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

[McIDAS #XMY-392143]: Mcidas-SFCPLOT behavior?



Hi Bill,

Long time no hear...

I apologize for not getting back to you immediately on your inquiry.  It
took me a bit of time to figure out a workaround for the problem you are
reporting.

re:
> This is probably something I should have learned in kindergarten, but I
> was watching birds that day:

:-)  Actually, the McIDAS training sessions have never spent enough time
going through the use of context files.  I am very happy that you were
curious enough to learn about them!  Have you looked at creating your own
custom maps using the MAP context files?  If you decide to look into creating
your own definitions in MAP.SITE or MAP.USER, please be aware that a bug
was just found where one can create a MAP definition in MAP.(USER|SITE|CORE)
that works nicely with MAP, but will result in an error when specifying the
MAP mnemonic as a MAP name for PTDISP and PTCON (which get run by all
macro commands like SFCPLOT, SFCCON, RAOBPLOT, RAOBCON, etc.).  The point
I am trying to make can be easily illustrated using the map definition
for CHINA that I included in the MAP.CORE file:

Try:

SFCPLOT T CHINA

You should get the text output:

Accessing Dataset Name = RTPTSRC/SFCHOURLY.ALL
PTDISP: Invalid MAP name specified = CHINA
SFCPLOT: PTDISP command failed
SFCPLOT failed, rc=1

MAP, however, is perfectly happy with this definition:
MAP CHINA
MAP: Completed frame 5

And, even more strange is that SFCPLOT can use the CHINA map just plotted
when doing an overlay:

ERASE
MAP CHINA
SFCPLOT T OLAY

I have reported this inconsistency to SSEC.

Back to your problem...

> SFCPLOT with the parameter PLOT puts the "Present Weather" in text form
> (I use SFCPLOT PLOT PTYPE=PMSL).  I would prefer symbols.

This appears to be a bug.  The current weather is _supposed_ to be plotted
as symbols given the specification in SFCPLOT.CORE:

PLOT   UNIT=AMERICAN METRIC \ 
       AMERICAN='F F X X MB KTS' METRIC='C C X X MB KTS' \
       COLOR='2 2 3 3 3 4 3 2' \
       FORMAT='I3 I3 SYMB SYMB K3 FLAG X I1' \
       LSIZE='6 6 6 6 6 9 5 6' \
       OFFSET='-8 -12 8 -12 0 -8 -20 -8 5 X X 8 5 X 14' \
       LOC=+ \
       TITLE=X 3 \
       FONT=' '\ 
       ISFC='T TD WX1 WX2 PSL WIND ID CIGC' \
       SVCA='T TD WX1 WX2 PSL WIND ID CCL' \
       SYN= na \ 
       ISHP=na

The specification of 'SYMB' in FORMAT= is specifying that the weather
should be plotted as symbols.  The PLOT option used to work correctly;
I will investigate why it is no longer plotting symbols and will likely
submit a bug report to SSEC.

> Individual
> plotting of symbols and values is possible to build a station plot
> (e.g., SFCPLOT WXS; SFCPLOT T; SFCPLT TD), but things like offset from
> station location are a pain if one wants to plot more than one parameter.
> You have already done all the work for me with the PLOT option.  I
> just want symbolic present weather.

OK.

> So, reading the friendly manual, I copied SFCPLOT.CORE to SFCPLOT.SITE,
> and under the PLOT.PMSL definition started playing around, changing WX1
> to WXS (and WXC) for the various data sources, with no success.

The correct formalism is to specify plotting of the current weather using
'WX1 WX2' and FORMAT of 'SYMB SYMB'.  This is the way it is done for WXS in
SFCPLOT.CORE:

WXS    FORMAT='SYMB SYMB' OFFSET='0 -4 0 8' \ 
       ISFC='WX1 WX2' SVCA='WX1 WX2' SYN='WX1 WX2' ISHP='WXP'

> At this
> point I will stop, because although FORTRAN format statements are still
> something I understand, the total construction of the SFCPLOT.CORE is
> only vaguely interpretable by me.  (Actually, if I had just gone ahead
> and built my own file of SFCPLOT commands with individual parameters and
> offsets empirically determined, I'd be done by now.  But I feel like I'm
> missing something, and I wonder if there's a more elegant way to
> accomplish this.)

I played around with the PLOT specification that I copied to my 
$MCDATA/SFCPLOT.USER
file (SFCPLOT.USER is used in perferance to SFCPLOT.SITE and SFCPLOT.CORE, but
it is intended to contain user-specific definitions.  Your use of SFCPLOT.SITE
is preferred when working as the user 'mcidas' who intends to setup defaults
that will be used by all McIDAS users).  I found that by putting the weather
symbol definitions first, I could get the plot I wanted.  Here is my 
SFCPLOT.USER
entry for a parameter I called TEST:

TEST   UNIT=AMERICAN METRIC \
       AMERICAN='X X F F MB KTS' \
       METRIC='X X C C MB KTS' \ 
       FORMAT='SYMB SYMB I3 I3 K3 FLAG X I1' \
       OFFSET='0 -7 -20 -8 -8 -12 8 -12 -8 5 X X 8 5 X 14' \
       LSIZE='6 6 6 6 6 10 6 6' \
       ISFC='WX1 WX2 T TD PSL WIND ID CIGC' \
       SVCA='WX1 WX2 T TD PSL WIND ID CCL' \
       SYN='WX1 WX2 T TD PSL WIND ID' \
       ISHP='WXP T TD PSL WIND IDA'

I would recommend using the TEST definition as a starting point from which you
can tweek settings.  After you have the look you want, simply change the name
TEST to PLOT in your SFCPLOT.SITE file.

I have attached a plot of TEST values over Colorado this morning (since several
stations are reporting light snow).

> Thanks.

No worries.  Thanks for identifying the PLOT bug.  I never use the PLOT option,
so I had not seen it.

Cheers,

Tom
****************************************************************************
Unidata User Support                                    UCAR Unidata Program
(303) 497-8642                                                 P.O. Box 3000
address@hidden                                   Boulder, CO 80307
----------------------------------------------------------------------------
Unidata HomePage                       http://www.unidata.ucar.edu
****************************************************************************


Ticket Details
===================
Ticket ID: XMY-392143
Department: Support McIDAS
Priority: Normal
Status: Closed

Attachment: sfcplot_TEST.GIF
Description: GIF image


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