DESERT PAVEMENT EVOLUTION OVER DUAL TIMESCALES: MODELING AND FIELD CALIBRATION

Desert pavements are widely used as a relative surface-dating tool because they are progressively more well-developed on surfaces ranging from a few thousand to hundreds of thousands of years old. More recently however, researchers have suggested that desert pavements are more dynamic. Healing experiments have shown that pavements reform over time scales of decades to centuries. Also, it has been suggested that in parts of the Mojave Desert above 400 m, all pavements were destroyed by latest Pleistocene vegetation advances, “re-setting” all desert pavement evolution in the Mojave above that elevation. In an effort to reconcile these seemingly disparate observations, we have developed an empirical-numerical model that captures the dual time-scale nature of desert pavement development and healing. To calibrate our model we conducted detailed geomorphic mapping, including pavement clast point-counting and integrated pavement-healing experiments from existing studies to capture the annual and millennial timescales. Our field work in Amargosa Valley, NV and CA highlights important differences in pavement evolution between sand-dominated and gravel-dominated parent materials. Our model treats desert pavement as a system that co-evolves with the underlying cumulic eolian epipedon over millennial and longer timescales. We emphasize the rate-controlling importance of development of the fine-grained inflationary eolian layer that is ubiquitous beneath well-developed desert pavements, and observe that “pavement-healing” from episodic disturbance occurs much more rapidly than the long-term development of the entire desert surface system. Furthermore, we conclude that the more well-developed the entire desert-surface system is, the more rapidly desert pavements are able to reorganize in response to transient disturbances.