A Conceptual Model of Future Volcanism at Medicine Lake Volcano, California – With an Emphasis on Understanding Local Volcanic Hazards

Medicine Lake Volcano (MLV) is most voluminous of all the Cascade Volcanoes (~600 km3), and has the highest eruption frequency after Mount St. Helens. Detailed mapping by USGS colleagues has shown that during the last 500,000 years MLV erupted \u3e200 lava flows ranging from basalt to rhyolite, produced at least one ash-flow tuff, one caldera forming event, and at least 17 scoria cones. Underlying these units are 23 additional volcanic units that are considered to be pre-MLV in age. Despite the very high likelihood of future eruptions, fewer than 60 of ~250 mapped volcanic units (MLV and pre-MLV) have been dated reliably. A robust set of eruptive ages is key to understanding the history of the MLV system and to forecasting the future behavior of the volcano. The goals of this study are to 1) obtain additional radiometric ages from stratigraphically strategic units; 2) recalculate recurrence rate of eruptions based on an augmented set of radiometric dates; and 3) use lava flow, PDC, ash fall-out, and lahar computational simulation models to assess the potential effects of discrete volcanic hazards locally and regionally. We identify undated target units (units in key stratigraphic positions to provide maximum chronological insight) and obtain field samples for radiometric dating (40Ar/39Ar and K/Ar) and petrology. Stratigraphic and radiometric data are then used together in the Volcano Event Age Model (VEAM) to identify changes in the rate and type of volcanic eruptions through time, with statistical uncertainty. These newly obtained datasets will be added to published data to build a conceptual model of volcanic hazards at MLV. Alternative conceptual models, for example, may be that the rate of MLV lava flow eruptions are nonstationary in time and/or space and/or volume. We explore the consequences of these alternative models on forecasting future eruptions. As different styles of activity have different impacts, we estimate these potential effects using simulation. The results of this study will improve the existing MLV hazard assessment in hopes of mitigating casualties and social impact should an eruption occur at MLV