Mapping and DOWNFLOW simulation of recent lava flow fields at Mount Etna Simone Tarquini ⁎, Massimiliano Favalli Istituto Nazionale di Geofisica e Vulcanologia, via della Faggiola 32, 56126 Pisa, Italy abstract article info Article history: Received 28 January 2011 Accepted 1 May 2011 Available online 9 May 2011 Keywords: LIDAR lava flow field lava flow simulation Digital elevation model Mount Etna In recent years, progress in geographic information systems (GIS) and remote sensing techniques have allowed the mapping and studying of lava flows in unprecedented detail. A composite GIS technique is introduced to obtain high resolution boundaries of lava flow fields. This technique is mainly based on the processing of LIDAR-derived maps and digital elevation models (DEMs). The probabilistic code DOWNFLOW is then used to simulate eight large flow fields formed at Mount Etna in the last 25 years. Thanks to the collection of 6 DEMs representing Mount Etna at different times from 1986 to 2007, simulated outputs are obtained by running the DOWNFLOW code over pre-emplacement topographies. Simulation outputs are compared with the boundaries of the actual flow fields obtained here or derived from the existing literature. Although the selected fields formed in accordance with different emplacement mechanisms, flowed on different zones of the volcano over different topographies and were fed by different lava supplies of different durations, DOWNFLOW yields results close to the actual flow fields in all the cases considered. This outcome is noteworthy because DOWNFLOW has been applied by adopting a default calibration, without any specific tuning for the new cases considered here. This extensive testing proves that, if the pre-emplacement topography is available, DOWNFLOW yields a realistic simulation of a future lava flow based solely on a knowledge of the vent position. In comparison with deterministic codes, which require accurate knowledge of a large number of input parameters, DOWNFLOW turns out to be simple, fast and undemanding, proving to be ideal for systematic hazard and risk analyses. © 2011 Elsevier B.V. All rights reserved. 1. Introduction Mt. Etna is among the most active basaltic volcanoes of the world, and its effusive activity poses an incessant threat to human communities living on its flanks. This volcano is also one of the most studied in the world, and constitutes a natural laboratory where new techniques and tools contributing to advancements in volcanol- ogy have often been experimented (e.g. Wright et al., 2008a; Favalli et al., 2010). Lava flow simulation is a front-edge technique which is increas- ingly used to mitigate the hazard posed by lava streams flowing downhill in populated areas (Crisci et al., 2003, 2010; Favalli et al., 2009a; Vicari et al., 2009; Favalli et al., Submitted). Mt. Etna has been one of the first natural scenarios where lava flow simulation codes have been tested and applied (Young and Wadge, 1990, Wadge et al., 1994). Since the early 1990s, at least 9 different lava flow simulation codes have been tested or extensively applied on this volcano: Dobran and Macedonio (Dobran and Macedonio, 1992), FLOWFRONT (Wadge et al., 1994), SCIARA (e.g. Crisci et al., 1999), LavaSIM (Hidaka et al., 2005; Proietti et al., 2009), DOWNFLOW (e.g. Favalli et al., 2005, 2009b), Costa and Macedonio (Costa and Macedonio, 2005a), ELFM (Damiani et al., 2006), FLOWGO (Harris and Rowland, 2001; Wright et al., 2008b), MAGFLOW (e.g. Vicari et al., 2007, 2009). Existing lava flow simulation codes can be roughly classified in two broad categories: deterministic codes based on transport theory (e.g. SCIARA, Crisci et al., 2010), and probabilistic codes based on the maximum slope (e.g. DOWNFLOW, Favalli et al., 2005; see also Costa and Macedonio, 2005b). While deterministic codes are based on the solution of the physical governing equation of the flowing lava, and hence need the knowledge of a large number of input parameters (e.g. lava discharge rate, temperature, viscosity, etc.), probabilistic codes need only the calibration of a few parameters and they promptly yield results which are often representative of a wide spectrum of possible events. The validation of the result of a lava flow simulation is obtained by comparing the area covered by the simulation output with the area covered by the corresponding actual lava flow (e.g. Vicari et al., 2007). An accurate mapping of real lava flows is therefore an essential element to assess the performances of a simulation code. Recently, LIDAR data processing proved to be a powerful technique for lava flow mapping (Mazzarini et al., 2007) and for the creation of high resolution digital elevation models (DEMs) of volcanic areas (Favalli et al., 2009c, 2010; Fornaciai et al., 2010). In this work, we apply, at Mt Etna, a combination of different techniques to obtain enhanced maps of several recent lava flows based on LIDAR data processing (Mazzarini et al., 2007; Favalli et al., Journal of Volcanology and Geothermal Research 204 (2011) 27–39 ⁎ Corresponding author. Tel.: + 39 050 8311932. E-mail addresses: tarquini@pi.ingv.it (S. Tarquini), favalli@pi.ingv.it (M. Favalli). 0377-0273/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.jvolgeores.2011.05.001 Contents lists available at ScienceDirect Journal of Volcanology and Geothermal Research journal homepage: www.elsevier.com/locate/jvolgeores