Morphologic analysis of scarp degradation can be used quantitatively to determine relative ages of different scarps formed in cohesionless materials, under the same climatic conditions. Scarps of tectonic origin as well as wavecut or rivercut terraces can be treated as topographic impulses that are attenuated by surface erosional processes. This morphological evolution can be modelled as the CQ!lvolution of the initial shape with erosion (or degradation) function whose width increases with time. Such modelling applies well to scarps less than !Om high, formed in unconsolidated fanglomerates. To a good approximation, the degradation function is Gaussian with a variance measuring the degree of rounding of the initial shape. This geometric parameter can be called the degradation coefficient. A synthetic experiment shows that the degradation coefficient can be obtained by least squares fitting of profiles levelled perpendicular to the scarp. Gravitational collapse of the free face is accounted for by assuming initial scarp s~opes at the angle of repose of the cohesion less materials (30° -35°). Uncertainties in the measured profiles result in an uncertainty in degradation coefficient that can be evaluated graphically. Because the degradation coefficient is sensitive to the regional slope and to three-dimensional processes (gullying, loess accumulation, stream incision. etc.), a reliable and accurate determination of degradation coefficient requires several long profiles across the same scarp. The linear diffusion model of scarp degradation is a Gaussian model in which the degradation coefficient is proportional to numerical age. In that case, absolute dating requires only determination of the propotionality constant, called the mass diffusivity constant. For Holocene scarps a few meters high, in loose alluvium under arid climatic conditions, mass diffusivity constants generally range between I and 6 m2/kyr. Morphologic analysis is a reliable method to compare ages of different scarps in a given area, and it can provide approximate absolute ages of Holocene scarplike landforms.
lNTRODUCfiONThe geomorphic expression of an active fault is usually characterized by offsets of features such as rivers, ridges, terrace levels and terrace risers, or moraines or simply of the sloping topographic surface. The kinematics of an active fault can be constrained by measurements of such offset features, combined with dating. Measurement of offsets is a relatively simple and easy task, especially with modem digital geodetic instruments. Dating of the morphology is less straightforward. In arid and semiarid regions, organic material (such as peat or charcoal) is rare, which makes 14C dating a painstaking endeavour. Based on the use of other cosmogenic isotopes (such as 36Cl, lOBe or 26Al), of thermoluminescence or of desert varnish development, new techniques may soon allow easier accurate dating. Morphologic dating is an alternate method that derives from the simple observation that the shape of a scarp is a function of its age [Wall...