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

[McIDAS #MOI-385613]: ADDE



Hi again Ioan,

re:
> Well, the image size is correct inside the directory. But the resolution
> does not seem to be, at least to me.
> 
> For example the directory for band 2 says it has 1 km resolution and a size
> of 21696x21696
> 
> If you look inside any netcdf file  for channel 2 for Rad variable yu can
> see
> 
> Rad:resolution = "y: 0.000014 rad x: 0.000014 rad" ;
> 
> These value for microradians corresponds to the resolution of 0.5 KM.
> McIDASX reports this correctly as you showed.
> 
> So both the NetCDF and McIDASX show the correct resolution but the
> directory shows 1 KM form band 2 as well as for the other bands (5,7,13,15)
> 
> Then again the definition of the directory struct from
> 
> https://www.ssec.wisc.edu/mcidas/doc/prog_man/2015/formats-1.html
> 
> says
> 
> The data in the directory is stored as 32-bit (4-byte) twos- complement
> binary integers or as ASCII characters.
> 
> so there is no way to write 0.5 into the header. This is why I suspected
> they multiplied the resolution by 2 :)))
> 
> This actually checks out as the real resolution in KM for bands is:
> 
> 2:0.5
> 5:1
> 7:2
> 13:2
> 15:2
> 
> while the directory reports
> 
> 2:1
> 5:2
> 7:4
> 13:4
> 15:4
> 
> As you pointed out, there is no 4 km data.
> 
> Maybe this helps a bit

Ah Ha!  I see the problem, and it is with poor wording for the meaning
of AREA directory words (1-based) 12 and 13.  A little better explanation
for the meaning of these words is:

    W12.  LINERES:  Line Resolution; spacing in image-lines between
                    consecutive area lines
    W13.  ELERES:   Element Resolution; spacing in image-elements
                    between consecutive area elements

Even this is phrased in a misleading manner.  The correct interpretation
is that words 12 and 13 are multiplicative factors for the base
resolution of the instrument, and the base resolution for GOES-16
imagery is defined in the McIDAS SATBAND file (BRes):

Sat 186
Cal ABIN 
BRes 0.5 0.5
1 DESC='0.47 um VIS aerosol-over-land' WN=21266 cm-1 WL=0.47 um
2 DESC='0.64 um VIS clouds fog, insol, winds' WN=15712 cm-1 WL=0.64 um
3 DESC='0.86 um Near IR veg/burn scar,aerosol, wind' WN=11575 cm-1 WL=0.86 um
4 DESC='1.37 um Near IR cirrus cloud' WN=7282 cm-1 WL=1.37 um
5 DESC='1.61 um Near IR cloud phase, snow' WN=6215 cm-1 WL=1.61 um
6 DESC='2.24 um Near IR land/cloud, vege, snow' WN=4459 cm-1 WL=2.24 um
7 DESC='3.89 um IR Sfc, cloud, fog, fire, winds' WN=2570 cm-1 WL=3.89 um
8 DESC='6.17 um IR High-level WV, winds, rainfall' WN=1621 cm-1 WL=6.17 um
9 DESC='6.93 um IR Mid-level WV, winds, rainfall' WN=1444 cm-1 WL=6.93 um
10 DESC='7.34 um IR Lower-level WV, winds & SO2' WN=1363 cm-1 WL=7.34 um
11 DESC='8.44 um IR Total WV cloud phase, dust' WN=1184 cm-1 WL=8.44 um
12 DESC='9.61 um IR Total ozone,turbulence,wind' WN=1041 cm-1 WL=9.61 um
13 DESC='10.3 um IR Surface & cloud' WN=968 cm-1 WL=10.3 um
14 DESC='11.2 um IR Imagery,SST,clouds,rainfall' WN=894 cm-1 WL=11.2 um
15 DESC='12.3 um IR Total water, ash, and SST' WN=815 cm-1 WL=12.3 um
16 DESC='13.3 um IR Air temp, cloud hgt and amt' WN=754 cm-1 WL=13.3 um

So, the actual resolutions for the bands are:
Band     BRes  Resolution
---------+----+----------
2  - 1 x 0.5 = 0.5
5  - 2 x 0.5 = 1
7  - 4 x 0.5 = 2
13 - 4 x 0.5 = 2
15 - 4 x 0.5 = 2

This used to trip me up all of the time!  And, as you surmised, this
was done because there is no way to represent a fractional number with
an integer ** without including/assuming a scale and offset **.  So,
in this case, words 12 and 13 must be interpreted as scale factors and
the offsets are zero.

It is interesting that you brought this up.  I am facing the exact
same issue when dealing with the SCMI format GOES-16 imagery that
is being delivered in NOAAPort.  There, all Full Disk images have
been sampled down to 6 km resolution, but there is no reference to
the base resolution of the instrument (ABI that is).

I hope that this helps!

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: MOI-385613
Department: Support McIDAS
Priority: Normal
Status: Closed
===================
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.