Improving the facies models for syn-eruptive fluvial succesions: Lessons from the Chaitén volcano and Blanco river, Chile.

UMAZANO, ALDO MARTÍN; MELCHOR, RICARDO NÉSTOR; BEDATOU, EMILIO; JAVIER MARCELO KRAUSE

Paraná

Congreso; XVII Reunión Argentina de Sedimentología-VIII Congreso Latinoamericano de Sedimentología; 2021

Asociación Argentina de Sedimentología

The stratigraphic record of syn-eruptive, fluvial successions without marine connection is relatively well stablished after Smith’s (1991) pioneer ideas. It is typically pyroclastic-rich and characterized by short-lived, braided fluvial channel bodies, which are interbedded with floodplain strata having common primary pyroclastic deposits and buried paleosols. In comparison with the inter-eruptive fluvial successions, the channel deposits have larger width/thickness ratio and composing facies that suggest shallower flows with high sediment concentration and minor lithological diversity. The channel deposits of inter-eruptive intervals register meandering patterns. These features are classically related with the pyroclastic sediment supply. Nevertheless, recent contributions had to evaluate other auto and allocyclic signals to make consistent paleoenvironmental interpretations. In comparison with this conceptual background, the present contribution shows important discrepancies in the syn-eruptive sedimentary record generated from the interaction between the Blanco River and the recent rhyolitic eruption of the Chaitén Volcano (Chile). During May 2008, the explosive eruption of this volcano affected three adjacent fluvial basins, in which there are short trunk rivers connected with the Pacific Ocean. In particular, the Blanco River catchment (~77 km2) was overfeeded with pyroclastic sediments that together with rains, steep topographic gradients and forest vegetation, triggered sudden geomorphic modifications and the generation of an episodic sedimentary record. Main perturbations included flooding, changes in dimension and pattern of fluvial channels, avulsion and later filling of abandoned channels, and coast-line progradation. In this syn-eruptive scenario, four main differences with the conceptual background are detected. First: An abandoned channel with meandering pattern, which was probably active no later than May 20 when the avulsion occurred. Then received episodic flows and fully filled with a succession up to ~2.75 m thick of reworked lapilli and ash (both pumice-rich) mainly deposited from diluted and hyperconcentrated flows. This abandoned channel was lateral to the new and active braided channel-belt that transports suspended fine-grained pyroclastic sediments, and paleovolcanic components with low to nule vesicularity as bedload. Therefore, the generation of stratigraphic intervals with coeval occurrence of meandering and braided fluvial channel deposits in syn-eruptive conditions can occur. Second: The presence of up to 1.50 m high eolian dunes (nebkhas) with superimposed ripples, both composed of well-sorted pumice-rich fine-grained ash on the abandoned fluvial channel. Their genesis represents wind reworking of recently deposited and loose pyroclastic substrates developed in a zone with wet climate (mean annual precipitation ~3300 mm year-1). Thus, the eolian dune deposits in syn-eruptive successions may also be present and does not necessarily represent a climatic proxy linked to aridity. Moreover, the eolian deposits could denote an additional feature of the mock-aridity hypothesis. Third: There are abrupt compositional changes in alluvial deposits along the longitudinal profile of Blanco River. The near-vent positions are characterized by the presence of volcaniclastic sand and gravel deposited from debris flows and dilute currents, which are replaced downstream by reworked ash and lapilli mainly deposited from hyperconcentrated and dilute flows. These proximal-distal changes are related to the temporal storage of sediments in logjams and generation of sediment-laden flows by logjam breakup or overflow. In these situations, the different floatability of pumice (lapilli and ash) and dense fragments (gravel and sand) generates that the latter were trapped in the logjams and preferentially concentrated upstream. Moreover, there are similar compositional variations in the deposits underlying those of the 2008 eruption, which suggest recurrent hydrosedimentary behaviour of the fluvial system. In this way, the pyroclastic-rich lithological homogeneity as important feature to recognize syn-eruptive fluvial successions should be carefully used. Fourth: Deposits of the delta plain are composed of volcaniclastic sand deposited from dilute flows with high and low sediment concentration. This compositional change is related to the mixing of sediments and represents a feature linked to the marine connection. As a summary, the Smith model is a good starting point to recognize and analyze syn-eruptive fluvial successions, but we must remember that there are discrepancies and each depositional system is unique.