The eruption of silicic lava domes on the Pacific rim over the last decade has focused interest and attention on the potential hazards (to life and property) associated with these volcanoes. However, the modelling of the emplacement of these flows has been hampered by the lack of accurate rheological data. As a first approximation, some physical properties can be deduced from field textural measurements and description. Recently, viscosity measurements have been obtained from natural rhyolitic obsidians. This study is a first attempt to combine field measurements, accurately determined laboratory rheological measurements, and numerical modelling results for specific lava flows.
Viscosity measurements (using the parallel-plate method) have been performed on samples from 4 different textural units within the Ben Lomond lava dome. Viscosity results are being used in tandem with (as a first approximation) an existing numerical cooling model to model temperature-time-depth profiles and viscosity-time-depth profiles within this lava flow. Viscosity measurements from specific textural units within lava flows, together with a numerical model, can be used to constrain the timing of formation of different textural units. As a first approximation, the timescale for bubbles to form in the finely vesicular pumice carapace may be estimated as the difference between the start time (t = 0) at the emplacement temperature, and the time elapsed when the magma crosses the rheological glass transition (Tg) as it cools. For the 90 m thick Ben Lomond flow, vesiculation occurs within the upper 10 m of the flow up to 2 years after emergence of the lava from the vent. In contrast, there is a lag-time of > 15 years for spherulites to form in the crystalline flow centre, as spherulites form just above and just below Tg. With this approach, the timescale for the formation of explosion pits and explosion breccias - potential hazards for these types of volcanoes - can be constrained. In addition, supportive experimentally-determined cooling rates through Tg using calorimetry, and water speciation, will aid in checking the reliability of the thermal models.