If he slowed down the water so that it has a chance to pecolate, it might help. But that only expands the capacity, and there may be days when that overflows.

If he slowed down the water so that it has a chance to pecolate, it might help. But that only expands the capacity, and there may be days when that overflows. I think Lancaster talks about the perpendicular flow, and also talks about ways to mitigate that with curves (as well as discussion about erosion patterns when water exits a structure too quickly, which in turn reshapes the terrain, sometimes going into degenerative erosion patterns). On-contour swales (or trenches) are good examples, especially when there are multiple tiers with overflow to increase the distance the water has to travel as it snakes back and forth across the contour and descends down a slope. There are multiple chances to slow down water and let it try to infiltrate. In his latest edition, Landcaster talks about gambesons (retaining walls), which he used to advocate. But because of long-term erosion patterns, those can fail pretty quickly.