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

19990524: MCIDAS neartime satellite data archive (cont.)

>From: Dave Dempsey <address@hidden>
>Organization: San Francisco State University
>Keywords: 199905241722.LAA00737 Unidata-Wisconsin imagery
 
Dave,

>A couple of comments about the GOES-10 visible images in area format from Mars
> hall Space Flight Center's ftp server:
>
>   (1) they are archived for about 24 hours only;

OK.  I wasn't aware of that.

>   (2) they do include images every 15 minutes (usually);

Right.

>   (3) after uncompressing them, the image files are generally two to four tim
> es larger (in some cases 10 times larger) than the visible image files we get
>  from SSEC;

Right.  The images from Marshall contain 2-byte (to store 10-bit) values.
They are also not sampled to get rid of the 1.7 times oversampling in the
element dimension.  These two features will yield a file size that is 4
times that in the UW stream.  Files that are larger most likely cover
a larger area than the ones in the UW stream.

>   (4) although the files are stored in a sub-directory called "4km" on the ft
> p server, they look to my eye like they have higher resolution than the 4km i
> mages we get from SSEC;

This is the oversampling in the element dimension.  The images in the UW
stream have been sampled to eliminate this (original GOES hemispheric images
will be egg shaped with the "top" - "bottom" diameter of the egg being 
coincident with the equator.

>   (5) the range of pixel values seems to be much smaller than the range we se
> e in the SSEC area files, requiring a different color-fill specification to p
> lot the images (which makes it difficult to loop them with SSEC images, among
>  other inconveniences).

The range of values is actually much greater.  The images in the UW stream
are 8-bit; the ones from Marshall are 10-bit.  It is possible that the
software that you are using to load the images is incorrectly interpreting
the 2-byte values.

I did send in an inquiry to SSEC to see if they would be willing to back
fill the UW datastream archive.  I havn't received a reply from them on
this yet.

Tom