[GRASS-user] r.watershed, blocking, and max_slope_length

[Ken Mankoff]

Hi List,

I can't seem to get blocking or max_slope_length to work as I would expect. I see no change to the output when I use them.

I'm calling =r.watershed= like this:

r.watershed -s -a --v --overwrite --verbose elevation=z_s flow=RU85 blocking=z_2000_r thresh=50 max_slope_length=11000 accumulation=tmp.acc

Where =v_200_r= is a raster of a single contour at 2000 made with:

r.contour input=z_s output=z_2000_v levels=2000 --o
v.to.rast --overwrite --verbose input=z_2000_v type=line output=z_2000_r use=z

Based on the =r.watershed= documentation, I would expect two things:

1. All flow routing stops at 2000 m and then restarts below.

2. No flow continues more than 11 km, which is not far because I'm on a 5 km grid.

Instead I see the exact same results as if I did not use either of those keywords.

Although I'd like to understand these keywords, I'm also curious if my planned usage makes sense. I need to route all surface water 1 cell downstream as part of this analysis (actually 1 cell downstream if slope is < 0.5). I plan to do this by generating an accumulation map with max distance of 7.5 km (on a 5 km grid). I assume this routes everything only 1 cell, whether 90 degrees (5 km) or 45 degrees (7.07 km). Then I subtract this accumulation map from the original, and the result is the input flow map but moved 1 cell downstream.

Thanks for any help you can provide,

  -k.

P.S. Using grass 7.0.3 on OS X installed with 'brew' and no Framework packages. Works quite nicely.