Terrace scarps can serve as a nearly ideal natural laboratory for the study of the evolution of slopes. This paper examines the effects of scarp size (height) and orientation (microclimate) by keeping constant variables such as age, lithology, and regional climate.

If a scarp degrades as a closed system, and downslope movement is directly proportional to surface gradient, the evolution of the scarp is modeled by the diffusion equation. For a group of scarps of same age and known starting angle, the diffusion-equation model predicts the relation between maximum scarp angle (𝛉) and scarp height (h). Late Pleistocene terrace scarps now as steep as 33.25°, as well as measured angles of repose for sand and gravel, require a starting angle as steep as 33.5°. For latest Pleistocene Idaho and Utah scarps, as h increases, 𝛉 is gentler (more degraded) than modeled by the diffusion equation with a constant rate coefficient. The degradation-rate coefficient (c) increases tenfold with scarp height; it should not change with scarp height if downslope movement is solely determined by surface gradient (to the first power). Soil wash appears to be responsible for this departure from the diffusion-equation model, for transport rate by soil wash is a function of scarp size (height).

South-facing scarps are less vegetated and more degraded than north-facing scarps. For scarps 2 m high, the degradation rate (c*) on S-facing scarps is 2 times that on N-facing scarps; for 10-m scarps, it is 5 times.

The observed dependence of the rate coefficient c* on scarp height can be removed by normalizing c* to values for west-facing scarps of the same height. The residual c* values calculated by this method correlate well with differences in incident solar radiation resulting from the different scarp orientations and maximum gradients. This correlation demonstrates the importance of orientation on slope processes and their rates through the differences in freeze-thaw cycles, soil moisture, and vegetative cover.

Scarp morphology may be used to estimate age, if one accounts for the effects of climate and for scarp height, orientation, and lithology. For example, using the dated Bonneville shoreline scarps for calibration and comparing only scarps of equal height, we estimate the Drum Mountains fault scarps to be 9,000 yr old. This age is about twice that produced by previous diffusion-equation calculations that have not accounted for the height as we have here, but it is the same as independent geologic estimates of their age.