[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
20000926: GRDIMAGE issues (cont.)
- Subject: 20000926: GRDIMAGE issues (cont.)
- Date: Thu, 12 Oct 2000 12:16:34 -0600
>From: Bill Fingerhut <address@hidden>
>Organization: Lyndon State
>Keywords: 200009251326.e8PDQ1b04272 McIDAS-X 7.612 GRDIMAGE
Bill,
I finally got enough time to login to your system as you and try some
grdimage.k invocations to see what I could see.
Previously, you wrote:
>>I have a problem with GRDIMAGE. When the source dataset position
>>is greater than one, the wrong gridfile is used. To illustrate:
>>
>>DSSERVE LIST MYDATA
>>Group/Descriptor Type Format & Range RT Comment
>>------------------------ ----- ------------------ -- --------------------
>>MYDATA/GRIDS GRID GRID 2001-2099 Local Scratch Grids
>>MYDATA/JGRDS GRID GRID 311-320 GLOBAL GRIDS
>>MYDATA/JIMGS IMAGE AREA 1800-1819 JET IMAGES
>>MYDATA/MGRDS GRID GRID 2001-2002 SCRATCH MODEL GRIDS
>>MYDATA/MIMGS IMAGE AREA 3001-3010 SCRATCH FOR JETSTREAM
>>MYDATA/SIMGS IMAGE AREA 2001-2099 SCRATCH IMAGES
>>DSSERVE: done
>>
>>GRDLIST MYDATA/GRIDS.4 NUM=ALL
>>Dataset position 4 Directory Title=
>>PAR LEVEL DAY TIME SRC FHOUR FDAY FTIME GRID PRO
>>---- --------- ------------ -------- ---- ----- ------------ -------- ---- ---
>>U MAXW 20 SEP 00264 00:00:00 AVN 0 20 SEP 00264 00:00:00 1
>>MERC
>>V MAXW 20 SEP 00264 00:00:00 AVN 0 20 SEP 00264 00:00:00 2
>>MERC
>>SPD MAXW 20 SEP 00264 00:00:00 AVN 0 20 SEP 00264 00:00:00 3
>>MERC
>>Number of grids listed = 3
>>GRDLIST - done
>>
>>->try to use an existing grid - position 4, grid 3
>>
>>GRDIMAGE MYDATA/GRIDS.4 MYDATA/MIMGS.5 RANGE=0 255 0 255 GRID=3 MAG=5 5
>>GRDIMAGE: Failed to read grid
>>TFILE CLOSE
>>-----------------------------------------
>>GRDIMAGE did not find the existing grid???
I did what should be essentially the same thing:
<login as you>
cd mcidas/data
mcenv -f 1@750x1450
cirrus:~/mcidas/data> dsserve.k LIST
Group/Descriptor Type Format & Range RT Comment
------------------------ ----- ------------------ -- --------------------
MYDATA/GRIDS GRID GRID 2001-2099 Local Scratch Grids
MYDATA/JGRDS GRID GRID 311-320 GLOBAL GRIDS
MYDATA/JIMGS IMAGE AREA 1800-1819 JET IMAGES
MYDATA/MGRDS GRID GRID 2001-2009 SCRATCH MODEL GRIDS
MYDATA/MIMGS IMAGE AREA 2001-2009 SCRATCH MODEL IMAGES
MYDATA/SIMGS IMAGE AREA 2001-2099 SCRATCH IMAGES
dsserve.k: done
cirrus:~/mcidas/data> grdlist.k MYDATA/GRIDS.4 NUM=ALL
Dataset position 4 Directory Title= SCRATCH FOR JETSTREAM
PAR LEVEL DAY TIME SRC FHOUR FDAY FTIME GRID PRO
---- --------- ------------ -------- ---- ----- ------------ -------- ----- ----
U MAXW 26 SEP 00270 00:00:00 AVN 0 26 SEP 00270 00:00:00 1 MERC
V MAXW 26 SEP 00270 00:00:00 AVN 0 26 SEP 00270 00:00:00 2 MERC
SPD MAXW 26 SEP 00270 00:00:00 AVN 0 26 SEP 00270 00:00:00 3 MERC
SPD MAXW 26 SEP 00270 00:00:00 AVN 0 26 SEP 00270 00:00:00 4 MERC
SPD MAXW 26 SEP 00270 00:00:00 AVN 0 26 SEP 00270 00:00:00 5 MERC
P MSL 26 SEP 00270 00:00:00 AVN 0 26 SEP 00270 00:00:00 6 MERC
Z MAXW 26 SEP 00270 00:00:00 AVN 0 26 SEP 00270 00:00:00 7 MERC
Number of grids listed = 7
GRDLIST - done
cirrus:~/mcidas/data> grdimage.k MYDATA/GRIDS.4 MYDATA/MIMGS.5 RANGE=0 255 0
255 GRID=3 MAG=5 5
Transferring AREA data outbound, bytes= 1042404
grdimage.k: Done
cirrus:~/mcidas/data> imglist.k MYDATA/MIMGS.5 FORM=EXP
Image file directory listing for:MYDATA/MIMGS
Pos Satellite/ Date Time Center Res (km) Image_Size
sensor Lat Lon Lat Lon
--- ------------- ------------ -------- ---- ---- ----- ----- ------------
5 DERIVED 26 SEP 00270 00:00:00 0 -1
Band: 1 No Information Available 28.0 28.0 721 x 1444
proj: 0 created: 2000286 175357 memo: SPD
type:PRD cal type:BRIT
offsets: data= 1280 navigation= 256 calibration= 768 auxillary= 0
doc length: 0 cal length: 0 lev length: 0 PREFIX= 0
valcod: 0 zcor: 0 avg-smp: N
start yyddd: 2000270 start time: 0 start scan: 0
lcor: 1 ecor: 1 bytes per pixel: 1 ss: 0
Image Center Point Res (derived) Lat: 27.80 (km) Lon: 27.80 (km)
imglist.k: done
cirrus:~/mcidas/data> ls -alt AREA2005
-rw-rw-r-- 1 fingerhu 3003 1042404 Oct 12 17:53 AREA2005
cirrus:~/mcidas/data> imgdisp.k MYDATA/MIMGS.5 REFRESH='EG;MAP H'
Beginning Image Data transfer, bytes= 1042404
IMGDISP: loaded frame 1
EG;MAP H
imgdisp.k: done
Erased graphic frame(s) 1-1
MAP: completed frame 1
cirrus:~/mcidas/data> frmsave.k 1 test.gif
Frame saved in /usr/homes/fingerhutb/mcidas/data/test.gif
cirrus:~/mcidas/data> grddisp.k MYDATA/GRIDS.4 GRID=3 NAV=C CINT=10 UNIT=KTS
grddisp.k: Done with graphic frame 1
GRDDISP - done
cirrus:~/mcidas/data> frmsave.k 1 test2.gif
Frame saved in /usr/homes/fingerhutb/mcidas/data/test2.gif
When I look at the overlay of the grid on top of the image representation
of the grid, it looks correct (I left test.gif and test2.gif in your
mcidas/data directory so you could take a look).
I verified that the version of grdimage.k your account uses is the current
one:
cirrus:~/mcidas/data> ls -alt /usr/local/mcidas/mcidas7.6/src/grdimage.pgm
-rw-r--r-- 1 mcidas udata 46691 Mar 27 2000
/usr/local/mcidas/mcidas7.6/src/grdimage.pgm
I believe that the correct GRID file from the MYDATA/GRIDS dataset is being
used. I say this because:
cirrus:~/mcidas/data> grdlist.k MYDATA/GRIDS.ALL FORM=FILE
DATASET NAME: MYDATA/GRIDS
Dataset Position Creation Date Max Grids Directory Title
---------------- ------------- --------- -------------------------------
1 2000286 159
2 2000279 50 scratch for modx
3 2000104 159
4 2000270 30 SCRATCH FOR JETSTREAM
10 2000271 159
40 2000270 30 SCRATCH FOR JETSTREAM
GRDLIST - done
This listing shows that the GRID file in MYDATA/GRIDS.1 has the creation
date of 2000286 and MYDATA/GRIDS.3 has a creation date of 2000270.
The image created by grdimage.k shows a date of 2000270 which matches
MYDATA/GRIDS.3:
cirrus:~/mcidas/data> imglist.k MYDATA/MIMGS.5
Image file directory listing for:MYDATA/MIMGS
Pos Satellite/ Date Time Center Band(s)
sensor Lat Lon
--- ------------- ------------ -------- ---- ---- ------------
5 DERIVED 26 SEP 00270 00:00:00 0 -1 1
imglist.k: done
Just to make sure, I did a listing of MYDATA/GRIDS.1:
cirrus:~/mcidas/data> grdlist.k MYDATA/GRIDS.1 NUM=ALL
Dataset position 1 Directory Title=
PAR LEVEL DAY TIME SRC FHOUR FDAY FTIME GRID PRO
---- --------- ------------ -------- ---- ----- ------------ -------- ----- ----
U MSL MB 12 OCT 00286 14:46:13 WAF 0 01 JAN 70001 00:00:00 1 NONE
V MSL MB 12 OCT 00286 14:46:13 WAF 0 01 JAN 70001 00:00:00 2 NONE
VR MSL MB 12 OCT 00286 14:46:13 WAF 0 01 JAN 70001 00:00:00 3 NONE
Number of grids listed = 3
These grids are totally different than the ones in MYDATA/GRIDS.4, so
there can be no mistaking them from the ones in MYDATA/GRIDS.4. The
same thing can be said for the grids in the other MYDATA/GRIDS members
(2, and 3) since they are CAPE products (MYDATA/GRIDS.2) and upper air
Z grids (MYDATA/GRIDS.3).
So, my conclusion is that the copy of grdimage.k that is used from your
account on cirrus is working correctly.
What am I missing?
Tom
>From address@hidden Fri Oct 13 13:17:42 2000
>cc: Mark Tucker <address@hidden>
>Subject: Re: 20000926: GRDIMAGE issues (cont.)
Tom,
I just tried a few things which should help solve this
problem.
1. I just ssh'd to cirrus and ran 'the' GRDIMAGE command.
Yes, GRDIMAGE grabbed the correct gridfile. I got what
you did, so you did nothing wrong.
When I reported that I ran McIDAS on cirrus, I did not
ssh to cirrus. I use an NT pc in my office with a vnc
viewer, that actually executes on elmo. GRDIMAGE grabs the
wrong gridfile when I execute on elmo. I ssh'd from elmo
to cirrus, then started mcidas. With this arrangement,
GRDIMAGE grabs the wrong gridfile.
So, using vnc GRDIMAGE grabs the wrong gridfile, and
without vnc it grabs the correct gridfile.
2. Now for the image just created by ssh'ing to cirrus. I
displayed it and overlayed contours. (I'm executing on
elmo thru vnc.) I believe the values are off by a constant
factor. I ran GRDINFO MYDATA/MYGRIDS.4 GRID=3 and I find
that the maximum value in the grid is .9482E02 at -45.00,17.50
(lat,lon); the units are shown as MPS, although I previously
multiplied the wind speeds by 1.94 (conversion vactor from
m/s to knots). In other words, I converted to knots but
did not change the units in the grid. In the infamous
GRDIMAGE command, values from 0 to 255 are supposed to be
mapped into 0 to 255; that is, the brightness values are
supposed to be equal to the raw grid point values. And
they are not. If I move the cursor to (-45,17.50), the
brightness value is shown at the bottom of the display
as 255. Could it be that GRDIMAGE is mapping the maximum
value in the grid to 255 and not using the RANGE= keyword?
3. When you overlayed contours with GRDDISP, you used the
keyword UNITS=KTS, which will multiply values by 1.94 .
Multiplying 94.82 by 1.94 gives 183.95 . When I look at
test2.gif, I see alot of contours, perhaps the largest
one is near 180. So the contours show a pattern similar
to correct and are understandable;
but, again, the image has a max brightness value of 255 when it
should be 94.82 . So, the image is wrong.
Whew!
Bill
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
< Bill Fingerhut, Professor PHONE: 802-626-6257 >
< Meteorology Dept FAX: 802-626-9770 >
< Lyndon State College >
< Lyndonville, Vt 05851 >
< >
< EMAIL: address@hidden >
< address@hidden >
< WWW: http://apollo.lsc.vsc.edu/ >
< >
< disclaimer: I know nothing - I only work here. >
vvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvv