The Pan-Amazonian Ucayali Peneplain, late Neogene sedimentation in Amazonia, and the birth of the modern Amazon River system

Abstract

We review the Neogene geologic history of lowland Amazonia in an attempt to focus attention on areas of agreement, as well as areas in dispute, in this research arena. We reinterpret pre-existing hypotheses, present new data, and discuss new insights intended to support a unified synthesis of the Amazon Basin as a single sedimentary basin, albeit on a vast scale, during the late Miocene to middle late Pliocene. We document the Ucayali Peneplain as an isochronous, Pan-Amazonian geologic feature that formed following the early to mid-Miocene Quechua I orogenic phase of Andean tectonism. Peneplanation began possibly as early as ∼ 15 Ma and terminated abruptly near the beginning of the late Miocene Quechua II orogenic event at ∼ 9.5–9.0 Ma. Subsequently, a thin cover of sediments comprising the Madre de Dios Formation began blanketing most of lowland Amazonia, excepting only the eastern Subandean Fold-and-Thrust Belt and isolated highlands within the basin. The buried peneplain is readily observed in river cutbanks throughout Amazonia as the marked, often angular Ucayali Unconformity that separates eroded, older, often folded, faulted, and weathered, moderately to well consolidated Tertiary formations from unconsolidated, near horizontal, upper Neogene deposits. The dates of formation of major unconformities and subsequent depositional events at widely separated areas within the Andes of Bolivia, Ecuador, and Peru are coincident with that of the Ucayali Unconformity and deposition of the Madre de Dios Formation and suggest that the events are linked to a common cause, which is interpreted to be the still on-going collision between the South American and Nazca tectonic plates. The Madre de Dios Formation has three members, the oldest of which documents a short-lived, high energy depositional environment followed by a moderate-energy depositional environment, both occurring at a time when drainage from the basin was unobstructed. The upper two members record fluctuations between moderate and low energy continental depositional environments during a period when drainage from the basin was obstructed, disorganized, and took place over long distances with extremely low gradients. The sedimentology of the Madre de Dios Formation, particularly the thick, massive beds of clay, and the widespread presence of paleodeltas and associated geomorphic features on the Amazonian planato are consistent with the hypothesis that much of the upper two members formed as lacustrine and deltaic deposits within a gigantic lake, Lago Amazonas, or, more probably, within a complex series of interconnected mega-lakes that occasionally united to cover most or all of lowland Amazonia to a shallow depth from the latest Miocene until ∼ 2.5 Ma. The presence of the Ucayali Unconformity and the relatively uniform lithostratigraphy basin-wide of the fluvial, fluviolacustrine, and lacustrine sediments of the upper Neogene Madre de Dios Formation are consistent with the hypothesis that the Amazon Basin acted as a single, undivided sedimentary basin in the late Neogene. The biostratigraphic correlation across important modern drainage divides of both micro- and macro-sized, late Miocene fossil vertebrates recovered from basal conglomerates of the Madre de Dios Formation, and the absence therefrom of fossil vertebrates of any other age, is also consistent with this hypothesis. Two 40Ar/ 39Ar dates on ash deposits within the Madre de Dios Formation corroborate the upper Miocene age of the basal horizons of that formation indicated by fossil vertebrates and support an upper Pliocene age for the youngest sediments of the formation. The modern Amazon River drainage system was established in the late Pliocene, at ∼ 2.5 Ma, by the breaching of the eastern rim of the sedimentary basin as a result of the basin being overfilled, or by headward erosion of the lower Amazon River, or both. Data on Cenozoic mass accumulation rates and the chemistry of terrigenous sediments in the Atlantic Ocean obtained by Ocean Drilling Project Leg 154, Ceara Rise support the postulated timing of the establishment of the modern Amazon River drainage system at ∼ 2.5 Ma, rather than the long-held view that this event occurred in the late Miocene. We discuss the important role of ocean currents and sea level fluctuations on terrigenous mass accumulation rates on the Ceara Rise. The postulated time of formation of the modern Amazon River is nearly coincident with the onset of the Plio-Pleistocene glacial climatic regime and the lowest sea level stands since the latest middle Miocene. This analysis indicates that modern Amazonia is a product of terrain development within an erosional regime since ∼ 2.5 Ma.