Landform characterization using Ruhe's nine slope shapes in GIS

In the paper "Landform characterization with GIS" from February of
1997 issue of the Journal of Photogrammetric Engineering and Remote
Sensing (and also earlier on GIS-L-search in GOOGLE group:
comp.infosystems.gis under Blaszczynski landform characterization) I
had an opportunity to suggest that continuous digital surfaces which
represent terrain morphology can be broken down into semantic (or, as
some others suggested: "cognitive") units that can be applied toward
developing better understanding of geomorphic processes. The units
proposed included: generally upper convex hillslopes and their crests,
generally lower concave slopes and their troughs (these included open
valleys and closed basins with interior drainage), and sloping as well
as horizontal flats. Much of what was proposed was based on dividing
terrain surface into "slope facets", particularly as they are defined
by their convexity or concavity. The centerpiece of the paper was a
"negative average slope" method for deriving information on the
concavity, convexity, or flatness of the surface by utilizing a moving
neighborhood (focal area) that calculated both the positive (upwards
from the center cell) and negative (downwards from the center cell)
slope gradients and then took their average. The method permitted
generalization of the features through expansion of the size of the
neighborhood and was originally based on nine basic combinations of
shapes found through empirical trials using a 3 x 3 cell neighborhood
and simple numbers. Around the time of the publication of the paper
the author became aware that these nine combinations of shapes were
very similar to the breakdown of slope shapes suggested by a
gemorphologist R. V. Ruhe in his publications from 1967 and 1975 (see
references section located at the internet address:
http://www.blm.gov/nstc/ecosysmod/reference.html ).



The above picture borrowed (with some changes) from Ruhe's 1975
publication shows the nine main slope shapes which he obtained by
considering the planimetric and profile curvatures of a
three-dimensional portion of a hillslope. Planimetric curvature is
the curvature of slope parallel to the contour lines, or, in geologic
terms, the strike of a hillslope. Profile curvature is the curvature
of a slope that is perpendicular to contours, or in the direction of
the dip of a hillslope.

Obtaining a map Ruhe's slope shape combinations is relatively easy to
accomplish in the context of Geographic Information System
capabilities and digital representations of terrain surfaces. I've
prepared an ARC MACRO LANGUAGE tool which is in public domain which I
am planning to put on the web at some future point but would like to
have it tested by other interested parties. I have a sequence of
potential applications of the tool but would welcome hearing any
additional comments and suggestions.

If you are interested in this AML send a message to my work email
jblaszcz@blm.gov requesting the AML with a short description of how
you think it might help you or others you know in their work. We
might also discuss it in the group.

Thank you,

Jacek Blaszczynski
jblaszcz@blm.gov