Downloaded from geology.gsapubs.org on March 6, 2011 Morphometry and evolution of arc volcanoes

Pablo Grosse1, Benjamin van Wyk de Vries2, Iván A. Petrinovic3, Pablo A. Euillades4, and Guillermo E. Alvarado5 1CONICET (Consejo Nacional de Investigaciones Científi cas y Técnicas) and Fundación Miguel Lillo, Miguel Lillo 205, (4000) de Tucumán, 2Laboratoire Magmas et Volcans, Centre National de la Recherche Scientifi que-Unité Mixte de Recherche (CNRS-UMR6524), Université Blaise Pascal, 5 Rue Kessler, 63038 Clermont-Ferrand, France 3CONICET–IBIGEO (Consejo Nacional de Investigaciones Científi cas y Técnicas–Instituto de Bio y Geociencias), Universidad Nacional de Salta, Mendoza 2, (4400) Salta, Argentina 4Instituto CEDIAC (Capacitación Especial y Desarrollo de la Ingeniería Asistida por Computadora), Universidad Nacional de Cuyo, Ciudad Universitaria, (5500) Mendoza, Argentina 5Área de Amenazas y Auscultación Sísmica y Volcánica, Instituto Costarricense de Electricidad, Apartado 10032-1000, Costa Rica

ABSTRACT /tephra ratio, and deformation, and ulti- Volcanoes change shape as they grow through eruption, intrusion, , and deforma- mately on underlying factors such as magma tion. To study shape evolution we apply a comprehensive morphometric analysis to fl ux and tectonic setting. two contrasting arcs, and the southern Central . Using Shuttle Radar Since Cotton (1944) there have been rela- Topography Mission (SRTM) digital elevation models, we compute and defi ne parameters for tively few studies of volcano morphology, plan (ellipticity, irregularity) and profi le (height/width, summit/basal width, slope) shape, as although Francis (1993) and Thouret (1999) well as size (height, width, volume). We classify volcanoes as cones, sub-cones, and massifs, gave broad overviews. The morphometry of and recognize several evolutionary trends. Many cones grow to a critical height (~1200 m) some specifi c volcano types has been studied in and volume (~10 km3), after which most widen into sub-cones or massifs, but some grow into detail, such as cinder cones (e.g., Wood, 1980; large cones. Large cones undergo sector collapse and/or gravitational spreading, without sig- Riedel et al., 2003), oceanic shields (e.g., Cul- nifi cant morphometry change. Other smaller cones evolve by vent migration to elliptical sub- len et al., 1987; Michon and Saint-Ange, 2008), cones and massifs before reaching the critical height. The evolutionary trends can be related to seamounts (e.g., Smith, 1996), and extraterres- magma fl ux, edifi ce strength, structure, and tectonics. In particular, trends may be controlled trial volcanoes (e.g., Plescia, 2004). Systematic by two balancing factors: magma pressure versus lithostatic pressure, and conduit resistance morphometric studies of polygenetic arc volca- versus edifi ce resistance. Morphometric analysis allows for the long-term state of individual noes are scarce at both individual and regional or volcano groups to be assessed. Morphological trends can be integrated with geological, scale (e.g., Wood, 1978; Lacey et al., 1981; geophysical, and geochemical data to better defi ne volcano evolution models. Carr, 1984; van Wyk de Vries et al., 2007), lead- ing to varying morphological classifi cations that lack consensus, with different and overlapping INTRODUCTION evolves depending on the prevailing processes. terms such as simple, composite, compound, Volcano edifi ce shape and size result from Thus, volcano morphology potentially contains complex, cluster, multiple, twin, shield-like, and the interplay between constructive and destruc- information on the balance of such factors as collapse scarred (e.g., compare classifi cations tive (erosional and deformational) processes age, growth stage, composition, eruption rate, given in Macdonald, 1972; Pike and Clow, (Fig. 1A). During a volcano’s life, its shape vent position and migration, degree of erosion, 1981; Francis, 1993; Simkin and Siebert, 1994; Davison and De Silva, 2000). Clearly, detailed morphometric studies are needed for a more Figure 1. A: Three-dimen- A rigorous quantitative classifi cation and a better sional (3-D) images de- rived from Shuttle Radar understanding of volcano shape evolution. Hone Topography Mission digital et al. (2007) went in this direction by means of elevation models show- cladistic analysis. 1 2 ing different shapes of arc We present a morphometric analysis of poly- volcanoes. 1—Concep- genetic volcano edifi ces from two continental ción (11.538°N, 85.623°W), simple symmetrical cone; arcs, the Central American Volcanic 2—Ollagüe (21.308°S, Front) and the southern Central Andes Volcanic 68.180°W), more complex Zone. We quantify, characterize, and classify cone; 3— volcanic edifi ce morphology, and then detect (21.225°S, 68.469°W), com- 3 4 posite volcano with a sub- shape evolution trends that we relate to evolu- conical shape; 4—Rincón tionary processes. Here we specifi cally look for de La Vieja (10.809°N, and interpret general trends; complementary 85.319°W), com plex mas- detailed analyses of individual volcanoes should sif. B: 3-D image of Ara- Summit area Elevation B contours Shape descriptors 6.0 Summit area be a subsequent step. car (24.297°S, 67.783°W) Height - Ellipticity index (ei) showing acquired mor- - Irregularity index (ii)

)mk ( 5.5 phometric parameters and Slopes ei (x10) MORPHOMETRIC PARAMETERS

Volume noi tave corresponding diagram of 5.0 We have used 90 m spatial resolution digi- ii (x10) elevation versus slope, el- Slope tal elevation models (DEM) from the Shuttle lipticity index, and irregu- lE 4.5 Radar Topography Mission (SRTM). This is larity index. See the Data Edifice outline Lowest closed Best-fit contour the best high-resolution global DEM data set Repository (see footnote 1) Area, width surface 4.0 for all locations and data. 10 15 20 25 30 (e.g., Rabus et al., 2003), and it is adequate for

© 2009 Geological Society of America. For permission to copy, contact Copyright Permissions, GSA, or [email protected]. GEOLOGY,Geology, July July 2009; 2009 v. 37; no. 7; p. 651–654; doi: 10.1130/G25734A.1; 5 fi gures; Data Repository item 2009151. 651 Downloaded from geology.gsapubs.org on March 6, 2011 morphometric studies of stratovolcanoes (e.g., noes have a wide variety of shapes and sizes. 0.25), and circular (low ei) and regular (low ii) Wright et al., 2006; Kervyn et al., 2008). We They are contrasting examples of continental plan shapes (Fig. 3). Average fl ank slopes are have analyzed 59 Central American Volcanic margin arcs: the Central American Volcanic 21º–34° and maximum interval slopes are 27º– Front and 56 southern Central Andes Volcanic Front is developed on thin to thick crust, contains 37° (Figs. 3 and 4). There is a ~300 m height Zone edifi ces (see the GSA Data Repository1 for many young and historically active volcanoes, interval at 1140–1430 m (corresponding to vol- , map, and additional material). Selected and has a humid, erosive climate; the southern umes of 9–13 km3) where there is a clear lack volcanoes have shown activity Central Andes Volcanic Zone is on thick crust, of cones (only one volcano, Azufre, is present) (Smithsonian Institution database, Siebert and most volcanoes are dormant or extinct, and it (Fig. 2). The cones above this “cone gap” have

Simkin, 2002) or are morphologically fresh. has a very arid, low-erosion climate. slightly lower H/WB, generally higher WS/WB, The seamless SRTM DEMs from the CGIAR- Figures 2–4 graphically display the morpho- and are more irregular and elliptical (Figs. 3 and CSI (Consultative Group on International Agri- metric features (also see Table DR1 in the Data 4). Within this large cone subgroup is a set of cultural Research–Consortium for Spatial Infor- Repository). Edifi ces of both arcs are grouped paired or twin cones (Atitlán-Tolimán, Fuego- mation) were used (Jarvis et al., 2008). into four main shape classes: cone, sub-cone, , and San Pedro–San Pablo), which

A basic morphometric uncertainty is the massif, and shield. This classifi cation is not are characterized by higher WS/WB ratios and ei selection of volcano extent, as the aprons can absolute, as there is gradation and overlap in the values (Fig. 3). merge with the surrounding landscape. We data; it is based on a fi rst-order grouping using thus restrict our analysis strictly to the edi- the H/WB ratio, then refi ned using the WS/WB Sub-Cones

fi ces, as they are generally clear landforms. ratio, the ei and ii, and the average fl ank slopes. Sub-cones have intermediate H/WB of 0.10–

Consequently, size data are an estimation of Field knowledge and qualitative evaluation of 0.16; their WS/WB, plan shapes and slope val- edifi ce size only and not total erupted volume. DEMs, satellite images, and geological maps ues are very variable, but are also intermediate The outline of each edifi ce was user-estimated were used to sort out quantitatively uncertain (Figs. 3 and 4). The larger sub-cones (volumes (details in the Data Repository). cases. The morphometric differences between > 13 km3) tend to be more irregular than the Morphometric parameters were acquired cones and massifs are clearly evident, while smaller ones (Fig. 4). With the exception of using an expressly written IDL (interactive sub-cones are transitional. Within each type, unusually large -Pajonales, the sub-cones data language) code (MORVOLC; see the differences between Central American Volca- have heights of 400–1400 m and volumes Data Repository for detailed descriptions of nic Front and southern Central Andes Volcanic between 1 and 46 km3. The lack of larger sub- the parameters used). Basal area and width are Zone edifi ces can be found, but are small com- cones with sizes equivalent to the larger cones obtained from the outline. The outline is also pared to differences between types. Shields are and massifs creates a “sub-cone gap” at heights used to compute a best-fi t surface from which only found in the Central American Volcanic >1400 m and volumes >46 km3 (Figs. 2 and 3). height and volume are derived (Fig. 1B). Front; they form a special subset of volcanoes There are different edifi ce types within the sub- The shape of elevation contours at 50 m with large that we do not consider here. cone class; some (e.g., ) have low ellip- intervals is described using two independent ticity and smaller summit areas, while others indexes (Fig. 1B): (1) ellipticity index (ei), Cones (e.g., , Aucanquilcha), are more elliptical which quantifi es contour elongation; and (2) Cones have a simple conical shape, with cir- and have larger summit areas. irregularity index (ii), which quantifi es con- cular base and steep, smooth concave profi le. tour irregularity or complexity. The ei and ii Their heights are 350–2250 m and volumes are Massifs 3 3 of successive contours defi ne two independent <1 km to 75 km (Fig. 2). They have elevated Massifs have low H/WB (<0.10), large sum- profi les that together summarize volcano plan H/WB (>0.15), small summit areas (WS/WB < mit areas (WS/WB > 0.30) and low average shape (Fig. 1B). slopes (average fl ank slopes <20°) (Figs. 2–4). Slope values are derived from the DEM, They are irregular and usually quite elliptical from which total, fl ank, and maximum interval 2.5 (Fig. 3). The smallest massif volumes are 5–6 Cones - CAVF 18° 3 average slopes are calculated, as well as aver- Cones - SCAVZ km , larger than the smallest cones and sub- age slopes as a function of height (Fig. 1B). Paired cones cones. The massif volume range is continu- 2.0 Sub-cones - CAVF 13° 3 A summit area is calculated as the area above Sub-cones - SCAVZ ous up to ~90 km ; fi ve larger massifs are then 3 which slopes strongly decrease (Fig. 1B). Massifs - CAVF 6° found with volumes >150 km (Fig. 2). These The average slope values between successive 1.5 Massifs - SCAVZ are the fi ve central Costa Rica volcanoes, which height intervals defi ne a profi le that, together are a particular case of huge massifs with more with height/width (H/WB) and summit width/ 1.0 shield-like shapes. Shape parameters of massifs Height (km) base width (WS/WB) ratios, summarize volcano do not vary systematically with size, except for profi le shape (Fig. 1B). a slight increase in irregularity (Figs. 3 and 4). 0.5 Shields CHARACTERIZATION OF VOLCANO DISCUSSION: THE EVOLUTION OF MORPHOMETRY 0 VOLCANO SHAPES 0 1 10 100 1000 The Central American Volcanic Front and the Volume (km3) The wide variety of volcano shapes and sizes southern Central Andes Volcanic Zone volca- probably represents different growth stages. As Figure 2. Height versus volume diagram the smallest volcanoes are all cones, a small showing different types of studied volca- 3 1GSA Data Repository item 2009151, Appendix nic edifi ces from Central American Volcanic (<1 km ) conical edifi ce can be considered a (morphometric parameter acquisition and descrip- Front (CAVF) and southern Central Andes morphometric starting point. From this simple, tion), maps of the Central American Volcanic Front, Volcanic Zone (SCAVZ). Curves correspond symmetrical, and smooth conical shape (e.g., and Table DR1 (location and morphometric details of to slopes of theoretical regular cones, which Izalco), a range of evolutionary trends of vol- all volcanoes used in this study), is available online approximately separate three main edifi ce at www.geosociety.org/pubs/ft2009.htm, or on request types. Straight line is threshold used to sep- cano growth can be recognized (Fig. 5). from [email protected] or Documents Secre- arate between small and large edifi ces in the The most easily recognizable trend is tary, GSA, P.O. Box 9140, Boulder, CO 80301, USA. text and in Figure 4. where conical shape is conserved with vol-

652 GEOLOGY, July 2009 Downloaded from geology.gsapubs.org on March 6, 2011 500 and they usually have a smooth conical profi le in one direction but are elongated ridges in the

100 opposite direction. Mid-sized sub-cones (e.g., , Lascar) have shapes similar to the ) 3 smaller sub-cones; they do not necessarily have more vents or a greater complexity, suggesting 10 that they evolve from mid-sized cones rather Volume (km than from smaller sub-cones. The cone–sub-cone–massif evolution is char- 1 acterized by volume increase with minor height

increase (H/WB decreases), enlargement of 0.3 0.00.050.100.150.200.25 0.0 0.1 0.2 0.3 0.4 0.5 1.0 1.5 2.0 2.5 3.0 1.0 1.2 1.4 1.6 1.8 20253035 1015 summit area (WS/WB increases), and increasing Height / width Summit width / base width Ellipticity index Irregularity index Average flank slope (°) complexity (ii and ei increase). Once mid-sized Figure 3. Volume versus several shape parameters showing variations of shape with size of massifs are formed (e.g., , El Hoyo) they different volcanic edifi ce types. Symbols as in Figure 2. can continue growing toward larger massifs with increasing complexity, producing a massif trend (Fig. 5). Larger massifs can also evolve 100 from mid-sized cones and sub-cones. Massifs 90 have many, generally aligned, vents; they tend

80 to form elongated ridges (e.g., Rincón de la

70 Vieja, Olca-Paruma), but can also form irregu- larly shaped clusters (e.g., Cerro Bayo). 60 The cone gap height interval coincides with 50 an interval of abundant sub-cones and mas-

Height (%) 40 Cones - small sifs (Fig. 2), while at greater heights the sub- 30 Cones - large cone gap occurs (Figs. 2 and 5). The cone gap Sub-cones - small 20 Sub-cones - large interval may refl ect a critical height range from 10 Massifs - small where two distinct evolutionary paths are pos- Massifs - large 0 sible; cones either continue growing upward and 1.0 1.5 2.0 2.5 3.0 1.0 1.1 1.2 1.3 1.4 1.5 1.6 5 10 15 20 25 30 become large cones, or they grow sideways and Ellipticity index Irregularity index Average slope (°) become large sub-cones and massifs, resulting Figure 4. Height versus plan (ellipticity and irregularity indexes) and profi le (slope) shape pa- in a scarcity of cones at this height range. rameters showing variation in average profi les of different volcanic edifi ce types and sizes. Which of these paths a cone takes may partly depend on the balance between magma pressure

(PM) and lithostatic pressure (PL), factors com- ume (Fig. 5). This simple evolution repre- Evolution from cones toward sub-cones and monly used to explain maximum edifi ce heights sents volcanoes that have one dominant vent massifs can occur before or at the cone gap (e.g., Eaton and Murata, 1960; Davison and De and that lack major structural complications. height interval (Fig. 5). Evolution from initial Silva, 2000). A pressure balance, P* = PM/PL This cone trend is continuous until the cone cones toward more complex shapes is supported can describe this effect: P* will tend to decrease gap. Even before reaching the cone gap com- by detailed studies of individual volcanoes with height (as PL increases) and summit erup- plexities do appear, but they do not alter sig- such as Lascar (Gardeweg et al., 1998) and tions will become increasingly less likely. Only nifi cantly the overall shape of the edifi ce: for Aucanquilcha (Klemetti and Grunder, 2008). those cones with high enough PM will be able example, El Tigre has a tectonic scarp cutting The smaller sub-cones (e.g., Conchaguita, to maintain a high P* and continue growing as its southern fl ank and thus has higher ii val- ) are elliptical and have large sum- large cones. Cones with lower P* will not be ues; Vallecitos has more than one vent and mit areas; they have more than one main vent able to erupt from their main vents, favoring thus is slightly elongated. and may evolve from cones by vent migration, shallow magma storage and the opening of new

Size Cones 3 3 3 3 1 km 5 km 10 km 50 km 2.0 Sub-cones Massifs Cone gap

Sub-cone gap 1.5 Cone trend P* and R* Cone gap

Cones P

1.0 yt ixelpmoc * and

s eno Height (km) Height R*

Massif trend c

-buS Sub-cone gap Cone trend

0.5 Shape

sf i

0 ss

0.1 1 10 100 1000 a Volume (km3 ) M Figure 5. Left: Height versus volume diagram showing fi elds of three main types of volcanic edifi ces and possible evolutionary trends. Right: Possible evolutionary growth paths of volcanoes starting from small simple cone. P* is pressure balance and R* is resistance balance

GEOLOGY, July 2009 653 Downloaded from geology.gsapubs.org on March 6, 2011 side vents. Such volcanoes will probably evolve tion and obtain a generalized model (Fig. 5). sensing: ASTER vs: SRTM: International Jour- toward sub-cones with increasingly complex We anticipate that this model will be applicable nal of Remote Sensing, v. 29, p. 6515–6538, doi: 10.1080/01431160802167949. shapes, larger summit areas, and more vents, to other volcanic settings. Klemetti, E.W., and Grunder, A.L., 2008, Volcanic evolution until eventually becoming massifs. From initial small cones several shape evo- of Volcán Aucanquilcha: A long-lived volcano in the Central Andes of northern : Bulletin of Another important factor is the balance lution trends are possible that depend on the Volcanology, v. 70, p. 633–650, doi: 10.1007/s00445- between conduit resistance (RC) and edifi ce prevailing processes, especially pressure and 007-0158-x. resistance (R ). We suggest a simple resistance resistance balances (P* and R*). If no tectonic Lacey, A., Ockendon, J.R., and Turcotte, D.L., 1981, On the E geometrical form of volcanoes: Earth and Planetary balance, R* = RE/RC, where if RC is low (e.g., complications arise, small cones grow until Science Letters, v. 54, p. 139–143. open magma-fi lled conduit), R* will be high reaching ~1200 m. Before reaching this height, Macdonald, G., 1972, Volcanoes: Englewood Cliffs, New and the cone will continue growing through its cones can evolve to sub-cones and eventually Jersey, Prentice-Hall, 510 p. Michon, L., and Saint-Ange, F., 2008, Morphology of Piton main conduit. In contrast, if R* is low, either massifs due to structural conditions or unusu- de la Fournaise basaltic shield volcano (La Réunion Island): Characterization and implication in the vol- because of a blocked conduit (high RC) or low ally low P*. At ~1200 m, cones reach a critical edifi ce resistance (low R ), then vent migration cano evolution: Journal of Geophysical Research, E height (low P* + R*) and most start growing v. 113, B03203, doi: 10.1029/2005JB004118. will dominate. RE will depend on cone material sideways by forming new vents, and evolve to Pike, R.J., and Clow, G.D., 1981, Revised classifi cation of ter- and the degree of fracturing and faulting, which sub-cones and massifs. Those with high enough restrial volcanoes and a catalog of topographic dimen- sions with new results on edifi ce volume: U.S. Geo- will be related to structural conditions. The cone P* and R* will continue growing as cones. logical Survey Open-File Report OF 81–1038, 40 p. gap may be a point where P* and R* reduce to a These larger cones will be prone to sector col- Plescia, J.B., 2004, Morphometric properties of Martian critical threshold. This threshold may be reached lapse and gravitational spreading, but they retain volcanoes: Journal of Geophysical Research, v. 109, E03003, doi: 10.1029/2002JE002031. earlier if RE is lowered by structural instabilities, their overall conical shapes. Rabus, B., Eineder, M., Roth, A., and Bamler, R., 2003, The favoring evolution toward small and medium- This study shows how volcano morphometry shuttle radar topography mission—A new class of digital elevation models acquired by spaceborne ra- sized sub-cones at heights below the cone gap. can be used to obtain information on processes dar: ISPRS Journal of Photogrammetry and Remote Large cones will be most prone to gravita- operating during volcano construction. It also Sensing, v. 57, p. 241–262, doi: 10.1016/S0924- tional spreading (e.g., Concepción) and sector contributes toward a more precise and quanti- 2716(02)00124-7. Riedel, C., Ernst, G.G.J., and Riley, M., 2003, Controls on the collapses (e.g., Ollagüe, ). Spreading tative classifi cation of volcanoes and a charac- growth and geometry of pyroclastic constructs: Jour- will slowly lower height and increase width, terization of shape evolution trends for arc vol- nal of Volcanology and Geothermal Research, v. 127, while sector collapse will rapidly reduce height canoes. Such a classifi cation, and its resultant p. 121–152, doi: 10.1016/S0377-0273(03)00196-3. Siebert, L., and Simkin, T., 2002, Volcanoes of the world: An and regularity. However, many edifi ces that interpretation of evolutionary trends, provides illustrated catalog of Holocene volcanoes and their have undergone these processes maintain their the framework for examining related structural, eruptions: Smithsonian Institution Global Volcanism Program Digital Information Series GVP-3: http:// conical shape, possibly because of growth after magmatic, and eruptive processes. www.volcano.si.edu/world/. (October 2008) or during these events (e.g., Ollagüe; Vezzoli et Simkin, T., and Siebert, L., 1994, Volcanoes of the world al., 2008); only cones that cease to be active for ACKNOWLEDGMENTS (second edition): Tucson, Arizona, Geoscience Press, 349 p. long periods will be signifi cantly modifi ed by Grosse is grateful to the Université Blaise Pascal and the CNRS (Centre National de la Recherche Sci- Smith, D.K., 1996, Comparison of shapes and sizes of sea- sector collapse or spreading, evolving toward entifi que, France) for funding a two-month stay. We fl oor volcanoes on Earth and “pancake” domes on Ve- sub-conical shapes (e.g., Maderas, ). thank CONICET (Consejo Nacional de Investigaciones nus: Journal of Volcanology and Geothermal Research, Científi cas y Técnicas), Fundación Miguel Lillo, and v. 73, p. 47–64, doi: 10.1016/0377-0273(96)00007-8. There are known geographical variations of Stoiber, R.E., and Carr, M.J., 1973, volcanic and Instituto CEDIAC (Capacitación Especial y Desarrollo edifi ce morphometry in the Central American tectonic segmentation of Central America: Bulletin of de la Ingeniería Asistida por Computadora) (Argentina) Volcanology, v. 37, p. 304–325. Volcanic Front (e.g., Stoiber and Carr, 1973; for their support, and B.D. Marsh, R.S.J. Sparks, and an Thouret, J.C., 1999, Volcanic geomorphology—An overview: Weyl, 1980). These can be often related to local anonymous reviewer for thoughtful reviews. Earth-Science Reviews, v. 47, p. 95–131, doi: 10.1016/ tectonics: for example, in Nicaragua, sub-cones S0012-8252(99)00014-8. van Wyk de Vries, B., 1993, Tectonics and magma evolution and massifs are located on zones, while REFERENCES CITED Carr, M.J., 1984, Symmetrical and segmented variation of of Nicaraguan volcanic systems [Ph.D. thesis]: Milton cones are on undisturbed crust (van Wyk de physical and geochemical characteristics of the Cen- Keynes, UK, Open University, 328 p. Vries, 1993; van Wyk de Vries et al., 2007). In tral American Volcanic Front: Journal of Volcanology van Wyk de Vries, B., Grosse, P., and Alvarado, G.E., 2007, and Geothermal Research, v. 20, p. 231–252, doi: Volcanism and volcanic landforms, in Bundschuh, addition, one of us (van Wyk de Vries, 1993) 10.1016/0377-0273(84)90041-6. J., and Alvarado, G.E., eds., Central America: Geol- showed that each morphological type of volcano Cotton, C.A., 1944, Volcanoes as landscape forms: Christ- ogy, resources and hazards, Volume 1: Netherlands, had different magma types and eruptive styles. church, Whitcombe and Tombs Publishing, 416 p. Balkema, p. 123–158. Cullen, A.B., McBirney, A.R., and Rogers, R.D., 1987, Struc- Vezzoli, L., Tibaldi, A., Renzulli, A., Menna, M., and Flude, S., Massifs may be complexes with shallow magma tural controls on the morphology of Galapagos shields: 2008, Faulting-assisted lateral collapses and infl uence storage, while cones develop predominantly Journal of Volcanology and Geothermal Research, on shallow magma feeding system at Ollagüe volcano (Central Volcanic Zone, Chile- Andes): Jour- deep magma chambers. These observations v. 34, p. 143–151, doi: 10.1016/0377-0273(87)90099-0. Davison, J., and De Silva, S., 2000, Composite volcanoes, nal of Volcanology and Geothermal Research, v. 171, show the potential for coupling tectonic, mag- in Sigurdsson, H., et al., eds., Encyclopedia of volca- p. 137–159, doi: 10.1016/j.jvolgeores.2007.11.015. matic, eruptive, and morphological phenomena noes: New York, Academic Press, p. 663–681. Weyl, R., 1980, Geology of Central America: Berlin, Born- Eaton, J.P., and Murata, K.J., 1960, How volcanoes grow: traeger, 371 p. into one unifi ed volcano evolutionary model. Science, v. 132, p. 925–938, doi: 10.1126/sci- Wood, C.A., 1978, Morphometric evolution of composite vol- ence.132.3432.925. canoes: Geophysical Research Letters, v. 5, p. 437–439, CONCLUSIONS Francis, P., 1993, Volcanoes: A planetary perspective: Ox- doi: 10.1029/GL005i006p00437. ford, Oxford University Press, 443 p. Wood, C.A., 1980, Morphometric evolution of cinder cones: Using morphometric parameters, volcano Gardeweg, M.C., Sparks, R.S.J., and Matthews, S.J., 1998, Journal of Volcanology and Geothermal Research, v. 7, morphology can be summarized and quanti- Evolution of Lascar Volcano, northern Chile: Geo- p. 387–413, doi: 10.1016/0377-0273(80)90040-2. fi ed. We fi nd that volcanoes can be grouped logical Society of London Journal, v. 155, p. 89–104. Wright, R., Garbeil, H., Baloga, S.M., and Mouginis-Mark, Hone, D.W.E., Mahony, S.H., Sparks, R.S.J., and Martin, P.J., 2006, An assessment of shuttle radar topography into distinct morphometric classes that suggest K.T., 2007, Cladistic analysis applied to the classifi - mission digital elevation data for studies of volcano distinct evolutionary trends. Despite different cation of volcanoes: Bulletin of Volcanology, v. 70, morphology: Remote Sensing of Environment, v. 105, p. 203–220, doi: 10.1007/s00445-007-0132-7. p. 41–53, doi: 10.1016/j.rse.2006.06.002. settings, the two studied arcs have volcanoes Jarvis, A., Reuter, H.I., Nelson, A., and Guevara, E., 2008, that are in similar morphometric classes. This Hole-fi lled SRTM for the globe, Version 4: CGIAR- Manuscript received 9 December 2008 suggests that volcano morphometry depends CSI SRTM 90m Database: http://srtm.csi.cgiar.org. Revised manuscript received 11 March 2009 (October 2008) Manuscript accepted 12 March 2009 on general processes. Hence, we can make Kervyn, M., Ernst, G.G.J., Goossens, R., and Jacobs, P., general statements about morphological evolu- 2008, Mapping volcano topography with remote Printed in USA

654 GEOLOGY, July 2009