Evidence for lahar-triggering mechanisms in complex stratigraphic sequences: the post-twelfth century eruptive activity of Cotopaxi Volcano, Ecuador

Cotopaxi volcano is situated in the Eastern Cordillera of the Ecuadorian Andes and consists of a symmetric volcanic cone that reaches an altitude of 5,897 m above sea level; it is capped over its upper 1,000 m by a permanent glacier. The volcano has erupted frequently in the past few centuries and, according to the archival records, has produced dozens of lahars by catastrophic snow and ice melting during eruptions. In this work, we present a detailed map and a stratigraphic study of the lahar deposits of the past 800 years in two different topographic settings. A thorough knowledge of the tephrostratigraphy of the explosive activity over the same time period was a first-order prerequisite for the complete reconstruction and dating of lahar activity and also allowed us to precisely link lahar units to eruptive phases of individual eruptions. Results indicate that, during the thirteenth to seventeenth centuries, highintensity eruptions (Plinian events or blast-like explosions) produced large debris flows that transported meter-sized boulders. A subsequent period of activity that started in 1742 was characterized by several lahar-generating eruptive episodes that were smaller in scale but with significant variability in size (the 1877 being the smallest and most recent). Analysis of events occurring in the eighteenth century suggests that eruption style affects the volume and energy of the resulting lahars, with different pyroclastic flow types causing different mechanisms of water release from the summit glacier. Lahars produced during this time period were triggered by: (1) dilute pumice and ash-rich radially distributed density currents and (2) column collapse-related radially distributed scoria and lithic-rich pyroclastic-flows. The former produced lahar deposits that are matrix-rich, block-poor, and valley-confined, while the high erosive capacity of the latter produced lahars that are block-rich, highly energetic, and widespread. The youngest (1853 and 1877) lahars were triggered by (3) confined scoria-flow lobes that had less capacity to scour and melt the glacier; resulting flows had lower energy and smaller volumes than lahars produced by eighteenth-century eruptions. We conclude that the dynamics of pyroclastic–density–current can exert a major control on the size and destructive capacity of lahar at ice-capped volcanoes. Moreover, the total extent of the glacier at the moment of eruption, which is commonly considered to exert a major control on lahar formation, may actually be a second-order factor compared with the way in which the eruptive products interact with the glacier.