How much steam is created when pyroclastic flows enter water? Does it matter?

To understand the physical and thermal processes that occur as pyroclastic flows interact with bodies of water, we performed a series of lab experiments and numerical simulations. The goal of the lab experiments is to quantify the energy balance that occurs as hot particles interact with water. In particular, we are interested in determining how much of the thermal energy of particles is transferred to steam. The volume expansion that accompanies the conversion of water to steam can create a blast when pyroclastic flows reach water and may also play a role in supporting the over-water surge. The goal of the numerical simulations is to examine the consequences of the lab-derived results and to compare model predictions with field observations. In the lab experiments we dropped hot pieces of pumice into cold water and determined how much of the thermal energy is transferred to steam. We considered the effects of particle size, temperature (between 100 and 700 °C), and density (we use glass spheres for comparison with the pumice). We find that the glass spheres generate almost no steam because they sink so quickly that any steam that forms on their surface recondenses in the water. Even though hot pumice becomes saturated with water and sinks very quickly, some of the steam that is generated escapes to the atmosphere. We find that about 10% of the thermal energy is used to generate steam this value is approximately constant for temperatures greater than about 200 °C and independent of particle size. The dynamics and evolution of pyroclastic flows that enter water was studied using a multifluid numerical approach. The multiphase model was adapted from the MFIX code and incorporates granular mechanics, turbulent compressible flow, and permits hydrous phase changes. Assuming that 10% of the flow's thermal energy generates steam, we can reproduce the size and location of the blast that accompanied the July 12, 2003 dome collapse at Montserrat (described in Edmonds and Herd, Geology 2005).