20040608: importing and contouring data on GOES imagery (cont.)

>From: Owen Cooper <address@hidden>
>Organization: Aeronomy Laboratory/NOAA
>Keywords: 200406080008.i5808gtK000586 McIDAS SCHEMA Flexpart

Hi Owen,

>Thank you for the quick reply. Before I try writing a new SCHEMA I nedd
>to get a better understanding of how the SCHEMA format actually works.

OK.

>I have attached  a  text file that my current schema converts to an MD
>file.  The first few line of the text file are:
>
>LON        LAT    CO
>UNIT=*-1 X X
> -141.5  9.5  1.55309999
> -140.5  9.5  0.997270942
> -139.5  9.5  0.696673512
> -138.5  9.5  1.1179049
> -137.5  9.5  0.6285339
> -136.5  9.5  0.888016224
> -135.5  9.5  2.95988703
> -134.5  9.5  4.80974627
> -133.5  9.5  7.33905411
> -132.5  9.5  6.97319412
> -131.5  9.5  4.3494668
> -130.5  9.5  3.28649211
>
>The SCHEMA file is:
>
>"       NAME    VSN     DATE    ID      TEXTID
>"       ----    ---     ----    --      ------
>SCHEMA  FLXP    1       01168   0       "FLEXPART CO CONCENTRATIONS
>
>"KEY    SCALE   UNITS   DESCRIPTION
>"---    -----   -----   -----------
>
>ROWS    90              "90 ROWS
>
>LAT     4       DEG     "LATITUDE (DECIMAL DEG)
>
>COLUMNS 360             "DEFAULT # OF COLS
>
>LON     4       DEG     "LONGITUDE (DECIMAL DEG)
>
>DATA                    "START OF DATA SECTION
>
>CO      2       CON     "COD CONCENTRATION
>
>ENDSCHEMA

>What I don't get is this:
>The SCHEMA thinks the latitude is a row with 90 entries, and the
>longitude is a column with 360 entries.

I would rephrase this to say that the schema dictates that the
latitude will be represented in the ROWS of a 3D matrix; longitude
will be written into the COLUMNS of a 3D matrix; and the CO values
will be written into the DATA dimension of a 3D matrix.  The schema
represents where these items are to be read/written.

>And then the CO values follow
>as data.  But the lat lon and CO are all written to the text file as
>side-by-side columns, each with 13,998 entries.  So I guess I just
>don't understand how it knows how to treat each column.

The first line in the data file defines what each column of data is.
In this case, column 1 values represent longitude (negative west);
column 2 values represent latitude (postive north); column 3 values
represent data at the Lat,Lon point.  The first line is read by
the conversion routine, TXT2MD, so that it knows what values in each
column represents.  The second line -- which is optional -- specifies
how to transform the values in the various columns.  In this case,
the longitudes are multiplied by '-1' to convert them to west positive
convention.  The LAT and CO values are left untouched (the 'X'
says take the default action which is do nothing).

TXT2MD also reads the schema that is registered in SCHEMA and sets
up write routines that know where the LAT, LON, and CO values should
be written in the MD file.


re: how often the output file needs to be created
>(but I need to make it for just
>one time per file...this is because I make these files in real-time as
>soon as the data are available, every 3 hours)

You don't need to make the MD files so that there is only one time per
file.  TXT2MD can append to an existing file, and I am suggesting that
you use DAY and TIME in the ROW headers.  Each time you have a new
TIME in the same DAY, you will simply be adding a new ROW to the
output MD file.

>With the new SCHEMA that you suggested would I enter the date
>info. first, and then tack on the lat lon and CO columns?  Like this?
>
>Day     Time
>2004160 12:00
>LON        LAT    CO
>UNIT=*-1 X X
> -141.5  9.5  1.55309999
> -140.5  9.5  0.997270942
> -139.5  9.5  0.696673512
> -138.5  9.5  1.1179049
> -137.5  9.5  0.6285339
> -136.5  9.5  0.888016224
> -135.5  9.5  2.95988703
> -134.5  9.5  4.80974627
> -133.5  9.5  7.33905411
> -132.5  9.5  6.97319412
> -131.5  9.5  4.3494668
> -130.5  9.5  3.28649211

No, you will have to include the DAY and TIME in each line, like:

LON        LAT    CO       DAY     TIME
UNIT=*-1 X X X X
 -141.5  9.5  1.55309999   2004160 12:00
 -140.5  9.5  0.997270942  2004160 12:00
 -139.5  9.5  0.696673512  2004160 12:00
 -138.5  9.5  1.1179049    2004160 12:00
 -137.5  9.5  0.6285339    2004160 12:00
 -136.5  9.5  0.888016224  2004160 12:00
 -135.5  9.5  2.95988703   2004160 12:00
 -134.5  9.5  4.80974627   2004160 12:00
 -133.5  9.5  7.33905411   2004160 12:00
 -132.5  9.5  6.97319412   2004160 12:00
 -131.5  9.5  4.3494668    2004160 12:00
 -130.5  9.5  3.28649211   2004160 12:00

Or, equivalently as:

DAY     TIME   LON    LAT  CO
UNIT=X X *-1 X X
2004160 12:00 -141.5  9.5  1.55309999
2004160 12:00 -140.5  9.5  0.997270942
2004160 12:00 -139.5  9.5  0.696673512
2004160 12:00 -138.5  9.5  1.1179049
2004160 12:00 -137.5  9.5  0.6285339
2004160 12:00 -136.5  9.5  0.888016224
2004160 12:00 -135.5  9.5  2.95988703
2004160 12:00 -134.5  9.5  4.80974627
2004160 12:00 -133.5  9.5  7.33905411
2004160 12:00 -132.5  9.5  6.97319412
2004160 12:00 -131.5  9.5  4.3494668
2004160 12:00 -130.5  9.5  3.28649211

It actually doesn't matter which column you put any particular parameter
value in as long as the first line correctly specifies the ordering.

Also, the spacing between the key names in the first line mean nothing,
so the following are equivalent:

DAY     TIME   LON    LAT  CO
DAY     TIME LON LAT CO
DAY TIME LON LAT CO

The only reason I put multiple spaces between the key names is so that
they can line up with the columns of values and be more readable.

One thing that is NOT optional, however, is the length of the key names.
These must be between 1 and 4 characters in length.

>Thanks

No worries.

Cheers,

Tom
--
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.