The Role of Tectonic Stress in Triggering Large Silicic Caldera Eruptions

We utilize 3-D temperature-dependent viscoelastic finite element models to investigate the mechanical response of the host rock supporting large caldera-size magma reservoirs (volumes > 10(2) km(3)) to local tectonic stresses. The mechanical stability of the host rock is used to determine the maximum predicted repose intervals and magma flux rates that systems may experience before successive eruption is triggered. Numerical results indicate that regional extension decreases the stability of the roof rock overlying a magma reservoir, thereby promoting early-onset caldera collapse. Alternatively, moderate amounts of compression (<= 10 mm/year) on relatively short timescales (<10(4)years) increases roof rock stability. In addition to quantifying the affect of tectonic stresses on reservoir stability, our models indicate that the process of rejuvenation and mechanical failure is likely to take place over short time periods of hundreds to thousands of years. These findings support the short preeruption melt accumulation timescales indicated by U series disequilibrium studies. Plain Language Summary Catastrophic caldera-forming volcanoes, often referred to as supervolcanoes, are one of the greatest natural hazards on Earth. Understanding how these explosive eruptions are triggered is critical for forecasting volcanic activity at the world's largest volcanoes and mitigating their hazards. Using sophisticated numerical models, the affect of plate tectonic stress is investigated to determine its role in triggering large caldera-forming eruptions. The numerical experiments indicate that calderas located in extensional tectonic settings erupt more readily than calderas located in compressional settings or in tectonically neutral settings. Additionally, calderas in extensional settings fail on shorter timescales than calderas located in compressional or tectonically neutral settings. The most important outcome of this work is that our numerical models show for the first time that the rock surrounding a large magma reservoir is only stable on timescales of centuries to thousands of years when new magma is actively being injected into the magma reservoir. This finding provides important constraints on the amount of time necessary to recharge and erupt a large, supervolcano size reservoir from the first indication of magmatic activity.