Local seismic and infrasound observations of the 2009 explosive eruptions of Redoubt Volcano, Alaska S.R. McNutt ⁎, G. Thompson, M.E. West, D. Fee, S. Stihler, E. Clark University of Alaska Fairbanks Geophysical Institute, United States abstract article info Article history: Received 12 June 2011 Accepted 20 March 2013 Available online xxxx Keywords: Explosive eruptions Infrasound Seismic event Reduced displacement Local seismic and infrasound data are used to characterize the March–April 2009 eruptions of Redoubt volca- no, Alaska. Data are from a network of ten seismic stations ranging from 2.6 to 21 km from the vent and a single infrasound sensor 12.2 km from the vent. Measurements were made for more than 30 explosive erup- tions including the 19 numbered events that were identified during the eruption response. More than 30 smaller explosions were also identified. Measured parameters include onset time, duration, reduced dis- placement (D R ), pressure, seismic energy, and acoustic energy. Results were compared with complementary gas and lightning data. Based on ratios of pressure to D R and acoustic to seismic energy we find that the initial group of explosive eruptions on 23–24 March was relatively stronger seismically and included the four highest D R values. We infer that these represent vent enlargement or stronger coupling between the magma and wall rocks. One event, at 12:31 UT on 23 March had an extremely high pressure of 250 Pa for the initial pulse and was likely weakly seismically coupled. A second group of explosive eruptions on 27– 29 March was relatively stronger acoustically. Deposits were finer grained and of different lithology, with less scoria and more dense clasts. These two groups may represent different modes of gas release. Cumulative acoustic energy as a function of time correlates well with comparable lightning and SO 2 data, suggesting that all three provide insight regarding different factors of gas release for the eruption as a whole. The number of events identified depends on criteria measured, and depends on the goals of the investigations. Overall the eruptions were substantially stronger than recent eruptions in Alaska, with some durations exceeding 1 h and pressures an order of magnitude higher. © 2013 Elsevier B.V. All rights reserved. 1. Introduction Seismic monitoring of volcanoes has long been an effective way to determine characteristics of eruptions. For example, the amplitude of volcanic tremor during eruptions, corrected for distance and instru- mental effects (called reduced displacement, D R ), has been shown to be proportional to the Volcanic Explosivity Index or VEI (Newhall and Self, 1982; McNutt, 1994). If the amplitudes are similar, the dura- tions of eruptions, as measured on seismograms, are approximately proportional to tephra volumes (McNutt and Nishimura, 2008). An index developed for Sheveluch Volcano, Kamchatka uses a combination of amplitude and duration to estimate eruption sizes (S. Senyukov, writt. comm. 2004). In recent years infrasound sensors (low frequency sound,b 20 Hz) have proved to be a useful complement to seismic data (Fee and Matoza, 2013; Johnson and Ripepe, 2011). Local infrasound sensors (within 15 km) have the advantage that the medium (the air) does not significantly distort the signal so that direct estimates of the source pressure time history are possible. This is in contrast to seismic data where multi-pathing, reflections and refractions may modify the seismograms, even over short distances. Seismic and infrasound data together are an effective combination, especially because they allow discrimination between air and ground wave energy which may vary with factors such as source depth, size, shape and bubble distribution (Garces et al., 1998; Johnson and Aster, 2005). Here we review seismic and infrasound data recorded during the March–April 2009 eruptions of Redoubt Volcano (60.4857 N, 152.742 W, 3100 m) south central Alaska (Fig. 1). The eruptions consisted of nineteen numbered events (Schaefer et al., 2012) and more than 20 other events that were documented during contempo- rary and later analyses (Fig. 2). Each event produced seismic and acoustic signals, and many of them also produced lightning (Behnke et al., this issue). The seismic data were recorded on stations from 2.2 to >20 km from the vent (Fig. 1). The close short-period stations were able to observe small signals that were too weak to register on distant stations, but the close stations became saturated (clipped) for larger signals which remained on scale on the more distant sta- tions. Thus using different stations provided some data for all events. The infrasound data were recorded on a single local sensor 12.2 km NNE of the vent (station DFR, Fig. 1). At this distance some small acoustic signals were not detectable above the noise level. All nine- teen numbered events showed clearly. Also, all nineteen numbered Journal of Volcanology and Geothermal Research xxx (2013) xxx–xxx ⁎ Corresponding author at: University of South Florida. E-mail address: smcnutt@usf.edu (S.R. McNutt). VOLGEO-05110; No of Pages 14 0377-0273/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jvolgeores.2013.03.016 Contents lists available at SciVerse ScienceDirect Journal of Volcanology and Geothermal Research journal homepage: www.elsevier.com/locate/jvolgeores Please cite this article as: McNutt, S.R., et al., Local seismic and infrasound observations of the 2009 explosive eruptions of Redoubt Volcano, Alaska, Journal of Volcanology and Geothermal Research (2013), http://dx.doi.org/10.1016/j.jvolgeores.2013.03.016