[GRASS-user] need help computing all flow routing for large DEM

[Ken Mankoff]

Hi GRASS list,

I'm seeking some advice about how to determine the eventual edge-cell destination of every pixel in a DEM. I have an algorithm that works on rasters up to 0.5 billion pixels but am now trying to do it on a region about 8x as large and it is failing.

I've spent a week exploring r.watershed and r.terraflow and can't figure out how to make it work.

What has worked for the 0.5 B pixel region is running r.watershed and then for every negative pixel, I run r.water.outlet and find the upstream contributors. But I can't run r.watershed on the entire large region. I can run it quickly on sub-regions, but some very large interior basins cause this method to fail.

I've run r.terraflow, but the direction raster from that has a massive number of internal gaps and I don't see how to get it to work.

Does anyone have any advice how to process large rasters with r.watershed, or r.fill.dir, or some other method that can trace all pixels to their edge destination?

Thanks,

  -k.