[GRASSLIST:3132] Re: [GRASS5] GRASS and MODIS satellite data: ISI N projection support?

[Markus Neteler]

On Wed, Feb 13, 2002 at 03:10:50PM -0500, John Gillette wrote:
> Where does  "select_vNsds" come from?  Is this source code from
> another program or from the reprojection tool?

I found it somewhere on a HDF-related NCSU page (have to dig
for the URL) and fixed the Makefile for Linux.
 
> I assume this allows you to select a specific band which you then import
> with r.in.bin.
Exactly. But the problem is that the MODIS is in that ISIN projection.
Means, I can see the data in a xy location with the ISIN deformation
making it useless. If GRASS supported ISIN (through PROJ4), I could
build a ISIN location and reproject later with r.proj.

The MODIS reprojection tool only eats 500m and 100m products, but not
the 250m MODIS products (tried and confirmed by USGS). So I have to
develop my own solution. Merging "select_vNsds" into r.in.bin would
solve the HDF/HDF-EOS problem for GRASS (I have already a scripted solution
running). Only the ISIN problem remains.

The ISIN formula I have in C, only I don't know how to hack the PROJ4.
On the related mailing list only silence...

Markus