Interpreting avulsion process from ancient alluvial sequences: Guadalope-Matarranya system (northern Spain) and Wasatch Formation (western Colorado)

Alluvial deposits of the Guadalope-Matarranya system (Oligocene, Ebro basin , Spain) and the Wasatch Formation (Eocene, western Colorado), provide time -integrated records of the process of river-channel avulsion, These sequenc es consist of isolated channel-belt sandstones incised into, and abruptly o verlain by, flood-plain siltstones, indicating deposition by avulsive river systems. The geometry and distribution of channel incisions suggest that a vulsion was not controlled by tectonics, climate, or base-level changes, bu t formed by autocyclic processes. Measurements from 221 channel fills in the Guadalope-Matarranya system and 38 from the Wasatch Formation allow us to statistically characterize channe l geometries me infer to be associated with establishment and abandonment o f individual river avulsions, Paleoflow depths in both systems average 1.4 to 1.6 m, Aggradation height (superelevation) of channel margin levees are, on average, 0.6 and 1.1 times paleoflow depth in the Guadalope-Matarranya and Wasatch systems, respectively, These results are consistent with values from recently avulsed modern rivers and suggest that (1) how depth is the appropriate parameter against which to scale the critical superelevation ne cessary for channel avulsion; and (2) the increase in potential energy due to channel perching drives the lateral instability that is needed for avuls ion to be successful. Numerous stacked channel fills indicate repeated reoccupation of the same s ite by avulsing channels, These reoccupation channels indicate that inherit ed hood-plain topography, here abandoned channel forms, was an important co ntrol on the arrival site of newly avulsed channels, Comparison of our results to others suggests two end-member types of avulsi on can take place, Incisional avulsion, seen here, is characterized by an e arly incision phase followed by infilling by migrating bar forms, Aggradati onal avulsion begins with aggradation followed in time by stream integratio n into a single downcutting channel, We suggest that the type of avulsion i s strongly influenced by whether or not the adjacent hood plain is well or poorly drained. In both cases subsequent aggradation and channel perching i ncrease the chances that some triggering event will lead to avulsion.