Timing Mismatch Between Facies Change and Provenance Change in Large River Systems: Implications for Reservoir Development in Evolving Catchments

Abstract

This paper explores the relationship between facies and provenance in large river systems and posits that in major drainage reorganisations, the water from the newly formed catchment (as reflected in depositional facies) can pre-date delivery of sand-grade material (reservoir sands transported as bedload) by periods of 1-10 Ma, potentially causing a major change in reservoir properties without a facies change. The main part of this work is derived from fieldwork on the Pliocene palaeo-delta of the Colorado River, which was deposited in the Fish Creek-Vallecito Basin of southern California. The delta deposits are well-exposed in the badlands of Anza-Borrego Desert State Park and record a complete delta cycle from pro-delta turbidites to delta plain deposits. There is a well-studied contrast between sands derived from the Colorado River (C-Suite) and locally derived sands (L Suite). The locally derived sands are lithic, while the C-Suite sands are quartzose; the two suites can also be distinguished on heavy mineral population and varietal mineral geochemistry. L-Suite sands underlie, are laterally equivalent to, and overlie C-suite Colorado deposits, recording initiation, deposition, and eventual abandonment of this first Colorado delta. The lower provenance change is transitional and takes place within the Pliocene Wind Caves member of the Latrania Formation. This change has been dated as younger than 5.33 Ma and used to support the lake-spillover hypothesis for initiation of the lower Colorado River and formation of the Grand Canyon. However, the main facies change is between locally derived subaerial conglomerates of the Split Mountain Group and well-sorted marine turbidites of the Lycium Member of the Latrania Formation; this change is well dated at 6.24 Ma (Messinian, Miocene). Since the depositional facies of the 100 m thick Lycium Member are consistent with those of the 200 m thick Wind Caves Member, we infer that Colorado water (representing the connected drainage) arrived in southern California about one million years before arrival of the bedload (C-Suite) sands. We will consider other examples of this phenomenon from Sakhalin (Neogene Amur River delta) and Antarctica (Permo-Triassic fluvial deposits) and draw general lessons for the effects of provenance change on hydrocarbon reservoirs that will not be apparent from facies analysis or petrophysics alone.

AAPG Datapages/Search and Discovery Article #90259 ©2016 AAPG Annual Convention and Exhibition, Calgary, Alberta, Canada, June 19-22, 2016