Age and Evolution of the Upper Cenozoic Andesitic Volcanism in Central-South Chile

F^NOSœMUMZAGA j Apartment of Geology, Universidad de Chile, Santiago, Chile

ABSTRACT Vergara (1972b) established the presence of the University of Sao Paulo, Brazil, by of two units in the late Cenozoic volcanic Munizaga. The first ten potassium-argon determina- series, which form elongated and parallel Ages have been interpreted as being rock- tions made on andesitic rocks of the Andes belts between lat 37°30' S. and 42° S.: (a) the formation events. There is no evidence of and Coastal ranges of central-south Chile, Coastal volcanic belt, deeply eroded, crop- thermal events that might have caused argon between latitudes 36° and 42° South, dem- ping out parallel to the present coast line, loss and no indication of excess radiogenic onstrate the existence of at least two main and (b) the Andean volcanic belt, consti- argon. The analytic errors quoted on Table 1 cycles of volcanic activity; one in the tuted by the plateau andesitic series and the are relatively large, especially for the Miocene and the other in the Pliocene to recent volcanic centers. The Andean belt is younger samples, because of the high con- Holocene. The volcanic activity lines seem located to the east of the Coastal belt and tent of atmospheric argon associated with to have changed eastward from the present crops out in the higher Andes. All the their analyses. Coastal volcanic belt in Miocene timss to analyzed samples are andesite, and their lo- However, the results seem to be valid, due about 100 km inland in Pliocene and cations are shown in Figure 1. to the concordance of the ages with the local Holocene times. During late Pleistocene to stratigraphy. Holocene, a gradual inversion in the direc- METHOD tion of migration of volcanism occurred. The K-Ar dating techniques employed RESULTS AND DISCUSSION The geochronological asymmetry of the vol- were described in detail by Amaral and The K-Ar ages here presented demon- canic chains seems to be a highly important others (1966). Crushed and sized whole- strate the presence of at least two main cy- feature in this continental border and plate- rock samples were fused by induction heat- cles of volcanic activity, one during the contact zone. ing in ultrahigh vacuum systems employing Miocene, and the other during the Pliocene Cu-CuO and titanium for gas purification. to Holocene. INTRODUCTION The gas released was "spiked" for isotope The Miocene volcanic cycle is represented The first results of K-Ar determinations dilution with pure 38 Ar. Mass spectrometry by the eroded Coastal volcanic belt. The belt from late Cenozoic andesitic volcanic rocks was carried out by the static method on a is characterized by andesitic lava flows with of the Andean central-south Chile region are Reynolds-type mass spectrometer. Potas- two types of pyroxene, amphibole andesite, presented in this work. The chronostrati- sium determinations were made by flame and pyroclastic rocks of similar composition graphic problems of that volcanism are photometry using a Baird-Atomic research (Vergara, 1972b). These rocks crop out in analyzed. These results can be useful to cor- flame photometer with a lithium internal horizontal or gently dipping layers, or as relate Andean tectonic-volcanic events with standard. The chemical procedure was es- volcanic necks. In the coastal area of Pacific sea-floor spreading. sentially that described by Brannock and Temuco and Chiloe, these rocks interfinger The region outlined in Figure 1 has; been, Berthold (1949). with marine sediments of Miocene age (Gar- from Mesozoic to Holocene, a continental The constants for final calculations are cia, 1968). In most of the studied area, the border and a plate-contact zone (Vergara, rocks are continental and overlie a X total = 0.530 x 10"9yr-', 1972a). The generation of the Andean metamorphic basement of Paleozoic age X K= 0.585 x lO-'Oyr"1, Mountains seems to be strongly conditioned (Aguirre and others, 1972). The roof of the and by intense late Cenozoic volcanism. The vol- volcanic formation corresponds to the pres- atm %40K in K =0.119. canic products unconformably ovetlie the ent surface of erosion. Its thickness has been metamorphic basement and the Mesozoic The ages presented in Table 1 were ob- estimated as more than 200 m (Fig. 2). On and Cenozoic granitic intrusives; the study tained in total-rock samples, due to the fine Table 1 and Figure 1, samples 1, 2, and 3 of their evolution will allow definition of grain and the impossibility of separating correspond to andesite with two types of some fundamental concepts regarding the mineral phases. The determinations were pyroxene from volcanic necks; and samples genesis of the Andes. made at the Laboratory of Geochronology 4, 5, and 6 are andesitic lava flows interca-

Geological Society of America Bulletin, v. 85, p. 603-606, 2 figs., April 1974

603

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/85/4/603/3443657/i0016-7606-85-4-603.pdf by guest on 28 September 2021 I

Figure 1. Generalized map of upper Cenozoic volcanic outcrop, modified from Vergara (1972b, Fig. 1).

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/85/4/603/3443657/i0016-7606-85-4-603.pdf by guest on 28 September 2021 UPPER CENOZOIC ANDESITIC VOLCANISM IN CENTRAL-SOUTH CHILE 605

TABLE 1. ANALYTICAL DATA out at the Pino Hachado area, east of Temuco (Fig. 1), is mainly constituted by Sample Locality Rock type A 1,0 ccSTP ln_7 Ar*° Age Error no. rad gr x lu Atm % (m.y.) (m.y.) andesitic flows, amphibole trachyandesite, and, in less proportion, olivine basaltic 1 10 km S. of Los Aphiric andesite 2.24 18.8 69.0 20.4 ±1.0 Angeles, andesite. Vergara (1972b) has also de- 37°35' S., 72°17' W. scribed intercalated intermediate to acid 2 1 km NW. of Temuco, Orthopyroxene and 1.58 14.08 53.4 22.0 ±1.6 38°42' S., 72°37' W. clinopyroxene pyroclastic flows. K-Ar ages were obtained andesite on samples 7 and 8. Sample 7 was collected 3 9 km S. of Temuco, Aphiric andes¡te 1.160 13.49 32.5 27.7 ±1.9 38°50' S., 72°36' W. in the upper part of the series, and sample 8 4 Colegual no. 2, Orthopyroxene and 0.332 2.75 60.9 20.7 ±1.1 in the lower part. This series crops out on the NW. Puerto Montt, clinopyroxene summit of the Pino Hachado Pass. The ages 41° $., 73°15' W. andesite 5 NM. Ancud, Aphiric ándesete 0.704 11.40 67.1 40.4 ±1.8 obtained from these samples, 2.45 m.y. and 41°48' S., 73°45' W. 1.35 m.y., respectively, are between Pliocene 6 NE. Ancud, Obsidian (rhyolitic- 3.20 0.950 94.3 0.76 ±0.21 and early Pleistocene in age, according to the 41°48' S., 73°45' W. dacitic glass/ Berggren (1969) time scale. However, they 7 Pino Hachado Pass, Aphiric andes'te 3.07 2.98 90.40 2.45 ±1.00 E. Temuco, exhibit large experimental errors (see Table 38°40' S., 70°55' W. 1), and their results can be considered con- 8 3 km E. of Pino Amphibole andesite 2.88 1.55 96.2 1.35 ±0.70 Hachado, E. Temuco, cordant, with a mean value of 1.90 ± 1.00 38°40' S., 70°54' W. m.y. 9 Cerro Campanario, Pyroxene andeiite 1.237 0.971 94.3 1.97 ±0.60 Pehuenche Pass, In the Laguna del Maule area, southeast 36° S., 70°25' W. of Linares, an extensive and complex 10 Laguna del Maule, Pyroxene andesite 0.823 0.416 80.8 1.27 ±0.34 36°05' S., 70°30' W. andesitic plateau series crops out. The pet- rology and stratigraphy of the area are being Note: K% determination made in duplicate. studied at present by Munizaga. K-Ar dat- ings were made in two samples. Sample 9 lated with large sequences of pyroclastic local extent, deep glacial erosion, and in- corresponds to later lava flows from Cam- rocks of continental origin. The other re- tense block faulting (Gonzalez and Vergara, panario Hill, which is a strongly eroded vol- sults, not considering sample 6, suggest that 1962). The approximate boundaries of this canic center located on the Chilean- the volcanic cycle represented by this chain unit, for the Chilean side, are shown in Fig- Argentine border. Sample 10 corresponds to was largely developed within the Miocene. ure 1. In Argentina, the boundary extension an intrusive plug that cuts the andesitic This indication corroborates the age previ- is unknown. Thickness is variable from 800 plateau series in this region. The resulting ously inferred by Garcia (1968), based on to 100 m (Gonzalez and Vergara,1962; Ver- ages are 1.97 m.y. and 1.27 m.y., respec- stratigraphic relations. gara, 1972b). From the stratigraphy, struc- tively. This indicates an early to middle Pleis- Andesitic lava flows crop out in Chiloe ture, and morphology it was thought that tocene age for these volcanic centers and Island, where they are intercalated in some these deposits were Pliocene-Pleistocene or provides a minimum age for the andesitic places with fossiliferous marine sediments of Miocene in age and therefore could be re- plateau series in this area. Present volcanic Miocene age. A sample collected at the lated to the Coastal series (Thiele and Kat- centers are young structures, normally Ancud beach (Table 1, no. 6) corresponds sui, 1969). However, the present radiomet- superimposed on the andesitic plateau to a rhyolitic-dacitic glass, and its age ric data demonstrate a chronological discon- series. In the upper Pleistocene to Holocene (760,000 yr) seems to represent a last and tinuity between the two volcanic cycles. evolution of the central-type volcanism, a isolated volcanic event in this series. Sarr.ple The andesitic plateau series, which crops gradual migration of the volcanic activity 5, with an age of 40 m. y. (late Eocene), corresponds to an andesite belonging to a COASTAL VOLCANIC BELT ANDEAN VOLCANIC BELT thick sequence of continental origin located MAINLY MIOCENE MAINLY PLIOCENE TO HOLOCENE in the northern part of the island. The thickness of the sequence in which the dated sample occurs suggests the possibility of an intense Eocene volcanism occurring in the southern part of the volcanic chain. Evalua- tion of this Chiloe Island volcanism requires 2,000 m more geochronological determinations. 1,000 m More recent volcanism is concentrated in PAC I FIC OCEAN I the Andean volcanic belt, located approximately 100 km east of the Coastal belt (Eig. Ba \ 1). Here two volcanic units are distin- guished: (a) the andesitic plateau series, and (b) stratovolcanoes of central type, which !) are superimposed on top of the plateau A ACTIVE VOLCANOES

series. ANDESITIC PLATEAU The name "andesitic plateau series" is VOLCANIC ROCKS given to a large sequence of horizontal £.nd ^ MIGRATION OF VOLCANIC ACTIVITY subhorizontal andesitic lava flows and (GONZALEZ, 15 70) pyroclastic rocks of continental origin CHANGE OF VOLCANIC ACTIVITY which rest discordantly on the Mesozoic a nd lower Mesozoic basement. This series dis- Figure 2. Generalized and schematic cross section showing the de- Ba PALEOZOIC. MC SO ZOIC, ANO plays, in some places, a gentle folding of velopment of the continental margin from Miocene to Holocene time. LOW E ff CENOZOIC BASEMENT

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/85/4/603/3443657/i0016-7606-85-4-603.pdf by guest on 28 September 2021 606 VERGARA AND MUNIZAGA

toward the west has been observed (Gon- rological asymmetry is a characteristic of the Survey Bull. 992, p. 1-14. zalez, 1967, 1970). Active volcanoes are lo- oceanic volcanic trench areas (Dickinson, D ckinson, W., 1971, Plate tectonic models of cated on the western border of the Andean 1971), the geochronological asymmetry of geosynclines: Earth and Planetary Sci. Let- volcanic belt (Fig. 1). the volcanic chains seems to be a highly im- ters, v. 2, p. 165-174. F;rrar, F., Clark, A. H., Haynes, S. J., Quirt, G. S., Figure 2, which is a generalized and portant feature in this continental border Conn, H., and Zentilli, M., 1971, K-Ar evi- schematic cross section of the area studied, and plate-contact zone. dence for the post-Paleozoic migration of the exhibits different types of volcanic activity Eastward changes of the lines of volcanic granitic intrusions foci in the Andes of and the changing positions of the belts in activity from Miocene in the Coastal vol- northern Ch ile: Earth and Planetary Sci. Let- relation to time. From data collected, we canic belt to Pleistocene-Holocene in the ters, v. 10, p. 60-66. conclude that radiometric ages from the Andean volcanic belt demonstrate changes Garcia, F., 1968, Estratigrafía del Terciario de Coastal volcanic belt and the Andean vol- in the location of the magmatic foci which Chile central, in Bello, A., ed., El Terciario canic belt are different. These series corre- gave origin to these rocks. We infer a strong de Chile-zona central: Santiago, Chile, Soc. spond to different volcanic cycles during the eastward change cf this plane because these Geológica de Chile, v. 2, p. 25-57. G onzalez, O., 1967, Migration, structural control late Cenozoic. magmatic foci arc located in the Benioff and petrographic provinces of the upper A change of the volcanism to the east seismic plane. The gradual inversion of the Cenozoic volcanism in Chile: Santiago, seems evident. The Coastal volcanic belt is volcanism toward the west, in the upper Chile, Chilean Committee of the Upper mainly Miocene; the Andean volcanic belt, Pleistocene to Holocene, allows us to infer a Mantle Project Progress Rept. no. 1, p. 1—3. whose age is Pliocene-Holocene, is located permanent oscillation of the magmatic foci 1970, Post-Miocene volcanic petrographic approximately 100 km to the east of the that gave origin to this late Cenozoic vol- provinces of west Antarctica and their rela- Coastal belt (Fig. 2). During late Pleistocene canism. This must be related to change or tion with the southern Andes of South to Holocene, however, an inversion in the oscillations of the Benioff seismic zone and America: Oslo, SCAR/IUGS Symp. Antarc- direction of migration of volcanism oc- to change or oscillations of the subduction tic Geol. and Earth Solid Geophys., p. curred. The presently active volcanic cen- zone. 187-195. Gonzales, O., and Vergara, M., 1962, Recon- ters are located in the extreme west of the ocimiento geológico de la Cordillera de los Andean volcanic belt, proving a gradual ACKNOWLEDGMENTS Arides entre los paralelos 35° y 38° latitud migration to the west. We thank U. G. Cordani and K. sur: Chile Univ. Inst. Geologia Pub., no. 24, Eastward change of the active volcanic Kawashita of :he Geochronological p. 20-108. lines during late Cenozoic, is similar to the Laboratory of the University of Sao Paulo, Ferron, E. M., and Hayes, D. E., 1969, A change that occurred during the develop- Brazil, who provided facilities for radiomet- geophysical study of the Chile Ridge: Earth and Planetary Sci. Letters, v. 6, p. 77-83. ment and evolution of the Andean géosyn- ric measurements described in this work. P tman, W. C, III, Herron, E. M., and Heirtzler, J. clinal basin between the Lower and Upper Claudio Dos Santos helped in the laboratory R. , 1968, Magnetic anomalies in the Pacific Cretaceous in this same region (Vergara, work. L. Aguirre, U. G. Cordani, and K. and sea-floor spreading: Jour Geophys. Re- 1972a) Farrar and others ¡1971) showed, Palmer read and criticized the manuscript. search, v. 73, p. 2069-2085. based on K-Ar evidence, the migration with Partial financial support for this work was Sjgimura, A., 1968, Spatial relations of basaltic time of the Mesozoic granitic intrusions to provided by the Multinational Project in magmas in island arcs, in Basalt: New York, the east in the northern Andes of Chile. Geology sponsored by the Organization of Interscience: Pubs., Inc., v. 2, p. 537-571. Therefore, the migration of the igneous ac- American States and the Department of Thiele, R., and K.atsui, Y., 1969, Contribución al tivity lines seems to be a dynamic charac- Geology, University of Chile. conocimiento del volcanismo post- teristic of this continental border region, at Miocènico de los Andes en la provincia de least since Mesozoic time. REFERENCES CITED Santiago: Santiago, Chile, Univ. de Chile, Dept. Geologia, pub. no. 35, p. 1-23. Sea-floor spreading has been confirmed Aguirre, L., Herve, F., and Godoy, P., 1972, Dis- Vergara, M., 1972a, Note on the paleovolcanism along the Chilean coast by strong magnetic tribution of metamorphic fades in Chile— in the Andean geosyncline from the central and chronometric evidence (Pitman and An outline: Kristalinikum, v. 9, p. 7-19. part of Ch:.le: Internat. Geol. Cong.. 24th, others, 1968; Herron and Hayes, 1969). It Amaral, G., Cordan , U. G., Kawashita, K., and Montreal 1972, sec. 2, p. 222-230. has been also postulated that this region was Reynolds, J. H., 1966, Potassium-argon 1972b, Note on the zonation of the upper a plate-contact zone during the Mesozoic dates of basaltic rocks from southern Brazil: Cenozoic volcanism of the Andean area of (Vergara, 1972a). The characteristics of this Geochim. et Cosmochim. Acta, v. 30, p. central-south Chile and Argentina: Conf. on 159-189. border zone, with an oceanic trench, vol- Solid Earth Problems, Buenos Aires, Inter- Berggren, W. A., 1969, Rates of evolution in some canic chains, and petrological transversal nat. Upper Mantle Project, v. 2, p. 381-397. Cenozoic planktonic foraminifera: Mi- asymmetry during the Mesozoic and cropaleontology, v. 15, no. 3, p. 351-365. Cenozoic volcanism, confer to it a character Brannock, W., and Berthold, S., 1949, The deter- of "allied mountain arcs," according to minations of sodium and potassium in sili- MANUSCRIPT RECEIVED BY THE SOCIETY AUGUST Sugimura's classification (1968). Just as pet- cate rocks by flame photometer: U.S. Geol. 30,1973

P-inted in U.S.A.

Downloaded from http://pubs.geoscienceworld.org/gsa/gsabulletin/article-pdf/85/4/603/3443657/i0016-7606-85-4-603.pdf by guest on 28 September 2021