Time-resolved seismic tomography detects magma intrusions at Mt. Etna

The continuous volcanic and seismic activity at Mount Etna makes this volcano an important
laboratory for seismological and geophysical studies. We used repeated three-dimensional
tomography to detect variations in elastic parameters during different volcanic cycles, before and
during the October 2002–January 2003 flank eruption. Well-defined anomalous low P- to S-wave
velocity ratio volumes were revealed. Absent during the pre-eruptive period, the anomalies trace
the intrusion of volatile-rich (Q4 weight percent) basaltic magma, most of which rose up only a few
months before the onset of eruption. The observed time changes of velocity anomalies suggest that
four-dimensional tomography provides a basis for more efficient volcano monitoring and shortand
midterm eruption forecasting of explosive activity.

1. B. R. Julian, A. M. Pitt, G. R. Foulger, Geophys. J. Int.
133, F7 (1998).
2. G. R. Foulger et al., J. Geophys. Res. 108, 10.1029/
2000JB000041 (2003).
3. G. Guerin, W. He, R. N. Anderson, L. Xu, U. T. Mello, in
2000 Offshore Technology Conference, paper 12101,
www.ces-enterprise.com/vpatch/bp/attachments/
OTC2000_Guerin.pdf.
4. R. O’Connell, B. Budiansky, J. Geophys. Res. 79, 5412
(1974).
5. N. I. Christensen, W. W. Wepfer, Geol. Soc. Am. Mem.
172, 91 (1989).
6. Z. W. Wang, A. Nur, Soc. Pet. Eng. Res. Eng. 3, 429
(1989).
7. J. Dvorkin, G. Mavko, A. Nur, Geophys. Res. Lett. 26,
3417 (1999).
8. G. Mavko, J. Geophys. Res. 85, 5173 (1980).
9. H. Sato, I. S. Sacks, T. Murase, J. Geophys. Res. 94, 5689
(1989).
10. C. O. Sanders, S. C. Ponko, L. D. Nixon, E. A. Schwartz,
J. Geophys. Res. 100, 8311 (1995).
11. M. Coltelli, P. Del Carlo, L. Vezzoli, Int. J. Earth Sci. 89,
665 (2000).
12. R. Clocchiatti, M. Condomines, N. Guenot, J. C. Tanguy,
Earth Planet. Sci. Lett. 226, 397 (2004).
13. N. Spilliaert, P. Allard, N. Me´ trich, A. V. Sobolev,
J. Geophys. Res. 111, 10.1029/2005JB003934 (2006).
14. C. H. Thurber, in Seismic Tomography: Theory and
Practice, H. M. Iyer, K. Hirahara, Eds. (Chapman and Hall,
London, 1993), pp. 563–583.
15. D. Patane` , C. Chiarabba, P. De Gori, A. Bonaccorso,
Science 299, 2061 (2003).
16. M. Laigle, A. Hirn, M. Sapin, J. C. Lepine, J. Geophys. Res.
105, 21,633 (2000).
17. C. Chiarabba, P. De Gori, D. Patane` , in Mt. Etna: Volcano
Laboratory, A. Bonaccorso, S. Calvari, M. Coltelli,
C. Del Negro, S. Falsaperla, Eds. (Geophysical Monograph
Series, American Geophysical Union, Washington, DC,
2004), pp. 191–204.
18. For igneous rocks, laboratory experiments found that the
VP /VS ratio generally decreases as temperature increases
and increases as pressure increases. An increase in the
VP /VS ratio is related to increases in temperature,
fracture, and especially partial melt, whereas a decrease
in the ratio can be associated with the presence of gas or
supercritical fluids [see (10) and references therein].
19. M. Aloisi, A. Bonaccorso, S. Gambino, M. Mattia,
G. Puglisi, Geophys. Res. Lett. 30, 10.1029/
2003GL018896 (2003).
20. D. Patane` , M. Mattia, M. Aloisi, Geophys. Res. Lett. 32,
10.1029/2004GL021773 (2005).
21. S. Gambino, A. Mostaccio, D. Patane` , L. Scarfı`, A. Ursino,
Geophys. Res. Lett. 31, 10.1029/2004GL020499 (2004).
22. D. Andronico et al., Bull. Volcanol. 67, 10.1007/s00445-
004-0372-8 (2005).
23. C. Martı´nez-Arevalo, D. Patane` , A. Rietbrock, J. Ibanez,
Geophys. Res. Lett. 32, 10.1029/2005GL023736 (2005).
24. Theory predicts that the velocity decreases and VP /VS
increases for increasing crack density (4). High VP /VS
values affecting portions of fault zones, such as at the San
Andreas Fault, are interpreted as evidence for the
presence of overpressured fluids (25). The influence of
pore pressure on seismic velocities is caused by the
tendency of cracks to remain open when they are
internally pressurized (26). Thus, fracturing and
increasing pore pressure cause a relatively marked
decrease of VS but only a small decrease of VP, which
yields an increased VP/VS ratio. The concentration of
fluids in a dike intrusion mainly occurs near its tip.
Moreover, a rapid migration of fluids in the surrounding
rock volume can also occur, especially when prefractured
zones exist, leading to a high VP/VS ratio and to an
increase of the seismicity (Figs. 2 and 4C).
25. C. H. Thurber et al., Geophys. Res. Lett. 24, 1591 (1997).
26. A. Nur, Bull. Seismol. Soc. Am. 62, 1217 (1972).
27. A. Michelini, T. V. McEvilly, Bull. Seismol. Soc. Am. 81,
524 (1991).
28. M. Reyners, D. Eberhart-Phillips, G. Stuart, Geophys. J.
Int. 137, 873 (1999).
29. We thank T. Caltabiano for SO2 flux data and anonymous
referees for helpful comments and suggestions on the
manuscript. This work was supported by grants from the
European Union VOLUME FP6-2004-Global-3 and INGV–
Department of Civil Protection V3/6 projects.