[GRASS-user] r.watershed & massive dataset

[Hamish]

ivan marchesini wrote:
> I'm trying to use r.watershed to obtain dainage and stream map of a
> very large dem..
> The region containing the dem is of around 260.000.000 of cells but
> masking out null cells (with a raster MASK) the number of residuals dem
> cells is of around 140.000.000.
> I have tried without the -m flag but the process stops after few
> minutes
> due to memory allocation problems (the computer has only 2 GB of RAM). 
> Now I have started the process with the -m flag....
> 
> Do you have any suggestion to optimize the process?


r.terraflow is better for massive datasets.


Hamish



      ____________________________________________________________________________________
Looking for last minute shopping deals?  
Find them fast with Yahoo! Search.  http://tools.search.yahoo.com/newsearch/category.php?category=shopping