Bill,

Your comments expanding on dynamic effective distance are right on the mark. As with all things GIS, the old adage that “what seems simple, ain’t” certainly applies. Treating slope as a static variable is often much too easy. While the consideration is a step above simple proximity “as-the-crow-flies,” it doesn’t encompass the real-world effects of effective distance—direction, accumulation and momentum. An old column (November, 1992) touched on issues surrounding dynamic effective distance…

“We all know that real movement involves a complex interaction of direction, accumulation and momentum. For example, a hiker walks slower up a steep slope than down it. And, as the hike gets longer and longer, all but the toughest slow down. If a long, steep slope is encountered after hiking several hours, most of us interpret it as an omen to stop for quiet contemplation.”

I have posted several of the earlier (ancient?) columns that describe the basic concepts and algorithms that are germane to the discussion… http://corp.pacificmeridian.com/basis2/, select Column Supplements, Beyond Mapping supplements. Most grid-based work for assessing dynamic effective distance utilizes a “look-up table” for updating the friction factors that is similar to a “turn-table” used in vector-based network analysis. For example, the supplement on effective distance algorithms notes that

“… under ideal circumstances you might hike three miles an hour in gentle terrain. When a "ring" encounters an adjacent cell which is steep (indicated on the slope map) and the movement is uphill (indicated on the aspect map), the normal friction is multiplied by the "friction multiplier" in the look-up table for the "steep-up" condition. This might reduce your pace to one mile per hour. A three-dimensional table can be used to simultaneously introduce fatigue-- the "steep-up-long" condition might equate to zero miles per hour.”

The notion that GIS modeling can be extended to trail and highway siting is a natural extension—that is if effective distance is unshackled from static assumptions as you suggest.

Yours truly,

Joe Berry

Berry & Associates

[jberry@innovativegis.com]