Mapping block-and-ash flow hazards based on Titan 2D simulations: a case study from Mt. Taranaki, NZ

摘要

Numerical models for simulation of mass flows are typically focussed upon accurately predicting the paths, travel times and inundation from a single flow or collapse event. When considering catchment-based hazards from a volcano, this is complicated by often being faced with several possible scenarios. Over the last 800 years at Mt. Taranaki/Egmont, a number of dome growth and collapse events have resulted in the genesis and emplacement of block-and-ash flows (BAFs). Each BAF was directed northwestward by a breach in the crater rim. The latest dome collapse events in the AD 1880s and AD 1755 inundated the northwestern flank and had run-out lengths 10 km from source. Future activity of this type could have a devastating effect on the Taranaki region’s communities, infrastructure and economy. Hazard planning has involved constructing volcanic hazard maps based upon the areas inundated by past volcanic flows, with little consideration of present-day topography. Here, a numerical geophysical mass flow modelling approach is used to forecast the hazards of future comparable BAF events on NW Mt. Taranaki. The Titan2D programme encompasses a “shallow water”, continuum solution-based, granular flow model. Flow mechanical properties needed for this approach include estimates of internal and basal friction as well as the physical dimensions of the initial collapse. Before this model can be applied to Taranaki BAFs, the input parameters must be calibrated by simulating a range of past collapse events. By using AD 1860 and AD 1755 scenarios, initial collapse volumes can be well constrained and internal and basal friction angles can be evaluated through an iterative approach from previous run-out lengths. A range of possible input parameters was, therefore, determined to produce a suite of potentially inundated areas under present-day terrain. A suite of 10 forecasts from a uniformly distributed range were combined to create a map of relative probabilities of inundation by future BAF events. These results were combined in a GIS package to produce hazard zones related to user-specified hazard thresholds. Using these input parameter constraints, future hazard forecasts for this scale and type of event can also take into account changing summit and topographic configurations following future eruptive or collapse events.