/dtheta).  However, I would like to try
>using the ISENT program to transform the grids to theta coordinates,
>and then use the old approach I had for calculating IPV from rawinsonde
>data.  It seems this might be straightforward *if* there is a way to
>convert GRIDS to MD files

Yes, this would seem to be straightforward IF there were such a conversion
routine.  Going from MDs to GRIDs is easy enough, but I don't know of
a routine that goes the other way.

>(seems in principle this is pretty easy, just
>gives a very dense MD field- I guess one obvious limitation is the
>current array size allowed?)

The limitation on the MD side would be the number of "stations" (grid
points) that would be possible to store.

>I went off and looked at the isent.pgm
>(in version 7.6, is this code that is on the verge of being sunset?)

Sunset items are discussed in:

http://www.unidata.ucar.edu/packages/mcidas/mcx/sunset.html 

ISENT is not on the list of items to be sunset by SSEC, but it probably
should be.

>Yesterday I went and read through the workshop notes, I thought that
>working with isentropic surfaces was an "advanced" topic, perhaps from
>previous workshops?

We occasionally had lab exercises that used ISENT and ISENP.

>It occurred to me that maybe this (converting to
>theta as the vertical coordinate) is something that has already been
>worked out with new routines like GRDDISP (although its not obvious to
>me how.)

I don't see it in the math functionality of the GRDxxx commands.

>I know we can calculate new parameters, but I don't see anyway
>that we can actually transform to a new coordinate system, since you
>always specify the level for the new parameter, this seems to lock you
>into the predefined levels of the grid.

Right, I don't see it.

>So, is this already doable, or will I have to find my own way to mimic
>ISENT for grids if I want to do this?

I guess that your problem is wanting/needing the MD files to be remote;
true?  I see by looking at the 7.6 version of the ISEN schema that it
has been updated to Y2K capable date formats:

SEE DCISEN
"ISEN (AMW) 0284 SCHEMA-- ISENTROPIC SURFACE DATA
"        NAME VSN DATE  ID "TEXTID
"        ---- --- ----- -- -------
  SCHEMA ISEN   3 98258 0  "ISENTROPIC SURFACE DATA

" KEY  SCALE UNITS
" ---- ----- -----

  ROWS 8

  DAY        CYD      "YEAR AND JULIAN DAY (CCYYDDD)
  TIME       HMS      "NOMINAL TIME (HH0000)

  COLUMNS 1500

  IDN                 "STATION NUMBER
  LAT  4     DEG          "RANGE (-90 => +90) POSITIVE NORTH
  LON  4     DEG          "RANGE (-180 => +180) POSITIVE WEST
  ZS         M            "ELEVATION
  ST         CHAR         "STATE ID

  DATA

  REPEAT 20
  THA        K        "THETA VALUE
  P    1     MB       "THETA SURFACE PRESSURE
  MIX  1     GPKG     "THETA SURFACE MIXING RATIO
  U    1     MPS      "THETA SURFACE U WIND COMPONENT
  V    1     MPS      "THETA SURFACE V WIND COMPONENT
  SY   2              "THETA SFC MONTGOMERY STREAM FNC E7*(CM/S)**2
  DELP 1     MB       "PRESSURE DIFFERENCE (CURRENT-PREVIOUS THETA SFC)
  Z          M        "HEIGHT OF THETA SURFACE (M)

ENDSCHEMA
SEE: Done...EOF Encountered.

Hopefully, this means that someone at SSEC has gone through ISENT and
modified anything that is not Y2K capable.  A quick look throught the
code shows that the DAY stuff has been updated from YYDDD to CCYYDDD,
so it looks like you can keep on using ISENT.  The downside, of course,
is that this is not an ADDE command.

Tom

>From address@hidden  Mon Nov 29 14:34:28 1999

re: going from grids to md files
>Well, I thought this was probably the case, but didn't hurt to
>ask.

re: The limitation on the MD side would be the number of "stations"
>Right, thats what I meant actually.

re: sunset items
>Well, I had looked here, and I didn't see ISENT, but I wondered
>because it would seem that its not a supported routine.  I mean,
>the only way I had ever used ISENT was in conjunction with subsequent
>MDX commands, and that it is being sunset, so frankly I was confused.
  
re: ISENP is to be sunset
>Maybe this is what I was remembering.

re: remote MD files
>I am not sure what you mean by this:
>"your problem is wanting/needing the MD files to be remote" 

>This sounds like ADDE-ese, right?  Actually, I don't really want to 
>make an MD file (if I did, I don't think it would matter if
>it was local or remote), since all I *really* want is just
>a grid file which has been transformed to theta-surfaces.  I guess
>if there is a way to work with ISENT, maybe it would be best to
>just figure out how to use the vertical interpolation info that
>allows you to calculate the values of variables on the new
>surface, and then write out new values for that surface.  
>Bet there are lots of others who would like to work on theta 
>surfaces, but SSEC must know this.  

re: ISENT has been updated to be Y2K capable
>Right.  Which just means it will only work on local data, right?