HAL Id: insu-00565059
https://hal-insu.archives-ouvertes.fr/insu-00565059
Submitted on 10 Feb 2011
HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.
Three-dimensional resistivity tomography of Vulcan’s forge, Vulcano Island, southern Italy
A. Revil, T. C. Johnson, Anthony Finizola
To cite this version:
A. Revil, T. C. Johnson, Anthony Finizola. Three-dimensional resistivity tomography of Vulcan’s
forge, Vulcano Island, southern Italy. Geophysical Research Letters, American Geophysical Union,
2010, 37, pp.15308. �10.1029/2010GL043983�. �insu-00565059�
Three-dimensional resistivity tom ography of Vulcan's forge, Vulcano Island, southern Italy
A. Revil,1,2 T. C. Johnson,3 and A. Finizola
4[1] 9,525 DC resistivity measurements were taken along 9 profiles crossing the volcanic edifice ofLa Fossa di Vulcano (the forge of God Vulcan in ancient Roman mythology), Vulcano Island (Italy) using a total of 9 58 electrode locations. This unique data set has been inverted in 3D by minimizing the L
2norm of the data misfit using a Gauss
Newton approach. The true 3D inversion was performed using parallel processing on an unstructured tetrahedral mesh containing 75,549 finite-element nodes and 398,208 elements to accurately model the topography of the volcanic edifice. The 3D tomogram shows a very conductive body (>0.1 Sim) comprised inside the Pietre Cotte crater with conductive volumes that are consistent with the position of temperature and CO
2anomalies at the ground surface. This conductive body is interpreted as the main hydrothermal body. It is overlaid by a resistive and cold cap in the bottom of the crater. The position of the conductive bo dy is consistent with the deformation source responsible for the observed 1990-1996 deflation of the volcano associated with a decrease of hydrothermal activity. Citation: Revil, A., T. C. Johnson, and A. Finizola (2010), Three-dimensional resistivity tomography of Vulcan's forge, Vulcano Island, southern Italy.
1. Introduction
[2] DC electrical res1st1v1ty imaging is a simple yet powerful technique to image and monitor active volcanoes and hydrothermal systems because resistivity is sensitive to porosity, water content, alteration ( cation exchange capacity), salinity (fDS), and temperature
[Revil et al.,2002, 2004, 2008;
Legaz et al.,2009]. Vulcano is an active volcanic island (3 x 7 km) located at the southernmost of the Aeolian Islands in the Tyrrhenian Sea in Italy (38
°24'N, 14
°58'E) along a major NW-SE tectonic fault named the Tindari
Letojanni fault. This fault crosses Sicily and the islands of Vulcano and Lipari and is associated with secondary faults, craters and volcanic tectonic depressions on these two islands (Figure la). On Vulcano island, La Fossa cone is a modest volcanic edifice located in its northern sector. It was formed in the last 6 Ky and gave its name to the ''vulcanian" type
1Department of Geophysics, Colorado School of Mines, Gol den, Colorado, USA.
2LGIT, UMR 5559, Universite de Savoie, INSU, CNRS, LeBourget
du-Lac, France.
3Energy Resource Recovery and Management , Idaho National Laboratory, Idaho Falls, Idaho, USA. 4
Laboratoire GeoSciences Reunion, UMR 7154, UR, IPGP, Saint
Denis, La Reunion.
of explosive activity. In the Roman mythology, Vulcano was considered to be the chimney to the forge of Vulcan (Hephaistos), the blacksmith of the roman gods. The activity of this volcanic island is characterized by both explosive phreatic eruptions and eruptions producing pumice fall deposits and lava flows. The morphology of La Fossa cone have been structured through five main periods of activity:
Punte Nere, Palizzi, Forgia Vecchia, Pietre Cotte, and Gran Cratere (Figure 1 b)
[Frazzetta et al.,1983 ]. Since its last eruption (1888-1890), Vulcano has been affected by two fumarolic crisis. The first one occurred in 1913-1923. The second episode began in 1977 and lasts until nowadays.
During this last fumarolic activity, magmatic fluid inputs have been evidenced by the geochemistry of the fumaroles in 1979-1981, 1985, 1988, 1996, and 2004- 2005 in association with swarms of seismic events. Since 1977, the temperature progressively increased and reached 690
°C in May 1993.
[3] Because of its dimensions and accessibility, this edifice has been a natural laboratory for geochemists, volcanologists and geophysicists interested to understand magmatic hydro
thermal systems
[Granieri et al.,2006;
Blanco-Montenegro et al.,2007]. In the present letter, we performed a 3D Gauss-Newton inversion of the resistivity data presented by
Revil et al.[2008] and we discuss the resulting tomogram in terms of localization of the hydrothermal body, fumarolic vents, and tectonic features. This is the first time that a true 3D resistivity inversion is performed over a volcanic edifice (usually inversions of DC or MT data are 1 D or 2D with an interpolation between the inverted data to form a pseudo-3D dataset of resistivity).
2. Data Acquisition and Inversion
[4] Resistivity data were measured in October 2005, May 2006, and October 2006 along nine profiles including roll
overs of the electrodes with a total of 9,525 resistivity mea
surements (Figure 2 ). We use a set of 64 brass electrodes with a take-out of 20 meters and a cable of 1.26 km. Each of these 9,525 resistivity measurements represents actually the mean of 8 to 16 distinct measurements stacked together with the same set of AB(current)-MN(potential) electrodes. Con
tact between the electrodes and the ground was improved by adding salty water at each electrode location but it was dif
ficult, most of the time, to inject more than 50 mA in the ground. We used the Wenner-a array for its good signal
to-noise ratio. The temperature measurements reported in Figure 2 were performed at a depth of 30 ± 2 cm
[Revil et al.,2008;
Barde-Cabusson et al.,2009]. This dataset was merged with the detailed grid temperature measurements performed with the same approach in August 1996 inside the crater of La Fossa cone by
Aubert et al.[2008]. Thanks to a multi
sensor permanent monitoring station at Stromboli volcano,
F igure 1 . Map o f the Is land o f Vu lcano ( I ta ly ) and i ts tec ton ic con tex t . (a ) Ma in tec ton ic fau l t sys tem c ross ing the Vu lcano and L ipa r i is lands ( f romMazzuo l i e t a l . [1995 ] and mod i f ied byG ioncada e t a l .[2003 ] ) . (b ) Loca t ion o f the ma in c ra te r r ims o f La Fossa Cone (mod i f ied f romFrazze t ta e t a l .[1983 ] ) . The d ig i ta l e leva t ion map o f Vu lcano is f romBa ld i e t a l . [2002 ] .
F igure 2 . Tempe ra tu re map a t a dep th o f 30 cm . (a ) Tempe ra ture map o f La Fossa d i Vu lcano w i th the pos i t ion o f the
res is t iv i ty pro f i les ( labe led 1 to 9 , see the sma l l wh i te do ts ) and the 958 e lec t rodes loca t ions ( the sma l l wh i te do ts ) . The
tempera ture map was ob ta ined merg ing our therma l da ta se t w i th the 2392 measuremen ts performed byAuber t e t a l .
[2008 ] . (b ) Zoom in the inne r pa r t o f the c ra te r . The sma l le r wh i te do ts come f romAuber t e t a l .[2008 ] and the la rge r
one f rom ou r s tudy . The b lack dash l ine ind ica tes the bo t tom o f the c ra te r . No te the co ld tempe ra tu re a t the bo t tom o f the
c ra te r . F1 , F2 , and F3 : loca t ion o f the ma in fuma ro l ic a reas . O the r symbo ls : same as F igu re 1 .
a p ro toco l o f co r rec t ion has been es tab l ished to co r rec t the 1996 measu remen ts f rom the seasona l tempe ra tu re chan - ges . The comb ined tempe ra tu re maps (F igu re 2 ) p rov ides an independen t ve r i f ica t ion o f the hyd ro the rma l ac t iv i ty in the v ic in i ty o f the g round su r face .
[
5] The fo rwa rd mode l l ing res is t iv i ty code so lves Po isson equa t ion fo r the e lec t r ica l po ten t ia l w i th the f in i te e lemen t me thod us ing the fo rmu la t ion o f Rücker e t a l .[2006 ] . We cons ide r res is t iv i ty as sca la r and no t as a second o rde r symme t r ic tenso r ( the re fo re the assume iso t ropy o f the ma te r ia l ) . We f i rs t compu ted the po le so lu t ion fo r each e lec t rode used as a cu r ren t sou rce o r s ink . The po ten t ia l d is t r ibu t ion fo r a d ipo le AB sou rce /s ink cu r ren t pa i r is then cons t ruc ted by sub t rac t ing the po le po ten t ia l assoc ia ted w i th the cu r ren t s ink . We imp lemen ted th is t rue 3D inve rs ion app roach in a pa ra l le l env i ronmen t us ing 480 p rocesso rs on the Ices to rm c lus te r loca ted a t the Idaho Na t iona l Labo ra - to ry . The p rocesso rs a re o rgan ized in a "mas te r /s lave"
con f igu ra t ion whe re one node se rves as the mas te r node , wh ich con t ro ls execu t ion o f the p rog ram and ins t ruc ts s laves conce rn ing wh ich rou t ines to execu te . A comp le te desc r ip - t ion o f the pa ra l le l inve rs ion a lgo r i thm is g iven byJohnson e t a l .[2010 ] .
[
6] The inve rs ion m in im izes an ob jec t ive func t ion , wh ich is the sum o f the L
2no rm o f the we igh ted da ta m is f i t and the L
2no rm o f the mode l regu la r iza t ion e r ro r . The regu la r iza - t ion ope ra to r is fo rmu la ted so tha t the regu la r iza t ion e r ro r inc reases as the es t ima ted conduc t iv i ty s t ruc tu re dev ia tes f rom homogene i ty . The re fo re , the va lue o f the ob jec t ive func t ion is gove rned by a t rade‐o f f be tween hono r ing the we igh ted res is t iv i ty da ta (wh ich impose he te rogene i ty ) , and hono r ing the regu la r iza t ion cons t ra in ts (wh ich impose homogene i ty ) . The re la t ive impo r tance p laced upon these compe t ing ob jec t ives is gove rned by the regu la r iza t ion o r t rade ‐o f f pa rame te r , wh ich sca les the regu la r iza t ion te rm o f the ob jec t ive func t ion . The t rade ‐o f f pa rame te r is in i t ia l ly se t to a la rge va lue the reby impos ing homogene i ty in the inve rse so lu t ion , and then dec reased as the a lgo r i thm p ro - g resses to a l low he te rogene i ty su f f ic ien t on ly to hono r the we igh ted res is t iv i ty da ta . Thus , the on ly he te rogene i ty d is - p layed in the f ina l so lu t ion is tha t wh ich the res is t iv i ty da ta a re ab le to reso lve . The ind iv idua l res is t iv i ty da ta a re we igh ted acco rd ing to da ta no ise us ing a da ta we igh t ing ma t r ix . The ob jec t ive func t ion is m in im ized w i th the Gauss‐
New ton a lgo r i thm as desc r ibed byJohnson e t a l .[2010 ] . [
7] To cons t ruc t the mesh , known topog raphy was used ove r the su rvey a rea and ex t rapo la ted fu r the r away to avo id edge e f fec ts . The compu ta t iona l mesh cons is ts o f 958 e lec t rode loca t ions , 75 ,549 f in i te‐e lemen t nodes , and 398 ,208 te t rahed ra l e lemen ts . The re la t ive ly la rge numbe r o f e lemen ts w i th respec t to the numbe r o f da ta in th is case was requ i red to accu ra te ly mode l the comp l ica ted su r face topog raphy . Howeve r , i t shou ld be kep t in m ind tha t mode l reso lu t ion a t a pa r t icu la r sca le is no t a func t ion o f the numbe r o f e lemen ts in the mode l . I t is a func t ion o f the in fo rma t ion p rov ided by the da ta . The inve rse compu ta t ions we re conduc ted on 480 nodes (1 mas te r and 479 s lave nodes ) , requ i r ing app rox ima te ly 2 hou rs (26 Gauss‐New ton i te ra t ions ) fo r conve rgence . Conve rgence c r i te r ia a re gen - e ra l ly se t based upon da ta e r ro r . Tha t is , when the L
2no rm o f the da ta m is f i t reduces to a va lue tha t is cons is ten t w i th
the es t ima ted da ta no ise , the inve rse so lu t ion is assumed to have conve rged .
3 . In terpre ta t ion
[
8] The e lec t r ica l res is t iv i ty tomog ram resu l t ing f rom the invers ion is shown in F igures 3b –3d (see a lso aux i l iary ma te r ia l ) .
1A h is tog ram o f the da ta m is f i t fo r the inve rse so lu t ion is shown in F igu re 3e . He re the e r ro r is de f ined as the res idua l no rma l ized by the obse rved va lue . W i th respec t to a no rma l d is t r ibu t ion , the e r ro r h is tog ram d isp lays e lon - ga ted ta i ls . These a re l ike ly due to the low leve l o f cu r ren t in jec ted in to the vo lcano , as exp la ined above , and resu l t ing no isy da ta . Because the re a re no c ross‐l ine measu remen ts be tween the 9 p ro f i les , the su r face res is t iv i ty s t ruc tu re is unce r ta in in many a reas and may con ta in a r t i fac ts . The deepe r s t ruc tu re is be t te r reso lved espec ia l ly in the cen t ra l pa r t o f the ed i f ice . The res is t iv i ty tomog ram shows a co r - respondence be tween the h igh conduc t iv i ty a reas (>0 .1 S /m ) and the pos i t ion o f the ma in su r face tempe ra tu re anoma l ies ev idenced by therma l infrared images ofCh iod in i e t a l . [2005 , F igu re 1a ] , o r in a de ta i led so i l tempe ra tu re s tudy a t 30 cm dep th [Auber t e t a l ., 2008 ] coup led w i th ou r da tase t (see F igu re 2b ) . F igu re 2b c lea r ly ev idences co ld tempe ra - tu res a t the bo t tom o f the p resen t c ra te r . Th is the rma l cha r - ac te r is t ic is a lso cons is ten t w i th the CO
2anoma ly d is t r ibu t ion desc r ibed byGran ier i e t a l . [2006 , F igu re 2 ] . Th is h igh conduc t iv i ty subvo lume is the re fo re assoc ia ted to the h igh tempe ra tu re o f the hyd ro the rma l f lu ids (>200°C ) ev idenc ing h igh pe rmeab i l i ty pa thways . Howeve r the p resence o f a magne t ized body ins ide the Fossa cone [B lanco‐Mon tenegro e t a l ., 2007 ] imp l ies tha t h igh tempe ra tu res a re con ta ined in ve ry l im i ted spaces and do no t a f fec t i ts bu lk inne r s t ruc tu re . Add i t iona l con t r ibu t ions to h igh conduc t iv i ty cou ld come f rom a l te ra t ion and sa l in i ty o f the b r ines .
[
9] The vo lume o f the h igh conduc t iv i ty body a t dep th is less than 0 .1 km
3in ag reemen t w i th the s ize o f the hyd ro - the rma l sys tem in fe r red byI ta l iano e t a l .[1984 ] us ing mass ba lance ca lcu la t ions and is comp r ised ins ide the P ie t re Co t te c ra te r . The pos i t ion o f th is body is a lso in ag reemen t w i th the subve r t ica l p ro la te sphe ro ida l sou rce in fe r red byGamb ino and Gug l ie lm ino[2008 ] f rom the obse rved 1990 –1996 de f la - t ion o f La Fossa d i Vo lcano . Th is re in fo rces the i r in te rp re - ta t ion tha t the f lu id loss f rom the sha l low hyd ro the rma l rese rvo i r shown in F igu re 3b was the mos t l ike ly cause o f the subs idence reco rded a t La Fossa Cone du r ing the pe r iod 1990–1996 . Indeed , the max imum o f tempe ra tu re occu r r ing in May 1993 cou ld be in te rp re ted as a consequence o f the las t magma t ic inpu t in 1988 , wh ich p ressu r ized the hyd ro the rma l sys tem , and the consequen t re lease o f the f lu id p ressu re and the assoc ia ted de f la t ion o f the ed i f ice du r ing the pe r iod 1990 – 1996 . In 1996 , a new magma t ic inpu t p ressu r ized aga in the sys tem .
[
10] The fuma ro les obse rved a t La Fossa cone a re a m ix o f a gas con t r ibu t ion re leased f rom a magma body (∼2 .5 km dep th [Nucc io and Paon i ta, 2001 ] ) and a sha l lowe r one fo rmed by the evapo ra t ion o f b r ines o f ma r ine o r ig in [Gran ier i e t a l ., 2006 ] . We no te a s t rong d i f fe rence in the pa t te rn and va lue o f the e lec t r ica l conduc t iv i ty be tween the SN p ro f i le (F igu re 3d )
1Auxiliary materials are available in the HTML. doi:10.1029/
2010GL043983.
and the EW p ro f i le (F igu re 3c ) . The EW p ro f i le shows res is t ive bod ies on the f lank o f the vo lcano . The SN p ro f i le is mo re conduc t ive . Th is is an expec ted as the g row th o f Vu lcano is con t ro l led by the NW–SE , T inda r i–Le to jann i s t r ike‐s l ip fau l ts and by the assoc ia ted N–S s t r ik ing no rma l fau l ts (F igu re 1a ) . Mo reove r , th is SN p ro f i le a lso c rosses h ighe r pe rmeab i l i ty a reas co r respond ing to o ld c ra te r bound - a r ies used as d ra ins by hyd ro the rma l f lu ids (Fo rg ia Vecch ia on No r the rn f lank and Pa l izz i on Sou the rn f lank (F igu res 1b
and 2 ) ) . The res is t ive body on the eas te rn s ide o f the vo lcano has been in terpre ted as an o ld dome s truc ture byBarde‐
Cabusson e t a l ., [2009 ] .
[
11] The 3D res is t iv i ty tomog ram cou ld be used a lso to
assess the CO
2p lumb ing sys tem o f the vo lcano as CO
2f lux
anoma l ies reco rded a t the g round su r face a re a lso cons is ten t
w i th the pos i t ion o f the conduc t iv i ty anoma ly . DC res is t iv i ty
is the re fo re a use fu l in fo rma t ion i f we wan t to mode l , w i th a
reac t ive t ranspo r t code , the CO
2f lux changes obse rved a t the
F igure 3 . 3D res is t iv i ty tomog ram o f La Fossa d i Vu lcano us ing the log o f the e lec t r ica l conduc t iv i ty . (a ) P ic tu re taken
f rom the No r th . C ra te rs : PN , Pun te Ne re ; FV , Fo rg ia Vecch ia ; PC , P ie t re Co t te ; GC , G ran C ra te re . F1 , F2 , and F3 : a reas o f
in tense fuma ro l ic ac t iv i ty . (b –d ) Res is t iv i ty tomog ram and i ts in te rp re ta t ion (ve r t ica l exagge ra t ion 2 :1 ) . No te the asymme t ry
o f the conduc t iv i ty d is t r ibu t ion be tween the EW and SN p ro f i les . (e ) Pe rcen tage e r ro r d is t r ibu t ion .
g round su r face byGran ier i e t a l .[2006 ] and co r respond ing to inc rease vo la t i le re lease f rom a s ta t iona ry magma sys tem . The anoma ly F1 in F igu re 3 co r responds to the pos i t ion o f the ma in fuma ro l ic f ie ld . The anoma ly F2 and F3 co r responds a lso to a reas o f the rma l and CO
2anoma l ies . These su r face anoma l ies can be ex tended downwa rd to the ma in hyd ro - the rma l body . I t wou ld be ve ry in te res t ing in fu tu re mea - su remen ts to assess how DC res is t iv i ty tomog raphy cou ld be used to assess changes in the hyd ro the rma l sys tem l ike i ts tempe ra tu re and the CO
2con ten t . A co r re la t ion be tween these changes and the se ism ic i ty , the magma t ic inpu t , and the de fo rma t ion o f the ed i f ice wou ld be o f the f i rs t in te res t . The a im wou ld be to be t te r cons t ra in the in f luence o f the hyd ro - the rma l sys tem o f La Fossa rega rd ing un res t and e rup t ive even ts .
4 . Conc lus ions
[
12] We pe r fo rmed a t rue 3D res is t iv i ty tomog raphy o f La Fossa d i Vu lcano (Aeo l ian Is lands , I ta ly ) us ing a non‐l inea r Gauss‐New ton a lgo r i thm and pa ra l le l p rog ramm ing . T ime lapse DC and comp lex res is t iv i ty tomog raphy a re expec ted to inc rease ou r concep tua l unde rs tand ing o f the a rch i tec tu re o f ac t ive vo lcanoes in com ing yea rs and w i l l o f fe r a powe r fu l me thod to mon i to r the fundamen ta l chem icohyd romechan ica l p rocesses occu r r ing in magma t ic hyd ro the rma l sys tems . In the case o f La Fossa d i Vu lcano , jo in t inve rs ion o f DC res is t iv i ty w i th magne t ic and g rav i ty da ta cou ld he lp to ge t a c lea re r p ic tu re o f the geo logy and tec ton ic fea tu res a f fec t ing th is ed i f ice and reduce the inhe ren t non ‐un iqueness o f po ten - t ia l f ie ld me thods .
[
13]
Acknowledgments. Financial support was provided by DOE (award GO18195), INSU‐CNRS, the Laboratoire GéoSciences Réunion (France), CNR, INGV, and the Dipartimento per la Protezione Civile (Project V3.5 Vulcano, 2005–2007)inItaly. Wethank S. Piscitelli, E. Rizzo, T. Ricci, A. Angeletti, A. Crespy, M. Balasco, S. Barde Cabusson, L. Bennati, S. Byrdina, N. Carzaniga, F. Di Gangi, J. Morin, A. Perrone, M. Rossi, E. Roulleau, B. Suski, Xavier Rassion, and Etienne Wheris fortheir partic- ipationtothethree field surveys and Alicia Hotovec forthe compilation of the data. A specialthanksto M. Marsellafor providing us withthe highres- olution digital elevation model of VulcanoIsland,J. Vandemeulebrouck and M. Todesco for fruitful discussions. Wethank Eric Calais andtwo anony- mous referees fortheir useful comments. IPGP contribution 2656.Re ferences
Aubert, M., S. Diliberto, A. Finizola, and Y. Chébli (2008), Double origin of hydrothermal convective flux variations in the Fossa of Vulcano (Italy),Bull. Volcanol.,70, 743–751, doi:10.1007/s00445-007-0165-y. Baldi, P., et al. (2002), Validation and comparison of different techniques
for the derivation of digital elevation models and volcanic monitoring (Vulcano Island, Italy),Int. J. Remote Sens.,23(22), 4783–4800, doi:10.1080/01431160110115861.
Barde‐Cabusson, S., et al. (2009), New geological insights and structural control on fluid circulation inthe Fossa cone (Vulcano, Aeolian Islands,
Italy),J. Volcanol. Geotherm. Res.,185, 231–245, doi:10.1016/ j.jvolgeores.2009.06.002.
Blanco‐Montenegro, I., R. De Ritis, and M. Chiappini (2007), Imaging and modelling the subsurface structure of volcanic calderas with high‐resolution aeromagnetic data at Vulcano (Aeolian Islands, Italy), Bull. Volcanol.,69, 643–659, doi:10.1007/s00445-006-0100-7. Chiodini, G., et al.(2005), Carbon dioxide diffuse degassing and estimation
of heat release from volcanic and hydrothermal systems,J. Geophys. Res.,110, B08204, doi:10.1029/2004JB003542.
Frazzetta, G., L. La Volpe, and M. F. Sheridan (1983), Evolution of the Fossa cone, Vulcano,J. Volcanol. Geotherm. Res.,17,329–360, doi:10.1016/0377-0273(83)90075-6.
Gambino, S., and F. Guglielmino (2008), Ground deformation induced by geothermal processes: A model for La Fossa Crater (Vulcano Island, Italy),J. Geophys. Res.,113, B07402, doi:10.1029/2007JB005016. Gioncada, A., R. Mazzuoli, M. Bisson,and M. T. Pareschi(2003), Petrology
of volcanic products youngerthan 42 ka onthe Lipari Vulcano complex (AeolianIslands,Italy): Anexample of volcanism controlled bytectonics, J. Volcanol. Geotherm. Res.,122, 191–220, doi:10.1016/S0377-0273(02) 00502-4.
Granieri, D., et al. (2006), Correlated increase in CO2fumarolic content and diffuse emission from La Fossa crater (Vulcano, Italy): Evidence of volcanic unrest or increasing gas release from a stationary deep magma body?,Geophys. Res. Lett.,33, L13316, doi:10.1029/2006GL026460. Italiano, F., P. M. Nuccio, and M. Valenza(1984), Mass geothermal energy
release at Vulcano, Aeolian Islands, Italy,Bull. Mineral. Rend. Soc. Ital. Mineral. Petrol.,39, 379–386.
Johnson, T. C., R. J. Versteeg, A. Ward, F. D. Day‐Lewis, and A. Revil (2010), Improved hydrogeophysical characterization and monitoring through high performance electrical geophysical modeling andinversion, Geophysics, in press.
Legaz, A., et al. (2009), A case study of resistivity and self‐potential sig- natures of hydrothermal instabilities, Inferno Crater Lake, Waimangu, New Zealand, Geophys. Res. Lett.,36, L12306, doi:10.1029/
2009GL037573.
Mazzuoli, R., L. Tortorici, and G. Ventura(1995), Obliqueriftingin Salina, Lipari, and Vulcano islands (Aeolian Islands, southern Italy),Terra Nova,7, 444–452, doi:10.1111/j.1365-3121.1995.tb00540.x.
Nuccio, P. M., and A. Paonita(2001), Magmatic degassing of multicompo- nent vapors and assessment of magma depth: Application to Vulcano Island (Italy),Earth Planet. Sci. Lett.,193,467–481, doi:10.1016/ S0012-821X(01)00512-X.
Revil, A., D. Hermitte, E. Spangenberg, and J. J. Cochémé(2002), Electrical properties of zeolitized volcaniclastic materials,J. Geophys. Res., 107(B8), 2168, doi:10.1029/2001JB000599.
Revil, A., A. Finizola, F. Sortino, and M. Ripepe(2004), Geophysicalinves- tigations at Stromboli volcano, Italy. Implications for ground water flow and paroxysmal activity,Geophys. J. Int.,157(1), 426–440, doi:10.1111/j.1365-246X.2004.02181.x.
Revil, A., et al. (2008), Inner structure of La Fossa di Vulcano (Vulcano Island, southern Tyrrhenian Sea, Italy) revealed by high resolution elec- tric resistivity tomography coupled with self‐potential, temperature, and soil CO2diffuse degassing measurements,J. Geophys. Res.,113, B07207, doi:10.1029/2007JB005394.
Rücker, C., T. Günther, and K. Spitzer (2006), Three‐dimensional mod- eling andinversion of DC resistivity dataincorporatingtopography—I. Modelling, Geophys. J. Int.,166,459–505, doi:10.1111/j.1365-246X. 2006.03010.x.
A. Finizola, Laboratoire GéoSciences Réunion, UMR 7154, UR, IPGP, 15 avenue René Cassin, BP 7151, F‐97715 Saint‐Denis CEDEX 9, La Réunion.
T. C. Johnson, Energy Resource Recovery and Management, Idaho National Laboratory, PO Box 1625, Idaho Falls, ID 83415‐2107, USA.
A. Revil, Department of Geophysics, Colorado School of Mines, 1500 Illinois St., Golden, CO 80401, USA. ([email protected])
