Ground Deformation Associated with the,.,/ Unrest of Asama Volcano, Japan

Size: px

Start display at page:

Download "Ground Deformation Associated with the,**.,**/ Unrest of Asama Volcano, Japan"

Dennis Knight
5 years ago
Views:

1 /* (,**/) 0 /1/ /2.,**.,**/,**/ / +-,**/ ++. Ground Deformation Associated with the,**.,**/ Unrest of Asama Volcano, Japan Yosuke AOKI, Hidefumi WATANABE, Etsuro KOYAMA, Jun OIKAWA and Yuichi MORITA Ground deformation associated with the most recent eruptions in Asama volcano started on September +,,**., is reported. The ground deformation observed by continuous Global Positioning System measurements is modeled by dike intrusion for two di#erent periods; one is between July,,**., to March,,**/, which represents overall deformation during the unrest, and the other is between November,,**., to March,,**/, which represents the deformation during the latter half of the unrest. We assumed a rectangular dike opening uniformly in elastic, homogeneous, and isotropic medium to model the deformation field. To solve a nonlinear optimization problem in which model parameters are nonlinear to observed deformation field, the Simulated Annealing inversion was employed. The uncertainties of and trade-o#s between the model parameters are estimated by the Bootstrap method. The results show that the deformation field is well modeled by a dike striking roughly east-west, strike of which is consistent with the regional stress field. Shape of the dike, that is, length, width, and thickness, is not well constrained due to the small amount of deformation, up to +* mm, and the scarcity of GPS sites, but volume of the dike is well constrained to be 0.2, and..0- million cubic meters for the whole period and the latter half, respectively. The estimated depth of the dike tip is roughly + km below the sea level ; the depth of hypocenters is consistent with a theory of dike-induced earthquakes that they occur near the dike tip due to the stress concentration. However, the comparison of the location of the modeled dike and the distribution of earthquakes clearly shows that the hypocenter distribution is inconsistent with the theory described above, that is, the hypocenters are distributed only in the eastern half of the modeled dike tip. The possible reasons for this inconsistency are either +) earthquakes exists in the west half of the dike tip as well, but they are not detected due to the sparse distribution of seismometers at the west of the flank,,) the western half of the modeled dike is not capable of generating earthquakes because the temperature is too high for brittle failure of rocks, or -) the di#erential stress in the western half is so low that the area cannot reach the critical stress field even by the introduction of dike-tip stress concentration. Current geophysical observations cannot identify the reason but future development of geophysical observations is expected to solve the puzzle. Key words : ground deformation, Asama volcano, dike intrusion, Global Positioning System, geodetic inversion + (Fig. +a) ++*2 +12-,* +30* , **-, Earthquake Research Institute, University of Tokyo, + + Yayoi +, Bunkyo-ku, Tokyo ++- **-,, Japan. -23 *+++,+,/ Sawada, +33. ; Fujita and Ida, +333 ; Aoyama and Takeo,,**+ Asama Volcano Observatory of Earthquake Research Institute, University of Tokyo,,+,/ Nagakura, Karuizawa, Nagano -23 **++, Japan. Corresponding author: Yosuke Aoki yaoki@eri.u-tokyo.ac.jp

576 (+33*), km / km (,**.) GPS +330,**. (,**/),**.,**-.,**.,**. 1 (GPS) 3 + 3 +. +0 +2 3,-,3 +* +* ++ +.,**. +, 3 GPS Aoki et al., +333 ; Dzurisin,,**-,**. GPS,**.,**/ Fig. +. (a) Tectonic setting around Asama volcano.

2 576 (+33*), km / km (,**.) GPS +330,**. (,**/),**.,**-.,**.,**. 1 (GPS) ,-,3 +* +* ++ +.,**. +, 3 GPS Aoki et al., +333 ; Dzurisin,,**-,**. GPS,**.,**/ Fig. +. (a) Tectonic setting around Asama volcano. Triangles indicate the spatial distribution of active volcanoes. EUR, NAM, PAC and PHS stand for the Eurasian, North American, Pacific and Philippine Sea plates, respectively. A rectangle shows the area of Asama volcano shown in Fig. +b. (b) Spatial distribution of continuous GPS sites analyzed in this study. The area is shown by the rectangle in Fig. +a. Circles and rectangles are the sites operated by Earthquake ResearchInstitute, University of Tokyo, and Geographical Survey Institute of Japan, respectively. A dashed line denotes the prefectural boundary., GPS GPS Fig. +b +33* GPS,**.,**.,**/ GPS,**.,**., ASM. (Fig. +b),**. 3 + ASM/ ASM0 (Fig. +b),**. 3,**.

3 ,**.,**/ 577 Fig.,. An example of GPS time series for ASM. and ASM0 with respect to 3/*,1,, locations of which are shown in Fig. +b. Vertical bars mark the dates of moderate-sized eruptions. (a) Time series for ASM.. The middle panel, the north-south component, shows that the deformation started in late July,,**., and has moved southward by approximately +* mm until March,,**/. (b) Time series for ASM0 built in response to the first eruption on September +,,**.. An apparent jump seen in all three components around February,,**/, is due to snow removal on the GPS antenna. Time intervals a#ected by snow accumulation on the antenna are shown by a solid line in the middle panel. *,*3/1 *,*32/ *-S*.0 (Fig. +b),**.,**/ -* + GIPSY-OASIS (Zumberge et al., +331) ITRF,*** (Altamimi et al.,,**,) GPS - / mm, +* mm Fig., ASM. ASM0 3/*,1, (Fig. +b) 3/*,1, + mm Fig.,a ASM.,**. 1 +*,**/ - / mm +* mm Fig.,b ASM0,**/ +, +/ mm GPS,,**. +, / mm - - +,** *,**/ - + +* (period +),** *,**/ - + +* (period,)

$-) "#./& A : MNG/ ;&$) OP @ K "#./& A QRD S T U @ ;& V ) W XY @ ;& Z. Q0KL [ 1 \KL ;&$) ^* _`a ].!! [ e.! f`! g bc>d h!i Aj kabc>d, lm!$

4 µ()!* +!v,!¹º»!¼ ASM/!ASM0 "# $%& '() &* + $,-./ -ῌ, ΐῐῒῌ ῑ (Mogi, +3/2) :;< +32/) (Okada, )./ A 8 9 B C ;& D: :DE A,**. FG,**/ D: D F4 HI J A GF KL (Fig. -) "#./& A : MNG/ ;&$) K "#./& A QRD S T ;& V ) W ;& Z. Q0KL [ 1 \KL ;&$) ^* _`a ].!! [ e.! f`! g bc>d h!i Aj kabc>d, lm!no 2 lkl /G abc>d$ NG/& Okada (+32/) "# >d /G abc>d pvdqr L K st uv * &* ` A ) NG/ ED abc>d ^ pvw x yfd< DGD z{ 15 >d }>~ k N Cervelli et al.,,**+m Simulated Annealing $L E (Basu and Frazer, +33*) pvw x y & Simulated Annealing / / abc>d l *G/&!Kst EDabc>d OPK & uv abc>d abc>d Table + / ) u"y$ ;D E E # `D )& D3 Table + u U ^*G/&.!!Z.FG^*G / : K }>~ abc>dkd pv x pv Simulated Annealing K D< D abc>d b > ) FG pv ; pv x y< ^* $_ y$ vst abc>d ; < & b > $ E l/ st abc>d <; < G > ^*G/ F u"y : y$ Annealing ªK D OPK & (Efron and Tibshirani, +33-) ^*& & Cervelli et al. (,**+) }>~ ) 15 >d $L / / ab c>d t«4k D<abc>d'? > ^* :K` Bootstrap & E ; ±./&abc>d t ^* + "# >d u"d abc>d ²³ E% $L pv }>~ ^*G/&u"y t ^* Fig. -. Comparison of the observed deformation ﬁeld with the modeled deformation ﬁeld with a intruded dike for (a) period + (between July,,**. and March,,**/) and (b) period, (between November,,**. and March,,**/). Thick lines in each panel indicate the horizontal projection of the modeled dikes. +-s error ellipses of the observed deformation are also shown. Dashed line marks the prefectural boundary. Simulated ) ^*G/&u"y t Bootstrap st r dobs-dcal ^* K d obs d cal / /"# u"d abc>dfg,-./ ² ³ KL

5 ,**.,**/ 579 Table +. Possible bounds for dike parameters and estimated model parameters. Bounds Optimum parameters : period + Optimum parameters : period, min. max. min. opt. max min. opt. max Length (km) Width (km) Depth (km) a Dip (degree) b Strike (degree) c East (km) d North (km) d Open (m) Volume (m - ) e +4* +4* *4+./4* +/*4* +*4* /4* *4* N A +*4* +*4* /4* +-/4* -*4* /4* 14* +4* N A,4+*.2.4*3*,,43//, +*+4./.. 0.4,+/, /41+., -4**31 *413,1 042, +* 0 +4./2, +4,*1, *4-+/- 134-,.3 2*4--*/ 24-//3,4*131 *4+*/ * /*,/./4.22+,4///1 /4/+1, *43.1/ +*4. +* , +4++/. *4,1/ ,1 +*4**** +4-1*0 *4*-.1 *430 +* */ -4*0,* 0*40-/2 3* ** 04+-,, *43.+,.40- +* ,,+0.422*. +,24*/ *,,4*0*3 04/,/3 *421,* 14.* +* 0 Column + and, : Minimum and maximum possible values for each model parameter, respectively. Column - to / : Optimal model parameters for the period + that represent the whole period of the activity (July,,**., to March,,**/). Column. represents the optimum values estimated by the Simulated Annealing, and columns - and / represent the minimum and maximum possible values within 3/ confidence intervals estimated by the Bootstrap method. Column 0 to 2 : Same as column - to / except for that these columns are for the period, that represent the latter half of the activity (November,,**. to March,,**/). a) Depth to the top of the dike from the surface. Because the altitude of the surface in the area is about,.* km above the sea level, the estimated depth must be subtracted by about,.* km when it is translated to the depth below the sea level. For example, the top of the dike for the period +,,.30 km below the surface, should read about +.* km below the sea level. b) 3* degrees for a vertical dike. A value less than and larger than 3* degrees indicate a north-dipping and south-dipping dike, respectively, for an east-west striking dike. c) Measured clockwise from north, that is, a value of./.* indicates a northwest-southeast striking dike. d) The location of the center of the dike measured from the summit of Asama volcano. e) Calculated by multiplying length, width, and thickness of the dike obtained by the inversion., r p + r p* p [p(+) p(,) p(-) p(.) p (/)] T p* p* [p(-) p(,) p(-) p(+) p(/)] T r* p* r* - d* d cal r* Simulated Annealing. + -,,***.. + Table + Fig. -,**. 1 linear trend, +, Fig. - (Fig. -a),**. ++ (Fig. -b)., +* mm

6 580 Fig... An example of covariance scatter plots for the period +. We showsome selected parameter pairs only. Each panel indicates covariance between (a) area and thickness, (b) depth and volume, (c) east component of the location and length, (d) east component of the location and depth, (e) width and depth, and (f) width and volume, of the modeled dike, respectively. Bootstrap Table + 3/ - km GPS (Table +) Fig.. period + Fig. / period, period + (Fig..a) Period, (Fig. -b) period + (Fig..) (Fig..a) Fig..b Period + GPS (Fig..a)

7 ,**.,**/ 581 Fig. /. Same as Fig.. except that these plots are for the period,. Fig..c, Fig..d GPS Fig..d GPS Fig..e.f / / + Fig ; Seno, +333 /, Fig. - period, period + Table + period, + Aoki et al. (+333) Fig. 0 3/*,,+ ASM. Fig. 0,**. 1 +*,**

8 582 Fig. 0. Temporal changes of the baseline length between 3/*,,+ and ASM.. Because the modeled dike strikes approximately east-west, change of the baseline length between 3/*,,+ and ASM., spanning the volcano to the north-south, is a good indicator of inflation and deflation of the volcano. Vertical bars represent the time of moderate-sized eruptions. Arrows represent the time when the magma injection rate was high. + + (,**/) * 3 1 GPS / -,**., +* 0 m -,**/,**. 1,**/ - 3/..0- +* 0 +.*. +* 1 m ,**/ /. Rubin and Gillard, +332 Gillard et al., +330 ; Rubin et al., +332 ; Hayashi and Morita,,**- (,**/) + km period +, - km (Table +),,*** m (,**/) - km - km (,**/) + km GPS. km Fig. 1 + km +,**/,

9 ,**.,**/ 583 Fig. 1. A vertical view illustrating the relation between an intruded dike and earthquake locations. -, +, - 0,**.,**/,**. 1,**/ -,**. ++,**/ - +* mm GPS, Generic Mapping Tools (Wessel and Smith, +332) (+02***,) Altamimi, Z., Sillard, P. and Boucher, C. (,**,) ITRF,*** : A new release of the International Terrestrial Reference Frame for earth science applications. J. Geophys. Res., +*1 (B+*),,,+., doi: +*.+*,3/,**+JB***/0+. Aoki, Y., Segall, P., Kato, T., Cervelli, P. and Shimada, S. (+333) Imaging magma transport during the +331 seismic swarm o# the Izu Peninsula, Japan. Science,,20, 3,1 3-*. Aoyama, H. and Takeo, M. (,**+) Wave properties and focal mechanisms of N-type earthquakes at Asama volcano. J. Volcanol. Geotherm. Res., +*/, ,. Basu, A. and Frazer, L. N. (+33*) Rapid-determination of the critical-temperature in Simulated Annealing inversion. Nature,,.3, +.*3 +.+,. Cervelli, P., Murray, M. H., Segall, P., Aoki, Y. and Kato, T. (,**+) Estimating source parameters from deformation data, with an application to the March +331 earthquake swarm o# the Izu Peninsula, Japan. J. Geophys. Res., +*0, ++,,+1 ++,,-1. Dzurisin, D. (,**-) A comprehensive approach to monitoring volcano deformation as a window on the eruption cycle. Rev. Geophys.,.+ (+), doi: +*. +*,3/,**+ RG***+*1. Efron, B. and Tibshirani, R. J. (+33-) An Introduction to the Bootstrap. Chapman & Hall, New York,.-0 pp. Fujita, E. and Ida, Y. (+333) Low attenuation resonance of a spherical magma chamber: source mechanism of monotonic volcanic tremor at Asama Volcano, Japan. Geophys. Res. Lett.,,0, -,,+ -,,.. (,**.) Gillard, D., Rubin, A. M. and Okubo, P. (+330) Highly concentrated seismicity caused by deformation of Kilauea s deep magma system. Nature, -2., Hayashi, Y. and Morita, Y. (,**-) An image of a magma intrusion process inferred from precise hypocentral migrations of the earthquake swarm east of the Izu Peninsula. Geophys. J. Int., +/-, +/ (+33*) 0/ 00/ 2*1.

10 584 Mogi, K. (+3/2) Relations between the eruptions of various volcanoes and the deformations of the ground surfaces around them. Bull. Earthq. Res. Inst., -0, (,**/),**. /* -* Okada, Y. (+32/) Surface deformation due to shear and tensile faults in a half-space. Bull. Seism. Soc. Am., 1/, ++-/ ++/.. Rubin, A. M. and Gillard, D. (+332) Dike-induced earthquakes : theoretical considerations. J. Geophys. Res., +*-, +*,*+1 +*,*-*. Rubin, A. M., Gillard, D. and Got, J.-L. (+332) A reinterpretation of seismicity associated with the January +32- dike intrusion at Kilauea Volcano, Hawaii. J. Geophys. Res., +*-, +*,**- +*,*+/. Sawada, M. (+33.) B-type and explosion earthquakes observed at Asama volcano, central Japan. J. Volcanol. Geotherm. Res., 0-, +++ +,2. Seno, T. (+333) Syntheses of the regional stress fields of the Japanese islands. Isl. Arc, 2, (+33+) Wessel, P. and Smith, W. H. F. (+332) New, improved version of the Generic Mapping Tools released. EOS Trans. AGU, 13, /13. (,**/),**. V*// **,. +1 (,**/),**. /* /+3 /--. Zumberge, J., Heflin, M. B., Je#erson, D. C., Watkins, M. M. and Webb, F. H. (+331) Precise point positioning for e$cient and robust analysis of GPSdata from large networks. J. Geophys. Res., +*,, /**/ /*+1.

Haruhisa N. (Fig. + ) + Graduate School of Environmental Studies, Nagoya University, Furo-cho, Chikusa-ku, Nagoya.0. 20+ Japan.

Haruhisa N. (Fig. + ) *+ Graduate School of Environmental Studies, Nagoya University, Furo-cho, Chikusa-ku, Nagoya.0. 20*+ Japan. /- (,**2) 0,+/,,+ Source Mechanism and Seismic Velocity Structure of Source Region of Deep Low-frequency Earthquakes beneath Volcanoes: Case Studies of Mt Iwate and Mt Fuji Haruhisa N AKAMICHI + +3 (Fig