20050818: GeoTIFF to AREA conversion (cont.)

>From: "Valentijn Venus" <address@hidden>
>Organization: ITC
>Keywords:  200508172300.j7HN0Bjo011078 McIDAS MAKEAREA MAKNAV BSQ

Hi Valentijn,

re:
>as per your instructions i have tried to add the GLC map, but get the
>following error message:
>
>First renamed gl-goodes-1km-landcover.bsq rm into GLCgoodes stored in
>/home/mcidas/data (as "mcidas"):
>
>MAKEAREA GLCgoodes 9002 17347 40031 0 1 0 13
>MAKEAREA - Begin
>Program terminated, segmentation violation
>
>Can you try to reproduce this error? You can find this global land cover
>map at http://glcf.umiacs.umd.edu/data/landcover/data.shtml

First, it is evident that you understand the process, so please forgive
the verbosity of this reply.  My detailed explanations are more than you
need, but may help the next person...

The first thing is that this image you chose is in the Goode's
Homolosine projection, and that projection is not supported in McIDAS.
Here is the metadata for the global, 1km landcover image:

Information 
Image Size:                     40031 Pixels 17347 Lines
Quantization:                           8-bit unsigned integer
Projection:                             Goode's Homolosine (meters)
Earth Ellipsoid:                Sphere, radius 6370997 m 
Upper Left Corner:              -20015500.000 E 8673500.000 N 
Lower Right Corner:             20015500.000 E  -8673500.000 N 
Pixel Size:                             1000.000 E 1000.000 N 

I think you will have to use the Lat,Lon versions instead.

This is what they are referring to in the Description page:

http://glcf.umiacs.umd.edu/data/landcover/description.shtml

with a representative plot labeled 'Geographic coordinates projection'.

Second, I was unaware of the BSQ format described in:

  "Binary BSQ format: This version of the data set is in binary format,
   specifically in Band SeQuential format. In this version, the data is
   considered a binary image by most commercial spatial software."

I did a Google (tm) search on the BSQ format and see that it should
be readable by McIDAS directly.

Next, I downloaded the 1 degree Global Land Cover Data Set in Lat,Lon
projection, uncompressed it, made a link to it so it could be read
by MAKEAREA, imported it into an AREA file, and then added navigation:

<as 'mcidas'>
mkdir ~/avhrr
cd ~/avhrr

ftp ftp://ftp.glcf.umiacs.umd.edu
  <user> anonymous
  <pass> my_full_email_address
  cd /glcf/Global_Land_Cover/Global/gl-latlong-1deg-landcover
  binary
  get gl-latlong-1deg-landcover.bsq.gz
  quit

gunzip -c gl-latlong-1deg-landcover.bsq.gz

ln -s gl-latlong-1deg-landcover.bsq test.bsq

<in McIDAS>

REDIRECT ADD *.bsq "/home/mcidas/avhrr
DMAP *.bsq

MAKEAREA test.bsq 2000 180 360 0 1

I got the shape of the image from the metadata file for this image
which reads:

  Spatial Coverage:
  -The coverage is global. Data in file are ordered from North to South and
   from West to East beginning at 180 degrees West and 90 degrees North. Point
   (1,1) represents the grid cell centered at 89.5 N and 179.5 W. 
   (UL=90 degrees Lat, -180 degrees Long, LR=-90 degrees Lat, 180 degrees Long)
  
  Spatial Resolution: 
  -The data are given in an equal-angle latitude / longitude grid that has a 
   spatial resolution of 1-by-1 degree latitude / longitude.  
  
  Data Format:
  -The database is provided in both binary and ASCII format. 
  -The dimensions of the database are 360 pixels by 180 lines
  -Quantized as 8-bit
  
  Requested Form of Acknowledgment:
  -Please cite the following publication whenever these data are used: 
  
       DeFries, R. S. and J. R. G. Townshend, 1994, NDVI-derived land cover  
       classification at a global scale. International Journal of
       Remote Sensing, 15:3567-3586.


As you can see, the metadata also has the information necessary to add
navigation to the image, so I added it with MAKNAV:

MAKNAV 2000 RECT=1 90 1 -180 1 1
IMGDISP MYDATA/IMAGES.2000 MAG=2
MAP SAT
 
Now, since the image represents land use, its data values values lie in
a very small range of brightnesses.  To accentuate the information, I
used the interactive Black and White enhancement utility, EB:

EB

I played around with this to the point where I could see the image
features and then saved the display as a Gif image:

FRMSAVE 1 gl-latlong-1deg-landcover.gif

The results of this exercise can be seen at:

http://www.unidata.ucar.edu/content/staff/tom/gifs/gl-latlong-1deg-landcover.gif

I assume that you are looking for a higher resolution version of
the landcover product, so I imagine that you would want to try using
the 1 km image in Lat,Lon projection instead of the 1 degree version.
The problem with doing this is the number of elements (pixels) in the 1
km image is 43200.  This is too large for MAKEAREA which is coded with a
limit of 40000.  This limit is probably the reason for you getting a
segmentation violation when you tried running MAKEAREA in your test.

So, I recommend that you try using the 8 km image instead.

Since the RECT= navigation requires the lat,lon difference in image
points in degrees, you will need to calculate the dlat and dlon using
the metadata information on the image shape:

 ...
Number_Of_Lines=2880;
Pixels_Per_Line=5760;
Pixel_Order=NOT_INVERTED;
Pixel_Format=BYTE;
Pixel_Sign=UNSIGNED;
 ...

dlat = 180 / 2880 = 0.0625
dlon = 360 / 5760 = 0.0625

So the sequence is:

- download the 8 km Lat,Lon image in gzipped BSQ format
- uncompresse the file using gunzip
- make a link to a name to satisfy MAKEAREAs' 8.3 file naming restriction
- run MAKEAREA to put the image values into an AREA file
- run MAKNAV to add navigation

Here is what I did:

<as 'mcidas' in the ~/mcidas/avhrr directory>
-- downloaded the 8km Lat,Lon image in BSQ format
gunzip gl-latlong-8km-landcover.bsq.gz
ln -s gl-latlong-8km-landcover.bsq 8kmland.bsq

<in McIDAS>
LWU DEL AREA2001
MAKEAREA 8kmland.bsq 2001 2880 5760 0 1
MAKNAV 2001 RECT=1 90 1 180 0.0625 0.0625
IMGDISP MYDATA/IMAGES.2001 MAG=-6
EB
MAP SAT
FRMSAVE 1 8kmland.gif

The result can be seen in:

http://www.unidata.ucar.edu/content/staff/tom/gifs/8kmland.gif

I hope that this helped...

Cheers,

Tom
--
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.