Fluvial terraces contain information about incision, deformation, and climate change. In this study, a chronostratigraphic record of Colorado River terraces is constructed from optically stimulated luminescence (OSL) dating of Pleistocene alluvium and real-time kinematic (RTK) GPS surveys of terrace form. This record is analyzed to relate terrace formation to late Pleistocene climate fluctuations, and terrain analyses and longitudinal profile patterns reveal recent salt-related activity in the northern Paradox Basin as well as patterns in Colorado Plateau incision. A well-preserved, correlative suite of mainstem (M) fluvial deposits exists along the Colorado River upstream of Moab, Utah. Absolute dates indicate sedimentation >70 ka (M7, M6/M5), 70-50 ka (M4), 50-40 ka (M3), and 35-25 ka (M2). The M4 and M2 formed during the crescendo to glacial maxima, but the M7, M6/M5, and M3 were deposited during variable climate of marine isotope stages (MIS) 5 and 3. Deposits include thin (<7 m) strath terraces and thick (10-20 m) fill terraces. Our results suggest that terrace sedimentation is linked to enhanced sediment flux during glaciations in Rocky Mountain headwaters (M4 and M2), but major deposits also formed during dryland tributary sediment loading with markedly different timing (M6/M5 and M3). Conversely, incision may be driven by higher deglacial flows. Clast provenance data demonstrate greater percentages of locally-sourced sediment in M6/M5 and M3 deposits. Valley-bottom geometry and neotectonics control terrace form, with strath terraces found in bedrock-restricted reaches and fill terraces in wider valleys. Previously speculated salt deformation in this area is confirmed by localized collapse preserved in M4 stratigraphy in the Cache Valley graben and ~15 m of broader subsidence upstream. Concavity and knickzone distributions in tributary profiles are discordant and represent subtle expressions of salt-tectonic activity. Finally, a surprisingly rapid incision rate of ~900 m/Ma over the past ~70 ka suggests that the Colorado River may be responding to flexural rebound in the central plateau, but is faster than that predicted by the debated bull's-eye pattern of regional incision. This locally high rate may also reflect a transient wave of incision, as suggested by increased Pleistocene rates interpreted by studies in Glen and Grand canyons.