PUNCTUATED SEDIMENT DISCHARGE DURING EARLY EVOLUTION OF THE COLORADO RIVER: EVIDENCE FROM REGIONAL STRATIGRAPHY AND SEDIMENTOLOGY

Late Miocene to early Pliocene deposits in the lower Colorado River (CR) corridor and western Salton Trough contain a record of sediment dispersal to the marine realm during the first ~0.5 m.y. of CR evolution. Abrupt first arrival of CR sand in the western Salton Trough is dated at ~5.3 Ma based on magnetostratigraphy and micropaleontology (Dorsey et al., 2007, 2011; this study), as recorded in sand-rich marine turbidites of the Wind Caves member of the Latrania Formation. The Wind Caves member is overlain by Coyote Clay, a regionally extensive marine claystone that records shut-down of sand delivery to the basin between ~5.1 and 4.8 Ma. Coyote Clay is overlain by a thick coarsening- and shallowing-up succession of marine sandstone and mudstone that records progradation of the CR delta and arrival of fluvial deposits (Palm Spring Group) at 4.25 Ma. Correlative deposits along the lower CR south of Blythe, CA, include: (1) Bouse Formation basal carbonate which accumulated in a transgressive intertidal to subtidal marine embayment at the north end of the Gulf of California; (2) Bouse green claystone coarsening up to thick cross-bedded CR channel sandstone that accumulated in the earliest through-flowing Colorado River; (3) upper Bouse bioclastic limestone, micrite, and fan gravels that formed in a large saline lake or marine estuary; and (4) cross-bedded pebbly sandstone (Bullhead Alluvium) that accumulated in the CR channel ca. 4.5-3.5 Ma (Howard et al., 2015). Bouse upper limestone records re-flooding of the lower CR valley after the river first flowed through it, and may correlate to limestone that overlies the 4.83-Ma Lawlor Tuff north of Buzzards Peak. Our stratigraphic synthesis suggests that dispersal of sand through the lower CR valley into the marine basin turned ON at ca. 5.3 Ma, turned OFF at ~5.1 Ma, and turned ON again at ~4.8-4.7 Ma. If the two stratigraphic records do not correlate to each other, the implied sequence of events would be more complex but not implausible. The discontinuous, start-and-stop history of sediment discharge is not predicted by existing models for CR initiation, and could have been controlled by variations in sediment discharge from the upper catchment (Colorado Plateau), changes in eustatic sea level, and/or changes in rate of fault-controlled subsidence along the lower CR corridor.