Modeling the dynamics and deposits characteristics of dilute pyroclastic density currents

Dilute pyroclastic density currents (PDCs) are one of the most dangerous and destructive types of flows generated by explosive volcanic eruptions. Such flows are typically unsteady, density-stratified, and turbulent, and tend to produce a wide range of deposit characteristics, including dune-like bedforms at a range of wavelengths. Controls on the dynamics and sediment transport and deposition of dilute PDCs remain poorly understood. In order to better understand dynamic controls and to provide a framework for interpreting deposit characteristics in terms of current dynamics, we developed an axi-symmetric model for flow of and sedimentation from a steady-state, vertically uniform density current. Equations for the conservation of mass, momentum, and energy are solved simultaneously, and the effects of atmospheric entrainment, particle sedimentation, basal friction, temperature changes, and variations in current thickness and density are explored. Additionally, we incorporate the Rouse number and Brunt-Väisäla frequency to estimate the wavelength of internal gravity waves in a density-stratified current. The model predicts realistic runout distances and bedform wavelengths for several well-documented field cases, although specific results are heavily dependent upon source conditions, grain-size characteristics, and entrainment and friction parameters. For example, increasing particle settling velocity, by increasing particle size and/or decreasing total particle concentration, decreases both predicted runout distance and bedform wavelength. Including the effects of enhanced entrainment of ambient air, due to increased velocity or slope, reduces runout distance and increases calculated bedform wavelength. Although preliminary runs for well-constrained test cases are consistent with observations, a number of processes and complexities must be considered as this work moves forward. In particular, because our existing formulation is based on depth-averaged equations, assumptions about density stratification and corresponding Brunt-Väisäla frequencies are likely an oversimplification. Furthermore, and perhaps more importantly, relationships between bedform wavelengths and internal gravity wave characteristics are not well-established at this time. In response to these difficulties, we propose a two-pronged plan that includes significant model development to quantify controls on density profile, along with experimental observation in a controlled laboratory setting to document the relationships among large-scale flow dynamics, density stratification and deposit characteristics.