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

20001115: MDXX problems in McIDAS-X 7.6x PROPLT (cont.)

>From: "Patrick S. Market" <address@hidden>
>Organization: University of Missouri-Columbia
>Keywords: 200011141520.eAEFKZO20569 McIDAS-X 7.6x PROPLT

Pat,

>Thanks again for your help.  The file MDXX4450 resides in our anonymous
>ftp site at:
>
>       solberg.snr.missouri.edu
>       
>in the     /pub/incoming     directory

Got it, thanks!

>I used the DSSERVE command and then ran PROFSECT.  I got the same result
>(a blank time-height) with the error now being
>
>       PROFSECT - Begin                                                       
>       Erased image and graphic frame  1                                      
>       PROFSECT: No data found matching search conditions                     
>       PROFSECT - Done

Hmm...  I just did the following:

<login as mcidas>
cd workdata
ftp solberg.snr.missouri.edu
  <user> anonymous
  <pass> address@hidden
  cd pub/incoming
  bin
  get MDXX4450
  quit

<start a McIDAS-X session>
DMAP MDXX4450         <- to make sure that my session could find the file

PROFSECT PLTC2 18 17 60 DATASET=MYDATA/PTSRCS.4450 DAY=97100
PROFSECT - Begin
Erased image and graphic frame  1
PROFSECT - Done
PROFSECT PLTC2 18 17 60 DATASET=MYDATA/PTSRCS.4450 DAY=1997100
PROFSECT - Begin
Erased image and graphic frame  1
PROFSECT - Done

Internally, PROFSECT rectifies the DAY= format so that both 97100 and
1997100 are acceptable.  Both invocations produced the desired result:
a time-height cross section of profiler data from PLTC2 (Platteville, CO).

The invocation:

PROFSECT WNCI2 18 17 60 DATASET=MYDATA/PTSRCS.4450 DAY=1997100
PROFSECT - Begin
Erased image and graphic frame  1
PROFSECT: No data found matching search conditions
PROFSECT - Done

matches your observation: no data were found for WNCI2 (Winchester, IL).

As a sanity check, I tried listing out values for WNCI2:

MDL 4450 ALL KEYS=DAY TIME IDA IDB DIR SPD Z=0 MODE=LOW IDA=WNCI2
MDL: DONE

To check my MDL invocation command line, I did the same thing for
PLTC2:

MDL 4450 ALL KEYS=DAY TIME IDA IDB DIR SPD Z=0 MODE=LOW IDA=PLTC2
   DAY   TIME  IDA  IDB   DIR   SPD
  97100      0 PLTC 2      36  7.70
  97100  10000 PLTC 2      44  6.00
  97100  20000 PLTC 2      60  5.29
  97100  30000 PLTC 2      58  4.69
  97100  40000 PLTC 2      46  4.11
  97100  50000 PLTC 2      52  4.00
  97100  60000 PLTC 2      30  3.90
  97100  70000 PLTC 2      52  3.08
  97100  80000 PLTC 2      45  3.04
  97100  90000 PLTC 2      70  2.53
  97100 100000 PLTC 2      68  2.15
  97100 110000 PLTC 2      74  2.04
  97100 120000 PLTC 2      58  1.76
  97100 130000 PLTC 2      20  2.53
  97100 140000 PLTC 2      23  2.84
  97100 150000 PLTC 2      25  2.85
  97100 160000 PLTC 2      35  3.81
  97100 170000 PLTC 2      32  3.12
  97100 180000 PLTC 2      29  3.64
  97100 190000 PLTC 2      25  4.30
  97100 200000 PLTC 2      25  4.17
  97100 210000 PLTC 2      29  3.74
  97100 220000 PLTC 2      30  3.85
  97100 230000 PLTC 2      31  4.85
MDL: DONE

So, it seems that there is no wind data for WNCI2 in the file.

>Interestingly, in trying (I changed the DAY to include '19')
>
>       PROPLT WNCI2 23 24 -60 MDF=4450 DAY=1997100 
>
>I now get data plotted for the latest hour only on the far left (there
>was *none* before).

This is weird given that the MDL lists indicate that there is absolutely
no wind data in the file for WNCI2!

>Could this be a Y2K problem still floating about?

I would not be suprised at all if PROPLT has a lingering Y2K bug in it.
I am currently rebuilding 7.613 to see if I can find the problem with
PROPLT.

>I did create the file in 1998.

This should not be a problem, and the fact that I can plot time-height
cross sections for a lot of the stations in the file (I didn't try all
of them) tells me that there is nothing systematically wrong with
the file itself.

I am somewhat at a loss to explain why you were unable to use PROFSECT
to plot out the cross sections.  An example output from a PROFSECT
invocation that I ran can be seen in:

http://www.unidata.ucar.edu/staff/tom/profsect.gif

>Thanks.

Let's see if we can't figure out why PROFSECT isn't working for you.
When I find the problem in PROPLT, I will provide the new source.

Tom

>From address@hidden Wed Nov 15 19:55:33 2000
>Subject: 20001115: MDXX problems in McIDAS-X 7.6x PROPLT (cont.) 

Pat,

I found the problem in PROPLT.  There was an implicit assumption in the
plotting loop that the data was "realtime".  "realtime" to McIDAS means
that the MD files for a particular type of data are in "circular
sets".  The circular set for realtime hourly summary FSL wind profiler
data is composed of the files: MDXX0081 - MDXX0090.  The result of the
assumption was the searching for the file MDXX0090 after the first hour
of data was plotted in the cross section.

I fixed the source for PROPLT (proplt.pgm) and have put it in the
unix/760/testcode directory of the passworded McIDAS FTP account.  If
you want to update your distribution with this modified version of
proplt.pgm, you, or whoever there is in charge of McIDAS, will need
to:

<login to the workstation on which McIDAS-X is installed as 'mcidas'>

cd mcidas7.6/src
ftp ftp.unidata.ucar.edu
  <user>
  <pass>
  cd unix/760/testcode
  get proplt.pgm
  quit

make proplt.k proplt.hlp
rm ~mcidas/bin/proplt.k
rm -f ~mcidas/help/proplt.hlp
ln proplt.k ~mcidas/bin
ln proplt.hlp ~mcidas/help

After buiding and installing the new version of proplt.k (the PROPLT
executable), the following invocations should work for you:

PROPLT PLTC2 18 17 60 MDF=4450 DAY=1997100
Erased image and graphic frame  1
PROPLT: DONE
PROPLT PLTC2 18 17 -60 MDF=4450 DAY=1997100
Erased image and graphic frame  1
PROPLT: DONE

Also, specification of the DAY= value can be done as either 1997100 or
97100.  PROPLT was always dealing with these two ways of specifying the
year and Julian day.

Tom