Magma supply path beneath Mt. Asama volcano, Japan

Minoru Takeo,1 Yosuke Aoki,1 Takao Ohminato,1 and Maki Yamamoto1,2

Received 20 March 2006; revised 9 May 2006; accepted 17 May 2006; published 8 August 2006.

[1] Obtaining a sharp image of magma supply paththrough dense geophysical observations is important forforecasting time and magnitude of hazardous futureeruptions. Here we reveal a clear magma plumbingsystem using dense seismic and geodetic networks aroundMt. Asama, central Japan. Magma intrusions occurredseveral times beneath the western flank of Mt. Asama,forming a WNW-ESE directed zone with 1 km below sealevel. The eastern end of this zone connects a narrowvertical pathway extending right under the summit crater,which erupted in 2004. Monitoring magmatic activity with awell-designed observational network is vital to mitigatefuture volcano hazards. Citation: Takeo, M., Y. Aoki,

T. Ohminato, and M. Yamamoto (2006), Magma supply path

beneath Mt. Asama volcano, Japan, Geophys. Res. Lett., 33,

L15310, doi:10.1029/2006GL026247.

1. Introduction

[2] Understanding a magma feeding system of a volcanois fundamental to assess a potential volcanic hazard. Thisgoal can be achieved by monitoring magma propagationwith well-designed networks of geophysical instruments.Some volcanoes generate various geophysical signals be-fore eruption. For example, long-period seismic signals areoften observed at shallow depths [e.g., Chouet, 2003], andearthquakes with higher frequency contents are also ob-served in volcanic areas [e.g., Rubin et al., 1998]. Also, avolcano deforms before an eruption due to magma injection[e.g., Aoki et al., 1999]. On the other hand, magmaplumbing systems beneath some volcanoes generate onlyweak signals which can be hidden by the signals associatedwith the migration of fluid in the shallow hydrothermalsystem [e.g., Battaglia et al., 2006]. In these cases, it isdifficult to identify a magma pathway using seismic andgeodetic measurements. There have been a few studies totry to delineate the magma system from geophysical obser-vations [e.g., Uhira et al., 2005]. Here we report a sharpimage of magma supply path beneath Mt. Asama volcano,central Japan, by integrating precisely relocated hypocentersand ground deformation data obtained from dense geophys-ical networks. In a way, we were lucky to be able to providethe sharp picture due to the active seismicity and theobvious crustal deformation.[3] Mt. Asama, which is one of the most active volcanoes

in Japan, is an andesitic volcano located in the center of thecountry. The summit elevation is 2560 m above sea level,and the size of the active summit crater is 450 m in diameter

and 150m in depth. To the west of Mt. Asama, there is a rowof older Quaternary volcanoes collectively known as EboshiVolcanoes. The volcanism near Mt. Asama appears to haveprogressed eastward, with Asama volcano as its eastern endand the youngest member of the row [Aramaki, 1963].[4] At 11:02 (GMT) on 1 September 2004, a moderate-

sized eruption occurred for the first time in the last 21 years.From 14–18 September, a continuous stromblian explosionemitted volcanic ash that reached as far as the Tokyometropolitan area about 130 km away. The volcanic activityseemed to have subsided thereafter, except for moderate-sized eruptions occurring on 23 September, 29 September,and 10 October, and some small-scale eruptions afterwardwhich became smaller and smaller with time. The lastmoderate-sized eruption occurred on 14 November, andsince then no eruption has occurred up to the end of 2005.

2. Observations

[5] The Earthquake Research Institute, University ofTokyo (ERI) and the Japan Meteorological Agency (JMA)operate seismic networks around Mt. Asama. These twoinstitutions exchange data for the purposes of scientificresearch and volcanic activity monitoring. Figure 1 showsthe spatial distribution of the seismic networks. Two summitstations (KAH2 and KAC2) were broken by the firsteruption. The stations coded with four characters (e.g.,ASMA), except KAC2 and KAH2, are operated by JMA.

3. Hypocenter Relocations

[6] Hypocenters of more than five-hundred volcanicearthquakes occurring from 1 January 2004 to 19 October2005, were determined in the routine data processing by theAsama volcano observatory (AVO), ERI. The routine hypo-centers (see auxiliary material1) spread vaguely beneath Mt.Asama, and it is difficult to infer a magma supply path fromthe routine result. We used the double-difference algorithm(DD method) and hypoDD (software) [Waldhauser andEllsworth, 2000] to obtain the precise relative distributionof these hypocenters. To regularize the ill-conditionedsystems of inversion problem, the DD method selects eventsthat are well linked to other events. Therefore, several dozenearthquakes that occurred beneath the eastern flank of Mt.Asama were excluded from the relocated list because theseevents were isolated from each other. The DD method takesa lower limit of travel-time assortments, so it essentiallyexcludes events with poorly picked data. In this paper, thelower limit of assortments was taken to be eight. Nearly halfof the events were relocated, and the main characteristic ofthe routine hypocenter distribution was included in the

1Auxiliary materials are available in the HTML. doi:10.1029/2006GL026247.

GEOPHYSICAL RESEARCH LETTERS, VOL. 33, L15310, doi:10.1029/2006GL026247, 2006ClickHere

for

FullArticle

1Earthquake Research Institute, University of Tokyo, Tokyo, Japan.2Now at NS Solutions Corporation, Tokyo, Japan.

Copyright 2006 by the American Geophysical Union.0094-8276/06/2006GL026247$05.00

L15310 1 of 5

relocated one. Using the DD method, Yamamoto et al.[2005] relocated the hypocenters of the earthquake swarmthat started at 6:10 (GMT) on 31 August, lasting until justbefore the first eruption. The stars and solid circles inFigure 2 represent the relocated hypocenters and the relo-cated swarm, respectively; those are independently deter-mined and plotted on the same figure. The relocateddistribution reveals a sharp image of seismicity composedof two groups. One group (Group-I) forms a WNW-ESEdirected zone at a depth range between 1 km and 1.5 kmbelow sea level. The eastern end of this seismic zone liesbeneath the summit crater and extends westward horizon-tally over 2 km in length. The other group (Group-II) formsa narrow vertical seismic zone extending from the easternedge of Group-I to just under the summit crater. The data ofthe two summit stations, KAC2 and KAH2, were notavailable after the first eruption. To examine whether themissing summit stations affect the hypocenter distribution,we relocated all these events under the condition that thesummit data were excluded from the catalog of travel-timedata, and confirmed that the effect of the missing stationswas negligibly small. Note that the hypocenters duringSeptember 2004 are missing from the relocated resultsbecause an extremely high and continuous seismicity asso-ciated with the summit eruptions prevented us from obtain-ing precise hypocenters.[7] The hypocenter distribution dramatically changed

before and after the eruption on 1 September 2004(Figure 3). Before the first eruption, almost all eventsoccurred just beneath the summit crater with a depthshallower than 1 km above sea level, and the activity inthe deeper part of Mt. Asama was relatively quiet. Thehypocenter distribution during this period corresponds tothe top 1km portion of Group-II. The hypocenter distribu-tion after the eruption was composed of two groups: Group-

I, and the deeper portion of Group-II. Group-I has beenactivated since the end of October 2004 and almost all ofthe events in this group are classified into an A-typeearthquake: the nature of seismograms is similar to thoseof the shallow tectonic earthquakes. The most of the eventsin Group-II are B-type earthquakes. Although they arecharacterized by emergent onsets, we could pick P-waveonsets at stations near the summit.

4. A Dike Model

[8] In addition to the seismic data, we also used contin-uous GPS data. We modelled the ground deformation fieldbetween June 2004 to March 2005 by inverting for length,width, depth, dip angle, strike direction, location, andamount of opening of an intruded rectangular dike in anelastic, homogeneous, and isotropic medium [Okada, 1985;Cervelli et al., 2001]. A pressure source model and a faultdislocation model cannot explain the pattern of the grounddeformation. The results show that the observations arewell explained by a dike intrusion to the western flank ofMt. Asama. The intruded volume is 6.8 � 106 m3, which isabout three times larger than the 2 � 106 m3 of magmaemitted during the eruption [Nakada et al., 2005]. Eventhough the total deformations are not large enough to wellconstrain every parameter, the horizontal location of thedike and the total volume of the intruded dike are con-strained relatively well; the estimated standard deviationsare ±0.7 km, and ±1.5 � 106 m3, respectively. The easternpart of the dike overlaps with Group-I of the relocatedhypocenter distribution (Figure 4). The depth at the top ofthe dike, whose estimated standard deviation is ±1.3 km,

Figure 1. Seismic network around Mt. Asama. The insetat the upper-right corner shows the location of Mt. Asama.

Figure 2. Locations for earthquakes which occurred from1 January 2004 to 19 October 2005. Stars and solid circlesrepresent the relocated hypocenters of the normal volcanicearthquakes and the swarm occurring just before the firsteruption, respectively. Crosses indicate the stations.

L15310 TAKEO ET AL.: MAGMA SUPPLY PATH BENEATH MT. ASAMA L15310

2 of 5

coincides with the depth range of Group-I. The distributionof dike-induced seismicity reflects the distribution of am-bient stresses that are near to failure, thus the seismicitymight be much more limited in extent than the dike thatproduced it [Rubin et al., 1998; Rubin and Gillard, 1998],being concentrated near the dike perimeter. The eastern endof Group-I is connected with the narrow vertical seismiczone, Group-II, extending from 1 km below sea level to justunder the eruptive summit crater. The swarm activityimplies the opening process of the blocked shallowest partof the vent. The relocated hypocenter distribution, in which

one part overlaps with the dike and the other part extendsvertically to the summit crater, represents the magma supplypath beneath Mt. Asama.

5. Discussion and Conclusions

[9] Figure 5 compares temporal variations in the monthlynumber of A-type earthquakes and that of all volcanicearthquakes with the changes in GPS baseline lengthbetween 950221 and 950268 from 1996 to 2004. BecauseEboshi-Asama volcanic row lies between these two GPS

Figure 3. Time series of relocated earthquakes. The time series is split before and after the first eruption, with theepicenter distributions and the north-south cross sections. (right) The red arrows indicate the timing of the moderate-sizederuptions.

Figure 4. (left) A dike model, which explains the total crustal deformation from June 2004 to March 2005, is shown by ared rectangle. A red shaded zone is the schematic magma supply path beneath Mt. Asama. (right) Parameters of the dikemodel and the comparison between the observed and calculated deformations are shown. Dashed ellipses represent theobservational error. A green triangle indicates the summit of Mt. Asama.

L15310 TAKEO ET AL.: MAGMA SUPPLY PATH BENEATH MT. ASAMA L15310

3 of 5

stations, the baseline extensions between these two GPSsites clearly indicate magma intrusions beneath the westernflank of Mt. Asama. The hypocenter distribution before2004 determined in the routine data processing by AVOrepresents that the A-type earthquakes occurred under thewestern side of Mt. Asama and the other volcanic earth-quakes occurred beneath the summit of Mt. Asama,although their precision was not as good as that after2004 due to lack of dense seismic network. The estimateddike location before 2004 was also beneath Eboshi-Asamavolcanic raw [Murakami, 2005]. It seems reliable that theoverall trend of hypocentral distribution and the dikelocation had not changed during the last decade. Beforethe eruption, we recognized three stages of increase innumber of A-type earthquakes: the latter half of 1996, fromOctober 2000 to April 2001, and from May 2002 to August2002. All the three GPS baseline extensions from 1996 to2003 were synchronized with these seismic activations,suggesting that the A-type earthquakes were associatedwith the intrusion of magma beneath the western flank ofMt. Asama.[10] On the other hand, the GPS baseline contractions

were also observed three times: from September 1997 toMarch 2000, from July 2001 to February 2002, and fromMarch 2003 to April 2004. The baseline contractionsindicate migrations of intrusive magma from under thewestern flank of Mt. Asama to other places. Although theexact direction of the magma migration is unknown becausethe change in GPS baseline length between 950221 and950264 is only sensitive to the inflation and deflation underthe western flank of Mt. Asama, it would be one of thepossibilities of two directions, the direction from whichmagma is injected into the dike, and the direction to thevent. The activity of the volcanic earthquakes was lowduring the first contraction period except for the last severalmonths of this period. On the contrary, the activity of B-type

earthquakes kept up in high level during the latter twocontracting periods. The maximum temperature of thecrater had exceeded 200�C from fall in 2002 [JapanMeteorological Agency (JMA), 2005], representing thatthe shallow part of vent had kept on high-temperature statefrom the middle of 2002. These observations suggest thatthere were certain essential differences between the firstcontraction stage and the last two contraction stages.[11] A sudden extension of GPS baseline length between

950221 and ASM4 was detected between 21–22 July 2004,suggesting a nearly vertical magma intrusion into the samedike beneath the western flank of Mt. Asama [Murakami,2005; Aoki et al., 2005]. Volcanic glows had begun to beobserved since the last ten days of July 2004 and themaximum temperature at the bottom of the crater exceeded500�C after the sudden extension of the baseline length[JMA, 2005]. These surface phenomena indicated that thetemperature rise in the shallow part of the vent succeededthe magma intrusion.[12] Integrating the observational facts as discussed

above, the two baseline contractions after 2000 appear tobe caused by the migration of the intrusive magma from thedike beneath the western flank to the vent of Mt. Asama.The gradual supply of magma into the vent had induced agradual activation of volcanic earthquakes from the middleof 2001 to the first eruption on 1 September 2004. Therelocated hypocenter distribution in Figure 3 shows thatalmost all earthquakes, which occurred from January 2004 tothe first eruption, lie in the shallower part of the vent with adepth shallower than 1 km above sea level; during this period,the seismicity was relatively high from May to June 2004.These facts suggest that magma had ascended about 1 kmabove sea level gradually by June 2004 at least and theascending magma may have increased the internal pressurein the top portion of the vent causing an enhanced seismicactivity at the shallower part of the vent. Before the eruptionon September 1, magma migration from the deep chamber tothe shallower portion of the vent was relatively slow probablydue to the sealing by a cap at the top of the vent. After theeruption, the removal of the cap may have made the magmamigrationmuch easier than before and caused the high A-typeearthquake activity. This speculation is supported by a factthat partially molten country rocks (rhyolite tuff) are foundamong the 2004 eruption products [Nakada et al., 2005].[13] The dense seismic and geodetic networks in and

around Mt. Asama have provided us with high quality data.Employing these data, we were able to make clearthe magma supply path beneath Mt. Asama shallower than2 km below sea level based on the precise distribution of thehypocenters and the crustal deformation before and after theeruption on 1 September 2004.

[14] Acknowledgment. The Japan Meteorological Agency kindlyprovided us with the seismic data.

ReferencesAoki, Y., P. Segall, T. Kato, P. Cervelli, and S. Shimada (1999), Imagingmagma transport during the 1997 seismic swarm off the Izu Peninsula,Japan, Science, 286, 927–930.

Aoki, Y., H. Watanabe, E. Koyama, J. Oikawa, and Y. Morita (2005),Ground deformation associated with the 2004–2005 unrest of AsamaVolcano, Japan, Bull. Volcanol. Soc. Jpn., 50, 575–584.

Aramaki, S. (1963), Geology of Asama Volcano, J. Fac. Sci. Univ. Tokyo,14, 229–443.

Figure 5. (top) Temporal variation in monthly number ofall volcanic earthquakes and (middle) that of A-typeearthquakes, which are counted using the monitoring recordat SAN, are compared with (bottom) changes in GPSbaseline length between 950221 and 950268 from 1996 to2004. The extension and contraction periods are shaded bydark gray and bright gray, respectively.

L15310 TAKEO ET AL.: MAGMA SUPPLY PATH BENEATH MT. ASAMA L15310

4 of 5

Battaglia, M., C. Troise, F. Obrizzo, F. Pingue, and G. De Natale (2006),Evidence for fluid migration as the source of deformation at CampiFlegrei caldera (Italy), Geophys. Res. Lett., 33, L01307, doi:10.1029/2005GL024904.

Cervelli, P., M. H. Murray, P. Segall, Y. Aoki, and T. Kato (2001), Estimat-ing source parameters from deformation data, with an application to theMarch 1997 earthquake swarm off the Izu Peninsula, Japan, J. Geophys.Res., 106, 11,217–11,237.

Chouet, B. (2003), Volcano Seismology, Pure Appl. Geophys., 160, 739–788.

Japan Meteorological Agency (JMA) (2005), Abstract of volcanic activitiesof Asamayama in 2004, Rep. of Coord. Comm. for Predict. of VolcanicEruption 89, pp. 11–23, Tokyo.

Murakami, M. (2005), Magma plumbing system of the Asama volcanoinferred from continuous measurements of GPS, migration preceding2004 eruption of Asama Volcano detected by GPS, Bull. Volcanol. Soc.Jpn., 50, 347–361.

Nakada, S., M. Yoshimoto, E. Koyama, T. Tsuji, and T. Urabe (2005),Comparative study of the 2004 eruption with old eruptions at AsamaVolcano and activity evaluation, Bull. Volcanol. Soc. Jpn., 50, 303–313.

Okada, Y. (1985), Surface deformation due to shear and tensile faults in ahalf-space, Bull. Seism. Soc. Am., 75, 1135–1154.

Rubin, A. M., and D. Gillard (1998), Dike-induced earthquakes: Theore-tical considerations, J. Geophys. Res., 103, 10,017–10,030.

Rubin, A. M., D. Gillard, and J.-L. Got (1998), A reinterpretation of seis-micity associated with the January 1 983 dike intrusion at Kilauea Vol-cano, Hawaii, J. Geophys. Res., 103, 10,003–10,015.

Uhira, K., T. Baba, H. Mori, H. Katayama, and N. Hamada (2005), Earth-quake swarms preceding the 2000 eruption of Miyakejima volcano, Japan,Bull. Volcanol., 67, 219–230.

Waldhauser, F., and W. L. Ellsworth (2000), A double difference earthquakelocation algorithm: Method and application to the Northern HaywardFault, California, Bull. Seismol. Soc. Am., 90, 1353–1368.

Yamamoto, M., et al. (2005), A unique earthquake activity preceding theeruption at Asama volcano in 2004, Bull. Volcanol. Soc. Jpn., 50, 393–400.

�����������������������Y. Aoki, T. Ohminato, and M. Takeo, Earthquake Research Institute,

University of Tokyo, 1-1-1 Yayoi, Tokyo 113-0032, Japan. (takeo@eri.u-tokyo.ac.jp)M. Yamamoto, NS Solutions Corporation, 20-15 Shinkawa 2-chome,

Tokyo 104-8280, Japan.

L15310 TAKEO ET AL.: MAGMA SUPPLY PATH BENEATH MT. ASAMA L15310

5 of 5