a 550-year-old plinian eruption at el chicho´n volcano ...raman/papers2/maciasecjgr.pdf · a...

A 550-year-old Plinian eruption at El Chichon Volcano, Chiapas,

Mexico: Explosive volcanism linked to reheating of the magma

reservoir

J. L. Macıas, J. L. Arce, J. C. Mora, J. M. Espındola, and R. Saucedo1

Instituto de Geofısica, Universidad Nacional Autonoma de Mexico, Mexico City, Mexico

P. ManettiDipartimento di Scienze della Terra, Universita degli Studi di Firenze, Florence, Italy

Received 23 April 2003; revised 14 August 2003; accepted 27 August 2003; published 23 December 2003.

[1] Some 550 years ago (1320–1433 A.D.), a powerful Plinian eruption at El ChichonVolcano in southern Mexico produced a widespread pumice fall deposit. We subdividedthe deposit into three parts on the basis of structural and textural characteristics,pumice lithology and density, granulometry, and petrologic-geochemical attributes. Thedeposit covers an area of 1500 km2 within the 1-cm isopach and has a minimumestimated bulk volume of 2.8 km3 (1.1 km3 dense rock equivalent (DRE)); its eruptivecolumn reached an altitude of �31 km. Consideration of field evidence, the presence andnature of mafic enclaves, and chemical data strongly suggest that the 550 year B.P.eruption is linked with the intrusion of a high-temperature basaltic magma intopreexisting but stagnated trachyandesitic magma beneath El Chichon. Thorough mixing ofthe two magmas produced a compositionally uniform hybrid trachyandesite magma(average SiO2 55.3 wt %), which subsequently underwent crystal growth and gasexsolution, ultimately overpressurizing the zoned magmatic system to erupt explosively.On the basis of El Chichon’s known eruptive history, the intrusion-mixing eventoccurred sometime after the 900 year B.P. eruption. The hybrid magma had apreeruption temperature of 820–830�C and was water undersaturated (5–6 wt % H2O)at pressures of �2–2.5 kbar. INDEX TERMS: 3620 Mineralogy and Petrology: Crystal chemistry;

8404 Volcanology: Ash deposits; 8414 Volcanology: Eruption mechanisms; 8434 Volcanology: Magma

migration; 8499 Volcanology: General or miscellaneous; KEYWORDS: Holocene, Plinian eruption, magma

mixing, El Chichon Volcano, Mexico

Citation: Macıas, J. L., J. L. Arce, J. C. Mora, J. M. Espındola, R. Saucedo, and P. Manetti, A 550-year-old Plinian eruption at El

Chichon Volcano, Chiapas, Mexico: Explosive volcanism linked to reheating of the magma reservoir, J. Geophys. Res., 108(B12),

2569, doi:10.1029/2003JB002551, 2003.

1. Introduction

[2] Plinian eruptions constitute phenomena of funda-mental importance in volcanology that have immediateand far-reaching consequences on the environment. Dur-ing the last third of the past century, the possibility ofusing modern methods of observation during Plinianeruptions led to a new understanding of the immediateand external characteristics of such events. What we knowabout the activity that takes place in the magmaticreservoirs and finally leads to a Plinian eruption is,however, still limited. To contribute to new knowledgein this subject, it is important to procure the informationcontained in the deposits yielded by ancient eruptions,

mostly from volcanoes that have shown frequent episodesof explosive activity. To this class of volcanoes belongs ElChichon volcano, widely known for its violent eruptionsof March–April 1982, which destroyed nine villages andkilled about 2000 people [Espındola et al., 2000;Sigurdsson et al., 1984]. The 1982 eruption of El Chichondeposited tephra as far away as the Yucatan Peninsula[Varekamp et al., 1984], and injected sulfur laden aerosolsinto the stratosphere [Devine et al., 1984; Matson, 1984;Rye et al., 1984], which circumnavigated the Earth severaltimes and reduced surface global temperatures by �0.5�C[Parker and Brownscombe, 1983]. The global influence ofthe 1982 El Chichon eruption gained enormous interestwhen scientist found sulfur-rich spikes in Greenland Icecores that were correlated with the volcano’s past activity[Zielinski, 1995; Zielinski et al., 1997]. The consequencesof the 1982 eruption and the inferred characteristics ofpast eruptions, suggest a large influence of El Chichoneruptions on the ancient Maya culture [Espındola et al.,

JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 108, NO. B12, 2569, doi:10.1029/2003JB002551, 2003

1Now at Instituto de Geologıa, Universidad Autonoma de San LuisPotosı, San Luis Potosi, Mexico.

Copyright 2003 by the American Geophysical Union.0148-0227/03/2003JB002551$09.00

ECV 3 - 1

2000; Ford and Rose, 1995; Gill, 2000], and theirpossible impact on the Classic Maya Collapse [Espındolaet al., 2000; Gill and Keating, 2002]. From a geologicpoint of view, the presence of anhydrite reported for thefirst time in juvenile pumice of the 1982 products beganan original discussion among scientists regarding thenature of the mineral [Carroll and Rutherford, 1987;Luhr, 1990; Luhr et al., 1984; McGee et al., 1987; Roseet al., 1984].[3] The 1982 eruption of El Chichon by stripping the

vegetation cover around the volcano exposed old pyroclas-tic deposits to further erosion and to investigation byvolcanologists. Several studies prompted by this eventshowed that the eruptive episodes in the volcano wererather frequent, with at least 11 eruptions during the last8000 years [Espındola et al., 2000; Tilling et al., 1984].Studies of the pre-1982 deposits have yielded severalradiometric ages and fragmentary stratigraphic data; yetthe scarcity of outcrops precluded the detailed reconstruc-tion of the parent events. A notable exception is unit B, apumice fall deposit produced some 550 years ago by apowerful explosive eruption. Tilling et al. [1984] brieflydescribed this deposit for the first time and provided sixradiometric ages from carbonized wood contained in thedeposit; the ages spanned the range 550 ± 60 to 700 ± 70years B.P. Espındola et al. [2000] reported similar agesfrom their collected samples, described the deposit inmore detail, and constructed approximate isopachs. Theypointed out that this eruption was probably greater thanthe one in 1982. In this paper, we present a detailedstratigraphic, depositional, and petrological study of theunit, as well as the reconstruction of the Plinian eruption.We find that the volume of this deposit is in fact an orderof magnitude larger than the volume of 1982 falloutlayers. In addition, through the study of the internalsubdivision of unit B, whole rock chemical analyses,and mineral chemistry of the pumice, we could constrainthe preeruptive conditions in the magma reservoir. Finally,we propose a model for the triggering mechanism of the550 year B.P. eruption in light of recent studies of similarvolcanic eruptions.

2. Previous Work

[4] El Chichon Volcano (17�220N, 93�140W; 1100 mabove sea level), the youngest member of the Chiapa-necan Volcanic Arc [Damon and Montesinos, 1978](Figure 1a), is located at the northwestern margin ofthe State of Chiapas, Mexico. The volcano consists of alarge somma crater with gently dipping outward slopes;the protruding rim of the 1982 crater rises in the centerof this crater. This crater is 1 km in diameter, has anaverage depth of 140 m, and contains a central lake(Figure 1b). El Chichon lies within an area of foldedLate Jurassic-Early Cretaceous marine sedimentary rocks,covered by Tertiary terrigenous formations [Canul et al.,1983], all affected by left-lateral east-west trending faults[Macıas et al., 1997].[5] Canul and Rocha [1981] were the first to report

past volcanic activity at El Chichon Volcano. Theydescribed old pyroclastic deposits containing carbonizedwood; however, they did not report on the age of these

materials. Several studies after the eruption reported olderdeposits and ages. Duffield et al. [1984] reported the firstradiocarbon date of �1250 years B.P. for a widespreadpyroclastic flow deposit around the volcano. Rose et al.[1984] used this date to present a tentative correlation ofvolcanic units between the crater and the surrounding apron.Tilling et al. [1984] presented the first composite section ofthe volcanic history of El Chichon, which included sixstratigraphic units and 12 dates obtained either from char-coal or from paleosoils. In their composite section, theylabeled the units from A (the youngest, i.e., the 1982deposits) to F (the oldest, corresponding to the �210 kasomma crater); the 550-year-old pumice air fall wasdesignated as unit B.[6] Macıas [1994] found two additional volcanic events

yielding radiocarbon ages 1400 and 900 years B.P. Morerecently, Espındola et al. [2000] reported new data andconstructed a revised composite stratigraphic column, con-cluding that the volcano has erupted at least 11 times duringthe past 8000 years. They also presented a preliminary studyof the 550 year B.P. fall unit B deposit, finding that the10-cm isopach covers approximately an area of 350 km2

with an estimated volume of 0.42 km3 and a total mass of40 � 1010 kg. They also estimated that the parent Pliniancolumn reached a height of �20 km. The deposit wasinitially described as a single fall unit [Tilling et al.,1984]. Espındola et al. [2000] recognized, however, thatat one section the unit was associated with a gray ash flowdeposit, which contains disseminated charcoal from whichthey obtained a 14C age of 550 ± 60 years B.P. (1320–1433 A.D., at 1s ± calibration).

3. Stratigraphy of the Deposit

[7] We studied the structural and textural characteristicsof unit B at 119 sections around the volcano (Figure 2).Unit B overlies Tertiary clay stones in the SE and NWportions of the volcano. At several other localities, unit Boverlies a thick dark brown silty paleosoil with abundantpottery shards developed on top of a gray, multilayeredpyroclastic flow deposit that was dated at 1200 years B.P.[Tilling et al., 1984] and unit D [Espındola et al., 2000].[8] Unit B is a yellow, pumice-rich, clast-supported fall

deposit (Figure 3). It has a maximum thickness of 110 cmsome 3 km northeast from the volcano (site 17, Figure 2).The deposit consists almost entirely (95 vol %) of yellowpumice (whitish on fresh surfaces) with rare gray andbanded pumice, crystals, gray dense andesite clasts, claystone fragments, amphibole cumulates (<2 cm in diame-ter), and dark gray rounded mafic enclaves (<5 cm indiameter). Both cumulates and mafic enclaves appeareither as fragments embedded in pumice or as loose debrisin the deposit.[9] From bottom to top unit B is reverse-to-normal

graded with the largest fragments in the middle of thedeposit (Figure 4a). Using this internal size variation,we divided the deposit into three parts, bottom (B1),middle (B2), and top (B3). The upper part of B3 isusually eroded, with the most complete exposures being5 cm thick and composed of well-sorted, coarse ashparticles.

ECV 3 - 2 MACIAS ET AL.: PLINIAN ERUPTION, CHICHON, MEXICO

[10] The yellow pumice is porphyritic, containing�12 vol % phenocrysts of plagioclase, hornblende, clinopy-roxene, titanomagnetite, rare illmenite, apatite, and spheneset in vesicular glass (Table 1). The gray pumices are also

porphyritic (�12 vol % phenocrysts) but with larger propor-tions of hornblende compared to the yellow pumices. Thedensities of the yellow pumices vary from 0.6 to 0.9 g/cm3

(vesicularity of 63–74 vol %), while the gray pumices are

Figure 1. (a) Location of El Chichon Volcano in southern Mexico. The dotted line is the 1-mm isopachof the 1982 fall A1 eruption. TVC, Los Tuxtlas Volcanic Field; CVA, Chiapanecan Volcanic Arc; CAVA,Central American Volcanic Arc; MAT, Middle American Trench. Triangles are volcanic structures.(b) View to the north showing the lake occupying the 1-km-wide crater produced by the 1982 eruption ofEl Chichon ( photograph by J. Pacheco, April 2001).

MACIAS ET AL.: PLINIAN ERUPTION, CHICHON, MEXICO ECV 3 - 3

denser varying from 1.0 to 1.2 g/cm3 (vesicularity of 51–58 vol %).

4. Granulometry

[11] Fifty samples belonging to B1 (10), B2 (22), and B3(18) were sieved in 1f spaced intervals (from �5f (32 mm)to 4f (0.0625 mm)). In most of the sites, we collected thesamples from the three parts of the deposit. Representativeresults are plotted in histograms and cumulative curves ofFigures 4b and 4c.

[12] At site 4, we carried out a complete vertical samplingthrough unit B (Figure 4a). Here, B1 has a bimodaldistribution with modes at �3 and 1f and a mediandiameter of �2.2f and is poorly sorted. B2 is bimodal withmodes at �4 and 1f, has a median diameter of �1.7f, andis poorly sorted. At other sites such as 176, B2 is unimodaland typical of pyroclastic fall deposits [Walker, 1971]. B3 isunimodal and has a mode of 1f, a median diameter of�0.5f, but is poorly sorted (Figures 4a and 4b). At somesites a second mode appears in B3 histograms, which often

Figure 2. Location of El Chichon Volcano showing main villages, towns destroyed by the 1982eruption, roads, and major rivers. Dots represent 119 stratigraphic sections studied in this work, andnumbers in boxes are selected sections mentioned in text.


lies either at �4f or at �3f (site 176, Figure 4b). Thebimodal distribution of the curve is likely related to strongwinds acting during the eruption. Cumulative curves for B1,B2, and B3 at sites 4 and 120 exhibit a significant decreasein grain size with distance from the vent as can be observedin Figure 4c.

5. Areal Distribution and Volume

[13] The deposit of unit B forms a blanket around thevolcano (Figure 5a). The best exposures of unit B are in theoutskirts of Volcan village, Chapultenango, and CarmenTonapac, at distances between 5 and 10 km from the source.Beyond 13 km, most of the deposit has been removed byerosion. Most exposures thinner than 10 cm are reworkedwith individual pumice fragments incorporated intoimmature soils or silty layers. Nonetheless, we were ableto construct reasonably well constrained isopach maps(Figure 5a).[14] Isopachs of layer B show two main dispersal axes.

Close to the crater (<4 km from the source) the dispersalwas almost to the east; however beyond 4 km the dispersalappears to shift to N30�E. This latter orientation coincideswith the dispersal axes of 1982 eruptive units [Carey andSparks, 1986; Sigurdsson et al., 1984].[15] The 20-cm isopach covers a minimum area of

240 km2 including the towns of Chapultenango, GuadalupeVictoria, Carmen Tonapac, Vicente Guerrero, Xochimilco,and other small settlements. It is worth noticing that thoselocalities received only �2 cm of tephra fall from the 1982eruption [Sigurdsson et al., 1984]. By using the extrapolatedLn thickness versus the square root of the area (A1/2)enclosed by the isopachs [Pyle, 1989], we obtained amaximum thickness of 118 cm for the deposit, which agreeswell with the 110 cm measured in the field. In addition, we

Figure 3. Composite stratigraphic column for El Chichonvolcano from 1200 years B.P. to present. Unit B overlieseither a white pumice flow deposit (unit C of Espındola etal. [2000]) or a light gray ash flow deposit that containspottery (unit D of Espındola et al. [2000] and Tilling et al.[1984]). Unit B underlies the soil and the 1982 deposits(unit A).

Figure 4. (a) Photo of unit B at site 4 that shows a remarkable difference in pumice grain size, the basisfor the designation of three different parts (B1, B2, and B3). The hammer used for scale is 30 cm long.(b) Histograms of the granulometric distributions of B1, B2, and B3. Notice the weakly bimodaldistribution of the two lower layers. (c) Cumulative frequency curves of B1, B2, and B3 at 4.5 and 12 kmfrom the source.


Table 1. Representative Whole Rock Composition of Unit B Sample Major and Trace Elementsa

Sample

B1 B2 B3 BG CHI9550 1A CHI99SAT CHI0101

Rock type pumice pumice pumice gray pumice pumice pumice mafic enclave mafic enclave

Major Elements, wt %SiO2 55.23 55.40 55.14 55.70 55.18 54.40 49.12 46.71TiO2 0.67 0.67 0.66 0.70 0.72 0.69 1.94 1.14Al2O3 18.84 18.94 19.15 18.49 18.60 18.50 13.43 13.82Fe2O3

b 6.65 6.71 6.67 6.76 6.93 6.81 9.93 12.07MnO 0.19 0.22 0.18 0.18 0.18 0.18 0.17 0.40MgO 2.26 2.29 2.21 2.42 2.36 2.38 9.42 7.48CaO 7.21 7.24 7.20 7.52 7.22 7.41 10.69 12.06Na2O 4.06 4.05 4.23 4.16 3.75 3.94 2.55 2.90K2O 2.21 2.21 2.31 2.51 2.40 2.44 1.94 1.31P2O5 0.34 0.33 0.34 0.36 0.29 0.37 0.31 0.31LOI 2.40 2.38 2.20 1.24 1.75 2.39 0.28 0.67Total 100.1 100.4 100.3 100.20 99.4 99.5 99.8 99.9

Minor Elements, ppmBa 756 760 783 754 593 720 582 417Rb 80 71 76 61 nd nd 50 35Sr 1012 1014 1028 1053 1433 1000 695 695Ta 1.7 1.7 1.5 1.0 nd nd 1 ndNb 14.1 14.2 7.9 nd 14 8 nd ndHf 4.1 4.4 4.1 3.9 nd nd 2.2 4.5Zr 124 114 102 121 nd nd 107 98Y 21 21 21 21 7 18 18 23Th 9.6 9.5 9.5 7.8 20.0 6.0 5.1 2U 3.0 3.2 3.2 2.8 nd nd 1.8 0.8Cr 5.5 7.4 6.9 10.1 21.0 7.0 516 374Ni 2 2 3 6 21 9 124 74Sc 10.4 10.5 9.9 10.9 nd 9.0 31.1 21.7V 180 181 182 169 164 190 264 246Cu 21 26 14 58 nd 34 39 42Pb 9 14 10 13 7 8 7 ndZn 91 89 76 72 62 90 93 118La 29.8 31.6 30.1 33.6 nd 29.0 23.6 24.5Ce 54.0 56.0 55.0 58.0 nd 54.0 41 55Nd 27.0 27.0 27.0 28.0 nd 31.0 22 28Sm 5.29 5.60 5.19 5.4 nd nd 4.69 5.66Eu 1.55 1.60 1.43 1.4 nd nd 1.36 1.62Tb 0.70 0.80 0.70 0.6 nd nd 0.6 0.7Yb 2.31 2.29 2.23 2.1 nd nd 1.69 2.63Lu 0.33 0.33 0.32 0.3 nd nd 0.25 0.3887Sr/86Sr 0.70409 0.70448143Nd/144Nd 0.51280 0.51279

Modes, vol %Plag ph 6.1 11.8 9.0 10.6Plag mph 2.3 1.3 2.9 –Cpx ph 0.8 1.0 1.5 1.3Cpx mph 1.5 – 1.2 –Amp ph 1.8 1.0 2.3 7.6Amp mph 0.7 – 0.9 –Oxi ph – 0.7 – 1.0Oxi mph 2.4 2.4 0.8 –Vs 59.6 53.9 55.9 43.0Gmass 25.0 28.0 25.3 36.3Total 100 100 100 100

aMajor and trace elements analyzed by inductively coupled plasma mass spectrometry (ICP-MS) and instrumental neutron activation analysis (INAA)(<0.01% major elements; Ba, 50 ppm; Cr, Pb, Nb, V, and Rb, 2 ppm; Ni, Sc, Sr, Y, and Zr, 1 ppm; Cu, Zn and Ta, 0.5 ppm; Hf and Th, 0.2 ppm; U, 0.1 ppm;La, Ce, Nd, Sm, Tb and Yb, 0.1 ppm; Eu, 0.05 ppm, detection limits) at Activation Laboratories, Ancaster, Canada. B1, B2, B3, BG (unit B), CHI99SAT,and CHI0101 (mafic enclaves) are samples analyzed in this study. CHI9550 (unit B) is from Espındola et al. [2000]; 1A (unit B) is from McGee et al.[1987]. Isotopic analyses performed on a Finnigan MAT 262 Mass Spectrometer, LUGIS, Instituto de Geofısica, UNAM. Measured values for Sr and Nd,laboratory standards are SRM987 = 0.710233 ± 17 and for La Jolla, 0.511881 ± 22, where the last two decimals units are n, number of analyses. Analyseswere performed by T. Trevino, G. Solıs, J. Morales, and M. Hernandez. The modal analysis was carried out by counting 800–1000 points. Abbreviationsare Plag, plagioclase; Cpx, clinopyroxene; Amp, amphibole; Oxi, Fe-Ti oxides; Vs, vesicles; ph, phenocrysts; mph, microphenocrysts; Gmass, groundmass;nd, not determined.

bFe2O3 is reported as total iron.


Figure 5. (a) Isopach map of unit B constructed from 85 selected stratigraphic sections (dots). Contourlines are every 10 cm. The dashed line represents the 3-cm isopach of the A1 fall deposit of the 1982eruption near the town of Chapultenango, where unit B is 20 cm thick. (b) Isopleth map of unit Bconstructed from the average of the five largest lithics measured at each section. Contour lines are incentimeters. Dotted lines represent inferred isopachs and isopleths.


calculated an extrapolated area of 1475 km2 for the 1-cmisopach.[16] The volume of the unit B deposit, determined using

the method of Fierstein and Nathenson [1992] as modifiedby Carey et al. [1995], is 2.8 km3, which corresponds to

1.1 km3 DRE, assuming an average pumice density of0.96 g/cm3 and a magma density of 2.5 g/cm3. It thusappears that the eruption of unit B produced an equivalentamount of material as the entire 1982 eruption, whichconsisted of three Plinian events accompanied by pyroclastic

Figure 6. (a) Modified maximum downwind range versus crosswind range graph [Carey and Sparks,1986], showing that the 550 year B.P. eruption of El Chichon Volcano occurred under strong windconditions similarly to the Plinian phases of the 1982 eruption. (b) Modified mass eruption rate versustotal column height diagram for tropical latitudes (method of Sparks [1986]). Assuming an averagetemperature of 800�C for the unit B magma, and a column height of 31 km, a mass discharge rate of108 kg/s is obtained.


flows and surges. Using an average density of 0.96 g/cm3

for the deposit, and a total volume of 28 � 108 m3, a totalmass of 1.05 � 1012 kg was ejected during the eruption.This figure is less than the total mass of the 1982 eruption of2.9 � 1012 kg as recalculated by Luhr and Logan [2002].

6. Column Height and Mass Eruption Rate

[17] The theoretical height of the eruptive column canbe calculated from an isopleth lithic map, using the

method of Carey and Sparks [1986]. At each outcropwe measured the major axis of the five largest lithics andthe average value was used to construct the isopleth mapof Figure 5b. The distribution is elongated with an axisoriented approximately 80� to the east. Assuming anaverage density of 2.5 g/cm3 for the lithic fragments, weobtained a column height of 31 km and a wind speed of25 m/s, which is similar to the measured winds during thephreato-Plinian phase of the 1982 eruption of 26 m/s[Carey and Sigurdsson, 1986], and those inferred for the

Figure 7. Classification diagrams of unit B rocks. (a) In the TAS diagram [Le Bas et al., 1986] themafic enclaves fall in the trachybasalt and basalt fields, while the pumice and glass of unit B aretrachyandesitic and rhyolitic, respectively. (b) Samples fall in the high-K field of calc-alkaline series[Gill, 1981]. (c–g) Harker diagrams of selected major elements.


1902 Santa Marıa eruption [Williams and Self, 1983] of17 m/s [Carey and Sparks, 1986] (Figure 6a).[18] We estimated mass eruption rate (MER) from col-

umn height and eruption temperature, using the analysis ofSparks [1986] (Figure 6b). Assuming an eruption temper-ature of 800�C (see section 7.2.4), a column height of 31 kmcorresponds to a mass eruption rate of 108 kg/s. Carey andSigurdsson [1989] deduced from the study of 45 Plinianeruptions worldwide that this eruption rate would produceonly small-volume pyroclastic flows or none at all. Thisagrees with our observations of the deposits associated tounit B.

7. Geochemistry

7.1. Whole Rock Composition

[19] Unit B pumice samples are trachyandesitic in com-position (Table 1) and plot in the high-K field for calc-alkaline rocks [Gill, 1981] (Figure 7b). No variations werefound between yellow (55.1–55.4 wt % SiO2) and graypumice (55.7 wt % SiO2), for which the color variations aredue to the different proportions of hornblende in the samplesand their vesicularity. Hereafter, we compare the composi-tion of unit B with 1982 eruptive products [McGee et al.,1987; Rose et al., 1984; Sigurdsson et al., 1984] and witholder products of the volcano [Espındola et al., 2000]. Theproducts of both unit B and other eruptions of El Chichonhave been fairly constant over the past 1250 years. Unit Bpumices have isotopic compositions of 87Sr/86Sr = 0.70409,and 143Nd/144Nd = 0.51280 (Table 1 and Figure 8).[20] Mafic enclaves found within unit B appear as loose

subrounded fragments, similar to those found in the Guaya-bal Tuff Cone [Macıas, 1994], an older volcanic structurelocated southeast of the 1982 crater (Figure 2). These latterenclaves have pillow-like, crenulated, and drop-like formswith chilled margins (Figure 9) and are associated withsandstone and limestone xenoliths, all hosted in trachy-andesite dome rocks [Espındola et al., 2000]. Unit Benclave (CHI0101), and the enclave from the GuayabalTuff Cone (CHI99SAT) have basaltic compositions. The

basaltic enclave has isotopic ratios of 87Sr/86Sr = 0.70448and 143Nd/144Nd = 0.51279 (Table 1 and Figure 8).

7.2. Mineralogy

[21] Modal analyses of pumice samples were obtained bycounting 748–1200 points. Each crystal was classifiedaccording to its size as ph, phenocryst >0.3 mm; mph,microphenocryst >0.03 mm and <0.3 mm; and Gmass,groundmass including microlites and glass. Microprobeanalyses in individual crystals were carried out from core(c) to interior (i) to rim (r).7.2.1. Plagioclase[22] Plagioclase is present as euhedral to anhedral pheno-

crysts (<4 mm) and microphenocrysts in proportions of6–12 vol % (Figures 10a–10c). Normal, reverse, or oscil-latory zonings occur in the phenocrysts (Figure 10d), withcompositions varying between andesine and labradorite,and rarely bytownite (Table 2 and Figure 11).[23] In B1, plagioclase is normally zoned from An46–65

in the cores to An35–40 in the rims. Microphenocrysts are

Figure 8. 87Sr/86Sr vs. 143Nd/144Nd diagram of ElChichon samples. The mineral data are for the 1982eruption [Tepley et al., 2000]. Notice that the mafic enclavehas a slightly higher Sr isotopic ratio than the unit BChichon sample.

Figure 9. Mafic enclaves in trachyandesitic dome lavas atthe head of the Agua Tibia valley. (a) Rounded enclave(center left), a crenulated enclave (center), and a smalltriangle-shaped enclave with sharp boundaries (blackarrow). (b) Rounded enclave rimmed by an assimilationaureole, both containing abundant plagioclase phenocrysts.The marker for scale is 14 cm.


unzoned at An55–58. The opposite ranges are found in B2plagioclase, in which phenocrysts exhibit slightly normaland reverse zoning (An41c–47r) while microphenocryst coreshave a wide spectrum of composition (An37–55) (Figure 11).Plagioclase in B3 is reverse and normally zoned, fromAn40–62 in the cores to An45–64 in the rims. Micropheno-crysts are constant in composition at An45.[24] Some microlites of plagioclase were found as inclu-

sions within hornblende phenocrysts, and have compositionsof An40–48. In contrast, discrete microlites in the ground-mass are anorthite-rich, with composition of An60–75.7.2.2. Hornblende[25] The most abundant mafic mineral is hornblende,

occurring mainly as phenocrysts (<3 mm) in concentrationsof�2 vol % (Figure 10c). Hornblende is relatively uniform incomposition, with little variation between crystals (Table 3).Overall, hornblende falls in the field of magnesian hasting-site, but two crystals have a tschermakitic composition.7.2.3. Pyroxene[26] Anhedral to subhedral clinopyroxene is present as

phenocrysts (<3 mm, 2 vol %, Figure 10c), micropheno-crysts, and rare microlites. Both normal and sector zoningare found (Table 4 and Figure 12). Some pyroxenes appearbroken and partially resorbed. All analyzed crystals arediopsidic, with En36–46 cores, En37 rims, and En34–46

microphenocrysts (Figure 12). Samples from B2 and B3have fewer diopside phenocrysts compared to B1 (Table 4and Figure 10c).7.2.4. Fe-Ti Oxides and Geothermometry[27] Anhedral titanomagnetite microphenocrysts are dis-

seminated in the groundmass and occur as inclusionswithin plagioclase, hornblende, and pyroxene phenocrysts(Figure 10c and Table 5). Ilmenite crystals were foundonly in B2.[28] Temperatures and oxygen fugacity were calculated

from coexisting ilmenite (60–66% X0Ilm) and titanomag-netite (16–19% X0

Usp) in sample (B2), following themethod of Andersen and Lindsley [1988] and the mineralrecalculation model of Stormer [1983]. The Fe-Ti oxidesmet the Mg-Mn equilibrium conditions of Bacon andHirshmann [1988]. Oxide pairs yield temperatures of820 to 830�C and oxygen fugacities in the range of�11.08 to �11.02. At those temperatures the oxygenfugacities are about 2.2 and 2.4 log units above theNNO buffer. The temperature agrees with the calculatedvalues for the 1982 magma: 750–850�C [Rye et al., 1984],785�C ± 23 [Luhr et al., 1984], and 800�C [Luhr, 1990].7.2.5. Groundmass and Water Contents[29] The groundmass is composed mainly of light brown

glass, vesicles, and plagioclase microlites. The glass is

Figure 10. Mineral assemblage, vesicle types and phenocryst textural characteristics of unit B pumice.Ph, phenocrysts; Mph, microphenocrysts; Ves, vesicles; plg, plagioclase; px, pyroxene; amph, amphibole;oxi, Fe-Ti oxides; and Gmass, groundmass. (a) Photomicrograph of pumice B1, plain polarized light.(b) Photomicrograph of pumice B1, crossed nicols. (c) Modal analysis of B1, B2, and B3 samples.(d) Photomicrograph showing a zoned plagioclase phenocryst in B1, crossed nicols.


rhyolitic in composition, with 6–10 wt % alkalies and66.6–69.6 wt % SiO2 (Table 6) and plots in the K-richfield of the calc-alkaline series (Figure 7). Plagioclasemicrolites are compositionally equivalent to phenocrystrims and microphenocrysts. Using the method of Houshand Luhr [1991], we estimate a preeruptive water concen-tration in the melt of 820–830�C of 5–6 wt %.7.2.6. Apatite and Sphene[30] Euhedral apatite crystals (<0.03 mm) are present in

plagioclase phenocrysts and as rare microphenocrysts in thegroundmass. Apatites are rich in F (2.43–2.22 wt %), with0.53–0.61 wt % Cl and 0.33 wt % SO3 (Table 7), similarabundances to those reported for apatites in the 1982magma [Luhr et al., 1984]. Sphene is present as discretemicrophenocrysts (<1 mm) dispersed in the groundmass ofall samples. The analyzed sphenes have compositionssimilar to those reported for the 1982 pumice [Luhr et al.,1984] but may be slightly enriched in CaO and slightlydepleted in SiO2 and TiO2 (Table 7).

8. Discussion

8.1. Episodes of Magma Recharge

[31] About 900 years ago, El Chichon Volcano explo-sively erupted producing a trachyandesitic pumice flowaround the crater (56.7% SiO2, unit C [Espındola et al.,2000]). According to the known history of El Chichon, noother eruptions occurred between the 900 and the 550 yearB.P. events. The absence of block-and-ash flow depositslinked to the volcanic stratigraphy of unit B suggests thatT

able

2.PlagioclaseCompositionsofUnitBPumices

byElectronMicroprobea

Type

Ph1c

Ph1r

Ph3c

Ph3i

Ph3r

Mph1

Mph2

Ph2c

Ph2i

Ph2r

Ml3

Ml4

Ph5c

Ph5r

Mph6

Mph7

Ph1c

Ph1i

Type

Ph1r

Ph2c

Ph2r

Ph4c

Ph4r

Ml1

Ml2

Ml3

bMl4

bMl5

bMph1

Sam

ple

B1

B1

B1

B1

B1

B1

B1

B2

B2

B2

B2

B2

B2

B2

B2

B2

B3

B3

Subunit

B3

B3

B3

B3

B3

B3

B3

B3

B3

B3

B3

WeightPercent

SiO

257.14

58.8652.30

58.36

59.58

52.44

53.53

59.0552.79

56.9349.8153.13

57.16

56.93

54.99

59.49

54.51

53.34SiO

254.95

53.33

52.75

57.43

55.52

52.25

52.74

58.95

58.40

57.41

57.67

Al 2O3

27.79

26.7930.81

26.90

25.75

30.06

30.05

26.5929.51

27.7631.9929.69

27.42

27.58

29.27

25.90

29.34

30.26Al 2O3

29.40

30.94

30.59

27.42

28.40

30.72

30.53

26.82

27.16

28.06

27.70

Fe 2O3c

0.21

0.28

0.19

0.20

0.42

0.56

0.31

0.16

0.39

0.34

0.71

0.34

0.23

0.38

0.38

0.27

0.28

0.44Fe 2O3c

0.33

0.44

0.30

0.40

0.36

0.67

0.53

0.29

0.30

0.49

0.26

CaO

9.20

8.0812.87

7.98

6.94

12.55

11.94

8.0211.82

9.2414.6011.86

9.17

9.19

10.70

7.33

11.02

12.58CaO

10.95

12.48

12.67

8.75

9.90

12.86

12.69

8.41

8.26

9.61

9.16

Na 2O

5.59

6.27

3.58

6.33

6.59

4.87

5.06

5.95

4.23

5.40

2.57

4.07

7.00

6.73

4.54

6.17

5.15

4.37Na 2O

5.08

4.05

3.80

6.82

6.21

3.75

4.48

5.75

6.20

5.48

5.76

K2O

0.47

0.66

0.24

0.65

0.79

0.27

0.30

0.68

0.29

0.56

0.12

0.29

0.47

0.52

0.40

0.84

0.34

0.30K2O

0.35

0.25

0.28

0.55

0.52

0.25

0.24

0.60

0.65

0.48

0.56

Total

100.40100.9499.99100.42100.07100.75101.19100.4599.03100.2399.8099.38101.45101.33100.28100.00100.64101.29Total

101.06101.49100.39101.37100.91100.50101.21100.82100.97101.53101.18

Mole

Percent

Ab

50.90

56.1033.00

56.70

60.20

40.60

42.70

54.9038.60

49.7024.0037.60

56.60

55.40

42.40

57.30

44.90

37.90Ab

44.70

36.40

34.60

56.80

51.70

34.00

38.50

53.30

55.40

49.40

51.50

An

46.30

40.0065.60

39.50

35.00

57.90

55.70

40.9059.60

47.0075.3060.60

40.90

41.80

55.20

37.60

53.10

60.30An

53.30

62.10

63.70

40.20

45.50

64.50

60.20

43.10

40.80

47.80

45.20

Or

2.80

3.90

1.40

3.80

4.80

1.50

1.70

4.10

1.80

3.40

0.70

1.80

2.50

2.80

2.50

5.10

2.00

1.70Or

2.00

1.50

1.70

3.00

2.80

1.50

1.40

3.70

3.80

2.80

3.30

aPlagioclaseanalyseswereperform

edontheJeolJX

A-8600electronmicroprobeat

theCNR-Florence

University

under

thefollowingconditions:accelerationpotential

of15kV;beam

currentof15nA;counting

time15–20s.Off-peakbackgroundcorrectionswereusedforallelem

ents.Naturalandsynthetic

mineralswereusedas

standards.AllFewas

assumed

tobeFe 2O3.Ph,phenocrysts;Mph,microphenocrysts;Ml,

microlites;c,

core;i,interior;r,rim.

bMicrolite

inclusionsin

amphibole

phenocrysts.

cFe 2O3isreported

astotaliron.

Figure 11. Plagioclase phenocryst compositions, showingchemical variation between cores and rims of single crystalsfrom B1, B2, and B3.


prior to the 550 year B.P. eruption the vent was not pluggedby a central dome but was rather an open crater throughwhich intense degassing took place, much like the present-day crater.[32] On the basis of our results we propose that before

the 550 year B.P. eruption, the 900 year B.P. trachyande-sitic magma residing beneath El Chichon was intruded bya hotter, more mafic magma of basaltic compositionproducing a hybrid trachyandesitic magma. Evidencefor this comes first from the presence of mafic enclaves(46.7 wt % SiO2) within unit B and from the zoningpatterns and chemical variations between cores and rimsin plagioclase phenocrysts (An40 – 53c, An59 – 60i, andAn41–53r) (Table 2 and Figure 11). Our results concurwith signs of mineral and isotopic disequilibrium reportedat the volcano by previous authors. McGee et al. [1987]described complex zoning of plagioclase phenocrystsassociated with anorthite spikes and inclusion-rich zonesof silicate glass and gas. Tilling and Arth [1994]described phenocryst phases that are not in isotopicequilibrium (clinopyroxene, apatite, anhydrite, and plagio-

clase). Espındola et al. [2000] reported mafic enclaves ofbasaltic composition (46.4 wt % SiO2) hosted in trachy-andesite rocks of the Guayabal Tuff Cone. Those maficenclaves, along with the Chapultenango trachybasalt(47.9 wt % SiO2) and the enclaves found in unit B(this work) represent the most mafic rocks recognized atthe volcano [Espındola et al., 2000]. Tepley et al. [2000]measured isotopic transects from core to rim in zonedplagioclase phenocrysts from the 1982 and the �210 kaeruptions (unit O [Espındola et al., 2000]). They discov-ered textural discontinuities marked by dissolution surfa-ces, large An fluctuations, and coeval decreases in87Sr/86Sr ratios with increasing Sr concentrations. Suchtextural and compositional fluctuations may have resultedfrom periodic recharge of high-temperature magmas withlow 87Sr/86Sr ratios and high Sr concentrations into coolerpreexisting magmas. Those magmas homogenized, pro-ducing compositionally homogeneous hybrids. Compara-ble results were obtained through in situ laser ablationisotopic analysis of plagioclase crystals of El Chichon[Davidson et al., 2000].

Table 3. Representative Hornblende Composition Determined by Electron Microprobea

Analysis

Ph1cb Ph1rb Ph3b Ph4cb Ph4rc Mph1b Mph3b Ph3c Ph4b Ph5b Ph7b Ph8b Ph3b Ph4b Ph7b Ph8b

Sample B-1 B-1 B-1 B-1 B-1 B-1 B-1 B-2 B-2 B-2 B-2 B-2 B-3 B-3 B-3 B-3

SiO2 41.40 41.38 41.20 42.41 42.63 40.51 42.69 42.60 41.29 40.55 40.59 40.58 41.69 42.28 40.84 40.79TiO2 2.56 2.35 2.44 1.77 1.49 2.22 1.56 1.75 2.23 2.52 2.50 2.34 2.03 2.17 2.34 2.15Al2O3 12.15 11.75 11.53 10.84 10.71 12.24 11.10 10.94 11.45 12.08 12.02 11.79 11.44 11.30 11.69 11.59FeOd 17.41 16.91 17.27 17.90 16.93 16.16 17.76 16.12 17.96 17.06 17.13 17.67 18.26 17.39 17.74 16.66MnO 0.55 0.52 0.44 0.71 0.62 0.51 0.61 0.53 0.57 0.64 0.51 0.48 0.54 0.56 0.57 0.42MgO 9.99 10.58 10.80 10.43 11.11 10.90 10.92 11.36 11.78 12.05 12.00 11.72 10.16 10.64 11.71 12.13CaO 11.60 11.62 11.31 11.67 11.53 11.72 11.49 11.57 11.60 11.76 11.58 11.55 11.77 11.54 11.59 11.75Na2O 1.70 1.94 1.91 1.83 1.85 2.14 2.00 1.61 2.51 2.51 2.39 2.11 1.95 1.76 2.33 2.46K2O 1.77 1.77 1.53 1.27 1.06 1.51 1.17 1.20 1.46 1.54 1.60 1.62 1.70 1.56 1.45 1.47Total 99.13 98.82 98.43 98.83 97.93 97.91 99.30 97.68 100.85 100.71 100.32 99.86 99.54 99.20 100.26 99.42Mge 57.00 59.31 63.32 59.53 64.47 62.01 63.54 65.87 67.87 69.70 70.27 69.92 56.56 60.71 68.94 68.86aIn weight percent. Amphiboles were calculated with MINPET program with the appropriate nomenclature of Leake [1978]. Amph, amphibole; Ph,

phenocrysts; Mph, microphenocrysts; c, core; r, rim.bMagnesian-hastingsite.cTschermakite.dFeO obtained by micropobe analysis.eMg = 100 Mg/(/Mg + Fe).

Table 4. Pyroxene Compositions Determined by Electron Microprobea

Analysis

Mph1 Ph2c Ph2r Mph2 Ph4c Ph4r Mph5 Ph5c Ph1 Ph2 Ph3 Ph4 Mph2 Mph1 Ph1 Ph2 Ph3 Ph4

Sample B1 B1 B1 B1 B1 B1 B1 B1 B2 B2 B2 B2 B2 B3 B3 B3 B3 B3

SiO2 52.52 50.53 49.71 50.75 51.18 51.90 51.80 51.89 51.80 53.52 51.59 51.84 51.16 51.58 50.69 51.25 51.58 50.69TiO2 0.31 0.53 0.51 0.57 0.54 0.25 0.15 0.38 0.15 0.29 0.30 0.21 0.21 0.38 0.53 0.20 0.38 0.53Al2O3 2.16 4.26 3.91 4.34 4.19 2.43 1.32 3.52 1.32 2.06 2.06 1.49 1.63 1.99 2.55 1.52 1.99 2.55FeOb 4.88 5.63 8.77 5.07 5.01 8.78 8.32 6.27 9.32 8.06 9.62 9.49 9.74 9.06 9.54 8.74 9.06 9.54Cr2O3 0.33 0.81 0.63 0.97 0.76 0.09 0.06 0.31 0.06 – – – – – – – – –MnO 0.17 0.14 0.47 0.20 0.16 0.82 0.96 0.12 0.96 0.62 0.62 0.98 0.84 0.62 0.76 0.76 0.62 0.76MgO 17.18 15.55 12.79 15.11 15.36 13.01 13.69 16.27 13.69 13.05 12.55 12.88 12.20 13.45 12.99 14.71 13.45 12.99CaO 23.01 23.04 22.42 23.52 23.73 22.63 23.26 21.53 23.26 22.06 22.68 22.82 23.54 22.96 22.66 23.01 22.96 22.66Na2O 0.19 0.39 0.50 0.32 0.28 0.66 0.49 0.50 0.49 0.53 0.51 0.45 0.54 0.53 0.63 0.58 0.53 0.63Total 100.8 100.9 99.7 100.9 101.2 100.6 100.1 100.8 101.1 100.3 99.9 100.2 99.9 100.6 100.4 100.8 100.6 100.4En 46.3 44.0 37.5 43.2 43.5 37.5 38.4 46.1 37.8 38.6 36.3 36.6 34.8 38.0 37.1 40.2 38.0 37.1aIn weight percent. Ph, Phenocrysts: Mph, Microphenocrysts; c, core; r, rim. En, enstatite content calculated with MINPET program [Richard, 1995].

MINPET pyroxene end-members are recalculated using the method of Cawthorn and Collerson [1974].bFeO obtained by micropobe analysis.


[33] The compositional gap observed between thetrachyandesite and trachybasalt products of El ChichonVolcano [Espındola et al., 2000; this work], and thecompositional homogeneity of the 550 year B.P. rocksindicate that melts involved in the mixing process hadenough time to hybridize. The time required needs not tobe long, as experimentally demonstrated by Kouchi andSunagawa [1985], who observed complete mixing ofandesitic and dacitic melts in a few hours. Magmatic watercontent between 2 and 4 wt % as reported for the 1982eruption [Luhr et al., 1984; McGee et al., 1987] would havereduced the viscosity of the magma, thereby improvingmixing efficiency [Tepley et al., 2000]. The mixing effi-ciency, however, is dependent upon the mixing proportionsof the two end-members [Koyaguchi, 1986]. In the case ofthe 550 year B.P. eruption, we performed petrologic mixingcalculations with the IGPET software that uses the leastsquares regression of major elements of Bryan et al. [1969].Considering intervals of 10% of mixing between thesemagmas, the hybrid magma formation would require�90 vol % of the stagnated trachyandesitic magma to mixwith 10 vol % of the hotter basaltic melt. In the case of the550 year B.P. eruption, compositional homogenizationwould have occurred only if the equilibrium temperaturebetween the two melts was sufficiently higher than the

solidus temperature of the mafic magma. If this last condi-tion had not been satisfied, the mafic magma would havebeen disaggregated into droplets (enclaves) in the coolerpreexisting magma.

8.2. Preeruptive Conditions of the Hybrid Magma

[34] Ongoing hydrothermal experiments with pumicesamples of unit B suggest that, prior to the eruption, themagma equilibrated at pressures around 2–2.5 kbar (�6–7.5 km) (J. E. Gardner, personal communication, 2001).These pressure conditions are slightly higher than theexperimentally obtained of value �2 kbar for the 1982eruption [Luhr, 1990]. The seismicity associated with the1982 eruption [Jimenez et al., 1999] shows a quiescentzone between 7 and 13 km below the volcano in theseismicity preceding the eruption that might represent adeeper magma reservoir (Figure 13). These observationsindicate that the location of El Chichon magma reservoirmay have shifted slightly from the 550 year B.P. to the1982 eruption toward shallower depths. A similar magmamigration with time was demonstrated experimentally formagmas of the last 4000 years at Mount St. Helens[Gardner et al., 1995].[35] It is reasonable to consider that the 550 year B.P.

hybrid magma stagnated at depths �6–7.5 km below the

Figure 12. Classification diagram of different clinopyroxene crystals of unit B samples. Allclinopyroxene crystals fall in the diopside field.

Table 5. Ti-Fe Oxides Chemical Compositions Obtained by Electron Microprobea

Analysis

Tm3 Tm4 Tm5 Tm6 Tm7 Ilm1 Ilm2 Tm8 Tm9 Tm10 Tm11 Tm12 Tm13 Tm14 Tm15 Tm16

Sample B1 B1 B1 B1 B2b B2b B2b B2b B2 B2 B2 B3 B3 B3 B3 B3

Weight PercentSiO2 0.08 0.08 0.05 0.08 0.07 0.06 0.01 0.08 0.04 0.10 0.02 0.15 0.16 0.10 0.13 0.07TiO2 6.00 6.09 5.92 5.72 5.95 31.82 35.20 6.00 6.10 5.99 5.82 6.13 5.59 5.82 5.88 5.93Al2O3 1.96 1.92 1.98 1.91 1.82 0.36 0.27 1.90 1.85 2.00 2.01 1.91 1.87 1.93 1.96 1.86FeOc 83.83 85.44 85.49 84.95 85.91 62.56 61.02 82.70 83.81 84.70 85.36 84.93 84.36 85.73 84.35 83.27MnO 1.04 0.97 0.92 0.93 1.03 0.62 0.30 1.00 1.05 1.08 0.78 0.97 0.94 1.08 0.99 0.93MgO 0.96 1.08 1.30 1.07 1.04 1.06 0.90 1.20 1.30 0.00 1.11 1.06 1.25 1.34 1.30 1.02Cr2O3 0.01 0.01 0.10 0.10 0.09 0.07 0.02 0.20 0.03 0.04 0.05 0.03 0.07 0.06 0.02 0.04Total 93.88 95.59 95.76 94.76 95.91 96.55 97.72 93.08 94.18 93.91 95.15 95.18 94.24 96.06 94.63 93.12

FeOc 34.34 34.88 34.47 34.34 34.83 26.17 29.76 33.74 35.39 35.80 34.58 34.88 33.86 34.32 35.63 34.02Fe2O3c 55.00 56.19 56.70 56.25 56.76 40.44 34.74 54.41 53.81 54.35 56.44 55.62 56.13 57.14 54.14 54.73X0Usp 17.56 17.4 16.77 16.53 16.91 17.59 18.77 18.45 16.78 17.66 16.09 16.31 18.25 17.43

X0Ilm 60.1 66.3aBy the magnetite-ulvospinel (spinel phase) molar proportions of cations normalized to a formula unit of three sites, by ilmenite-hematite (rhomboedral

phase) molar proportions of cations normalized to a formula unit of two sites, FeOc and Fe2O3 c, and X0Usp and X0

Ilm were calculated with the model ofStormer [1983]. FeOc, Fe2O3c were calculated.Tm, titanomagnetite; X0

Ilm, ilmenite content; X0Usp, ulvospinel content.

bTemperature is 820–830�C, calculated with the Andersen and Lindsley [1988] geothermomether.cFeO obtained by micropobe analysis.


summit at a temperature of 820–830�C and had H2Ocontents of 5–6 wt %, allowing plagioclase, hornblende,and clinopyroxene to crystallize. The essentially uniformcompositions of hornblende and clinopyroxene suggest thatthe hybrid magma did not experience further injections butrather evolved allowing crystal growth under stable con-ditions. Crystal growth and gas exsolution, increased theproportions of gas bubbles that overpressurize the systemleading to the explosive eruption [Bower and Woods, 1997;Eichelberger, 1995; Tait et al., 1998].[36] However, several volatile species must have played a

significant role in the dynamics of the system for theestimated temperature and pressure conditions; the 550 yearB.P. magma was undersaturated in water (�80%) [Wallaceand Anderson, 2000]. Experimental data obtained for mag-

mas erupted at Mount St Helens indicated that many of thedacitic magmas were also significantly undersaturated inwater [Rutherford and Devine, 1988]. Rutherford andDevine concluded that an excess vapor phase was presentbefore the eruption. The high sulfur emissions recordedduring the 1982 Chichon eruption [Rye et al., 1984] suggestthat SO2 and/or CO2 could have also been present duringthe 550 year B.P. eruption. Excess CO2 can be explained byrepeated injections of basaltic magma into the magmasystem [Fogel and Rutherford, 1990] and excess SO2 bythe interaction of the magma with Mesozoic evaporitesunderneath El Chichon [Duffield et al., 1984; Rye et al.,1984].[37] The vesicularity of the pumice fragments ranges

greatly (51–74 vol %) and records the overpressurization

Table 6. Glass Composition Obtained by Electron Microprobea

Analyses

gg 1 gg 2 gg 4 gg 5 gg 7 gg 8 gg 9 gg 10 gg 11 gg 12 gg 13 gg 14 gg 15 gg 16

Sample B1 B1 B2 B2 B2 B2 B2 B2 B3 B3 B3 BG BG BG

SiO2 69.13 69.46 69.41 69.59 68.50 68.41 68.52 68.55 66.71 66.63 68.79 66.78 68.65 67.64TiO2 0.21 0.25 0.25 0.30 0.25 0.21 0.22 0.25 0.24 0.22 0.24 0.47 0.42 0.41Al2O3 16.32 16.14 16.23 16.32 16.65 16.26 16.65 16.23 16.26 16.31 16.78 16.34 16.02 16.15Cr2O3 0.02 0.02 – 0.02 0.01 – – – – – 0.02 – 0.02 0.04FeO 1.22 1.32 1.39 1.48 1.86 1.46 1.39 1.36 1.25 1.53 1.55 2.73 2.44 2.79MnO 0.05 0.09 0.15 0.05 0.10 0.11 0.04 0.09 0.08 0.11 0.10 0.12 0.12 0.13MgO 0.18 0.26 0.17 0.20 0.24 0.27 0.21 0.26 0.22 0.17 0.21 0.89 0.41 0.52CaO 1.65 1.84 1.59 1.66 1.94 1.82 1.68 1.83 1.78 1.90 1.90 2.62 2.37 2.13Na2O 2.71 2.78 2.28 2.03 4.72 3.25 3.66 2.11 4.65 4.36 2.40 3.15 3.09 2.76K2O 4.54 4.62 4.26 4.23 4.86 4.52 4.67 3.96 4.41 4.89 4.38 4.65 4.03 4.52SrO 0.10 0.07BaO 0.13 0.11F 0.07 0.08 0.03SO3 0.05 0.05Cl 0.19 0.22 0.20 0.22Total 96.03 96.78 95.99 96.18 99.64 96.31 97.04 94.64 96.05 96.12 96.37 97.75 97.57 97.09aIn weight percent. Analyses were performed on the Jeol JXA-8600 electron microprobe at the CNR-Florence University. Analyses were performed with

a focused beam of 5 or 10 mm in diameter; gg1, glass in the groundmass.

Table 7. Apatite and Sphene Compositions Obtained by Electron Microprobea

Analyses

Average Composition Luhr Compositionb Diff.

Analyses

Average Composition Luhr Compositionb Diff.Ap1 Ap2 Sp1 Sp1 Sp2

Sample B1 B1 B2 B1 B3SiO2 0.33 0.04 0.19 0.22 0.04 29.58 29.64 29.72 29.65 30.04 0.39TiO2 – – – – – 35.88 36.07 35.66 35.87 36.84 0.97Al2O3 0.04 0.02 0.03 0.01 �0.02 1.27 1.18 1.18 1.21 0.92 �0.29Cr2O3 0.01 0.01 0.01 0.02 0.01 – – – – – –FeO 0.10 0.12 0.11 0.13 0.02 1.56 1.27 1.21 1.35 1.30 �0.05MnO 0.14 0.18 0.16 0.13 �0.03 0.14 0.11 0.15 0.13 0.13 0.00MgO 0.09 0.04 0.07 0.06 �0.01 0.06 – 0.02 0.03 0.02 �0.01CaO 52.70 55.60 54.15 54.20 0.05 26.95 27.63 28.05 27.54 26.95 �0.59Na2O 0.23 0.09 0.16 0.09 �0.07 – 0.05 0.11 0.05 – –K2O 0.03 0.03 0.03 0.04 0.01 0.03 – – 0.01 – –SrO 0.14 – 0.07 0.13 0.06 – – – – – –F 2.43 2.22 2.33 2.24 0.09 – – – – – –Cl 0.53 0.61 0.57 0.64 0.07 – – – – – –SO3 0.33 – 0.17 0.34 0.18 – – – – – –P2O5 40.39 41.45 40.92 42.04 1.12 – – – – – –Total 97.49 100.4 98.955 100.29 95.47 95.95 96.10 95.84 96.20aIn weight percent. Sp, sphene; Ap, apatite.bLuhr et al. [1984] = [1]. Diff. is variation between [1] and this work analysis.


of the magma system upon eruption, overlapping theproposed range for pumice disruption in explosive eruptions[Gardner et al., 1996]. Yellow and gray pumice of unit Bhas two populations of bubbles that might record twodistinct processes. Large irregular-shaped vesicles mighthave exsolved and coalesced within the magma reservoir,whereas the small semirounded vesicles might be theproduct of nucleation during sudden decompression of themagmatic system, as experimentally demonstrated [Gardneret al., 1999].

8.3. Hazard Implications

[38] Nowadays, the area blanketed by the 550 falloutlayer is occupied by many small villages and several largertowns, with a total population of about 66,000 people. Forcomparison, these towns were covered by only 3 cm offallout (A1) from the 1982 eruption. If we consider the fineash transported by predominant winds during the 1982eruption, we should recall that a dense cloud of ashdarkened Villahermosa, the capital city of the State ofTabasco, at midday during 29 March 1982. Furthermore,a large part of the Yucatan peninsula was obscured by far-reaching ash, affecting the important cities of Campeche,Merida, and Chetumal. The 1-mm isopach of the eruptioncovered an area of 45,000 km2, affecting many cities of theregion as Villahermosa and Palenque [Varekamp et al.,1984]. Because the 550 year B.P. eruption was at least1 order of magnitude larger than the 1982 air fall events,these cities would be affected by an estimated 2 cm of ash.The disruption caused by the 550 year B.P. eruptioncaused a great impact to the prehispanic centers locatedin the area, but no legends or archaeological evidence ofabandonment of the area are known. In fact, such a bigeruption in proximity of important cultures such as thepostclassical Maya, must have had a serious impact on

their development, and thus have important bearing onarcheological studies in the area. It is interesting to notethat such events as the 1450–1454 A.D. drought followedby famine in the Yucatan peninsula are well recorded[Ford and Rose, 1995; Gill, 2000]. As we have seen theeruption investigated in this work occurred sometime inthe period 1320–1433 (at 1s ± calibration [Espındola etal., 2000]). A Plinian eruption of this magnitude wouldaffect extensive areas within a radius of 30 km around thevolcano, posing a serious volcanic hazard to more than66,000 inhabitants.

9. Conclusions

[39] The last explosive eruption at El Chichon Volcanobefore the 1982 event occurred �550 years ago. Thiseruption started under open vent conditions with the gener-ation of minor pyroclastic flows, which ultimately openedthe magma conduit and gave rise to a Plinian column thatrose to a height of 31 km. The fallout from this plume wasdispersed by predominant NE 30� winds, producing tephradeposits with a total volume of �1.1 km3 DRE.[40] The eruption was triggered by a magma-mixing

event involving the injection of a hot mafic magma into acooler stagnating trachyandesite magma. The mixing pro-duced a hybrid trachyandesitic magma that then evolved inthe reservoir, allowing crystal growth and gas exsolution,which increased the proportions of gas bubbles, and ulti-mately led to the explosive eruption. Before the eruption,the magma was in equilibrium at a temperature between820 and 830�C, and pressures of �2–2.5 kbar (6–7.5 kmbelow the volcano).[41] The 550 year B.P. event was 1 order magnitude

larger than the Plinian events of the 1982 eruption. Com-parison between data from both the 550 year B.P. and 1982

Figure 13. Cartoon illustrating the storage conditions beneath El Chichon Volcano before majorexplosive events 550 years ago and in 1982 (see text discussion). References are 1, Carey and Sigurdsson[1986]; 2, Jimenez et al. [1999]; 3, Luhr et al. [1984]; and 4, Luhr [1990].


eruptions makes clear that it is important to know theevolution and behavior of the magma system betweeneruptions.

[42] Acknowledgments. This research was supported by CONACYTgrants (27993-T to J. L. M., 32312-T to J. M. E., 40396 A-1 to J. C. Mora),a CONACYT-CNR bilateral project to J. L. M., and a GSA ‘‘LipmanResearch Fund’’ to J. L. A. The help of several people is gratefullyacknowledged: F. Olmi provided technical support during the microprobeanalyses performed at CNR at University of Florence, G. Valdez,T. Hernandez, G. Solıs, and P. Schaff provided technical and instrumentalsupport. We also thank W. A., Duffield, J. Davidson, R. I. Tilling,J. Gardner, F. Tepley, J. Luhr, G. Boudon, and C. Bacon for their carefulrevision of the manuscript.

ReferencesAndersen, D. J., and D. H. Lindsley, Internally consistent solutions modelsfor Fe-Mg-Mn-Ti oxides: Fe-Ti oxides, Am. Mineral., 73, 714–726,1988.

Bacon, C. R., and M. M. Hirshmann, Mg/Mn partitioning as a test forequilibrium between coexisting Fe-Ti oxides, Am. Mineral., 53, 57–61,1988.

Bower, S. M., and A. W. Woods, Control of magma volatile content andchamber depth on the mass erupted during explosive volcanic eruptions,J. Geophys. Res., 102, 10,273–10,290, 1997.

Bryan, W. B., L. W. Finger, and F. Chayes, Estimating proportions inpetrographic mixing equations by least squares approximation, Science,163, 926–927, 1969.

Canul, R. F., and V. L. Rocha, Informe geologico de la zona geotermica de‘‘El Chichonal,’’ Chiapas, Com. Fed. de Electr., Morelia, Mexico, 1981.

Canul, R. F., A. M. Razo, and V. L. Rocha, Geologıa e historia volcanolo-gica del volcan Chichonal, Estado de Chiapas, in El Volcan Chichonal,Rev. Inst. Geol. Univ. Nac. Autom. Mex., special issue, pp. 3–22, Inst. deGeol., Univ. Nat. Autom. Mex., Mexico City, 1983.

Carey, S. N., and H. Sigurdsson, The 1982 eruptions of El Chichon volcano,Mexico, 2: Observations and numerical modeling of tephra fall distribu-tion, Bull. Volcanol., 48, 127–141, 1986.

Carey, S. N., and H. Sigurdsson, The intensity of Plinian eruptions, Bull.Volcanol., 51, 28–40, 1989.

Carey, S. N., and R. S. J. Sparks, Quantitative models of fallout and dis-persal of tephra from volcanic eruption columns, Bull. Volcanol., 48,109–125, 1986.

Carey, S. N., J. E. Gardner, and H. Sigurdsson, The intensity and magnitudeof Holocene Plinian eruptions from Mount St. Helens volcano, J. Volca-nol. Geotherm. Res., 66, 185–202, 1995.

Carroll, M. R., and M. J. Rutherford, The stability of igneous anhydrite:Experimental results and implications for sulfur behavior in the 1982 ElChichon trachyandesite and other evolved magmas, J. Petrol., 28, 781–801, 1987.

Cawthorn, A. F., and K. D. Collerson, The recalculation of pyroxene endmember parameters and estimation of ferrous and ferric iron content frommicroprobe analyses, Am. Mineral., 59, 1203–1208, 1974.

Damon, P., and E. Montesinos, Late Cenozoic volcanism and metallogen-esis over an active Benioff Zone in Chiapas, Mexico, Ariz. Geol. Soc.Dig., 11, 155–168, 1978.

Davidson, J., F. Tepley, Z. Palacz, and S. Meffan-Main, Magma recharge,contamination and residence times revealed by in situ laser ablation iso-topic analysis of feldspar in volcanic rocks, Earth Planet. Sci. Lett., 184,427–442, 2000.

Devine, J. D., H. Sigurdsson, A. N. Davis, and S. Self, Estimates of sulfurand chlorine yield to the atmosphere from volcanic eruptions and poten-tial climatic effects, J. Geophys. Res., 89, 6309–6325, 1984.

Duffield, W. A., R. I. Tilling, and R. Canul, Geology of El Chichon volcano,Chiapas, Mexico, J. Volcanol. Geotherm. Res., 20, 117–132, 1984.

Eichelberger, J. C., Silicic volcanism: Ascent of viscous magmas fromcrustal reservoirs, Annu. Rev. Earth Planet. Sci., 23, 41–63, 1995.

Espındola, J. M., J. L. Macıas, R. I. Tilling, and M. F. Sheridan, Volcanichistory of El Chichon Volcano (Chiapas, Mexico) during the Holocene,and its impact on human activity, Bull. Volcanol., 62, 90–104, 2000.

Fierstein, J., and M. Nathenson, Another look at the calculation of fallouttephra volumes, Bull. Volcanol., 54, 156–167, 1992.

Fogel, R., and M. J. Rutherford, The solubility of carbon dioxide in rhyo-litic melts: A quantitative FTIR study, Am. Mineral., 75, 1311–1326,1990.

Ford, A., and W. I. Rose, Volcanic ash in ancient Maya ceramics of thelimestone lowlands: Implications for prehistoric volcanic activity in theGuatemala highlands, J. Volcanol. Geotherm. Res., 66, 149–162,1995.

Gardner, J. E., M. Rutherford, S. Carey, and H. Sigurdsson, Experimentalconstraints on pre-eruptive water contents and changing magma storageprior to explosive eruptions of Mount St. Helens volcano, Bull. Volcanol.,57, 1–17, 1995.

Gardner, J. E., R. M. E. Thomas, and C. Jaupart, Fragmentation of magmaduring Plinian volcanic eruptions, Bull. Volcanol., 58, 144–162, 1996.

Gardner, J. E., M. Hilton, and M. R. Carroll, Experimental constrains ondegassing of magma: Isothermal bubble growth during continuousdecompression from high pressure, Earth Planet. Sci. Lett., 168, 201–218, 1999.

Gill, J. B., Orogenic Andesites and Plate Tectonics, 358 pp., Springer-Verlag, New York, 1981.

Gill, R. B., The Great Maya Droughts: Water, Life and Death, 464 pp.,Univ. of N. M. Press, Albuquerque, 2000.

Gill, R. B., and J. P. Keating, Volcanism and Mesoamerican archaeology,Ancient Mesoamerica, 13, 125–140, 2002.

Housh, T. B., and J. F. Luhr, Plagioclase-melt equilibria in hydrous systems,Am. Mineral., 76, 477–492, 1991.

Jimenez, Z., V. H. Espındola, and J. M. Espındola, Evolution of the seismicactivity from the 1982 eruption of El Chichon Volcano, Chiapas, Mexico,Bull. Volcanol., 61, 411–422, 1999.

Kouchi, A., and I. Sunagawa, A model for mixing basaltic and daciticmagmas as deduced from experimental data, Contrib. Mineral. Petrol.,89, 17–23, 1985.

Koyaguchi, T., Textural and compositional evidence for magma mixing andits mechanism, Abu volcano group, southwestern Japan, Contrib.Mineral. Petrol., 93, 33–45, 1986.

Le Bas, M. J., R. W. Le Maitre, A. Streckeisen, and R. Zanettin, A chemicalclassification of volcanic rocks based on the total alkali-silica diagram,J. Petrol., 27, 745–750, 1986.

Luhr, J., Experimental phase relations of water-and-sulfur saturated arcmagmas and the 1982 eruption of El Chichon volcano, J. Petrol., 31,1071–1114, 1990.

Luhr, J., and M. V. A. Logan, Sulfur isotope systematics of the 1982 ElChichon trachyandesite: An ion microprobe study, Geochim. Cosmochim.Acta, 66, 3303–3316, 2002.

Luhr, J. F., I. S. E. Carmichael, and J. C. Varekamp, The 1982 eruptions ofEl Chichon Volcan, Chiapas, Mexico: Mineralogy and petrology of theanhydrite-bearing pumices, J. Volcanol. Geotherm. Res., 23, 69–108,1984.

Macıas, J. L., Violent short-lived eruptions from small-size volcanoes:El Chichon, Mexico (1982) and Shtyubel’, Russia (1907), Ph.D. disserta-tion thesis, State Univ. of New York at Buffalo, Buffalo, 1994.

Macıas, J. L., J. M. Espındola, Y. Taran, and P. A. Garcıa, Explosivevolcanic activity during the last 3,500 years at El Chichon Volcano,Mexico, field trip guide, 53 pp., IAVCEI General Assem., Puerto Vallarta,Mexico, 1997.

Matson, M., The 1982 El Chichon volcanic eruptions: A satellite perspec-tive, J. Volcanol. Geotherm. Res., 23, 1–10, 1984.

McGee, J. J., R. I. Tilling, and W. A. Duffield, Petrologic characteristics ofthe 1982 and pre-1982 eruptive products of El Chichon volcano, Chiapas,Mexico, Geofis. Int., 26, 10–85, 1987.

Parker, D. E., and J. L. Brownscombe, Stratospheric warming following theEl Chichon volcanic eruption, Nature, 301, 406–408, 1983.

Pyle, D. M., The thickness, volume and grainsize of tephra fall deposits,Bull. Volcanol., 51, 1–15, 1989.

Richard, L. R., MINPET Program, version 2.02, MINPET Geol. Software,Quebec, 1995.

Rose, W. I., T. J. Bornhorst, S. P. Halsor, W. A. Capaul, S. Lumley, S. De laCruz-Reyna, M. Mena, and R. Mota, Volcan El Chichon, Mexico: Pre-1982 S-rich eruptive activity, J. Volcanol. Geotherm. Res., 23, 147–167,1984.

Rutherford, M. J., and J. D. Devine, The May 18, 1980, eruption of MountSt. Helens: 3. Stability and chemistry of amphibole in the magma cham-ber, J. Geophys. Res., 93, 11,949–11,959, 1988.

Rye, R. O., J. F. Luhr, and M. D. Wasserman, Sulfur and oxygen isotopicsystematics of the 1982 eruptions of El Chichon Volcano, Chiapas,Mexico, J. Volcanol. Geotherm. Res., 23, 109–123, 1984.

Sigurdsson, H., S. N. Carey, and J. M. Espındola, The 1982 eruptions ofEl Chichon volcano, Mexico: Stratigraphy of pyroclastic deposits,J. Volcanol. Geotherm. Res., 23, 11–37, 1984.

Sparks, R. S. J., The dimensions and dynamics of volcanic eruption col-umns, Bull. Volcanol., 48, 3–15, 1986.

Stormer, J. C., The effects of recalculation on estimates of temperature andoxygen fugacity from analysis of multi-component iron-titanium oxides,Am. Mineral., 68, 586–594, 1983.

Tait, S., R. Thomas, J. E. Gardner, and C. Jaupart, Constraints on coolingrates and permeabilities of pumice in an explosive eruption jet fromcolour and magnetic mineralogy, J. Volcanol. Geotherm. Res., 86, 79–91, 1998.


Tepley, F. J., III, J. P. Davidson, R. I. Tilling, and J. G. Arth, Magma mixing,recharge and eruption histories recorded in plagioclase phenocrysts fromEl Chichon Volcano, Mexico, J. Petrol., 41, 1397–1411, 2000.

Tilling, R. I., and J. G. Arth, Sr and Nd isotopic compositions of sulfur-richmagmas of El Chichon volcano, Mexico, paper presented at IAVCEICongress, Middle East Tech. Univ., Dep. of Geol. Eng., Ankara, Turkey,1994.

Tilling, R. I., M. Rubin, H. Sigurdsson, S. Carey, and W. A. Duffield,Prehistoric eruptive activity of El Chichon volcano, Mexico, Science,224, 747–749, 1984.

Varekamp, J. C., J. F. Luhr, and K. L. Prestegaard, The 1982 eruptions of ElChichon volcano (Chiapas, Mexico): Character of the eruptions, ash-falldeposits, and gas phase, J. Volcanol. Geotherm. Res., 23, 39–68, 1984.

Walker, G. P. L., Grain size characteristics of pyroclastic flows, J. Geol., 79,696–714, 1971.

Wallace, P., and A. T. Anderson, Volatiles in magmas, in Encyclopedia ofVolcanoes, edited by H. Sigurdsson, pp. 149–170, Academic, San Diego,Calif., 2000.

Williams, S. N., and S. Self, The October 1902. Plinian eruption of SantaMarıa Volcano, Guatemala, J. Volcanol. Geotherm. Res., 16, 33–56,1983.

Zielinski, G. A., Stratospheric loading and optical depth estimates ofexplosive volcanism over the last 2100 years derived from the GreenlandIce Sheet Project 2 ice core, J. Geophys. Res., 100, 20,937–20,955, 1995.

Zielinski, G. A., J. E. Dibb, Q. Yang, P. A. Mayewski, S. Whitlow, M. S.Twickler, and M. S. Germani, Assessment of the record of the 1982 ElChichon eruption as preserved in Greenland snow, J. Geophys. Res., 102,30,031–30,045, 1997.

��J. L. Arce, J. M. Espındola, J. L. Macıas (corresponding author), and J. C.

Mora, Instituto de Geofısica, UNAM, Coyoacan 04510, Mexico City, D.F.,Mexico. ([email protected]; [email protected]; [email protected]; [email protected])P. Manetti, Dipartimento di Scienze della Terra, Universita degli Studi di

Firenze, Via G. Moruzzi, 1, Pisa, Toscana I-56124, Italy. ([email protected])R. Saucedo, Instituto de Geologıa, Universidad Autonoma de San Luis

Potosı, Av. Dr. Manuel Nava No. 5, Zona Universitaria, San Luis Potosi78240, Mexico. ([email protected])


a 550-year-old plinian eruption at el chicho´n volcano ...raman/papers2/maciasecjgr.pdf · a...

Documents