A statistical approach to evaluate the tephra deposit and ash concentration from PUFF model forecasts

In this paper we present a new statistical approach which provides tephra deposit load and ash concentration
using PUFF, a Lagrangian model widely used to forecast volcanic ash dispersal during volcanic crisis. We
perform a parametric study in order to analyze the influence of each input parameter on model outputs. For
this test, we simulate two eruptive scenarios similar to the 2001 (Scenario 1) and 1998 (Scenario 2) Etna
eruptions using high resolution weather data and a domain of 170×170 km. Results show that for both
scenarios, we are able to calculate the tephra deposit load and ash concentration but the use of millions of
particles is required. Specifically, up to 33 and 220 millions of particles were necessary to accurately predict
the tephra deposit and ash concentration in air, respectively. This is approximately two orders of magnitude
larger than the values typically considered running PUFF. The parametric study shows that the horizontal
diffusion coefficient, the time step of the simulations, the topography and the standard deviation of the
particle distribution greatly affect the model outputs. We also validate the model by best-fit procedures.
Results show a good comparison between field data of the 2001 Etna eruption and PUFF simulations, being
inside 5 and 1/5 times the observed data, comparable with results of Eulerian models.
This work will allow to reliably outline the areas of contaminated airspace using PUFF or any other Lagrangian
model in order to define the No Fly Zone and ensure the safety aviation operations as required after the
Eyjafjallajökull eruption