Full Terms & Conditions of access and use can be found athttp://www.tandfonline.com/action/journalInformation?journalCode=tigr20

Download by: [UNAM Ciudad Universitaria] Date: 03 May 2017, At: 15:31

International Geology Review

ISSN: 0020-6814 (Print) 1938-2839 (Online) Journal homepage: http://www.tandfonline.com/loi/tigr20

New stratigraphic, geochronological, andstructural data from the southern GuanajuatoMining District, México: implications for thecaldera hypothesis

Ángel Francisco Nieto-Samaniego, Javier Antonio Báez-López, GillesLevresse, Susana Alicia Alaniz-Alvarez, Carlos Ortega-Obregón, MargaritaLópez-Martínez, Benito Noguez-Alcántara & Jesús Solé-Viñas

To cite this article: Ángel Francisco Nieto-Samaniego, Javier Antonio Báez-López, GillesLevresse, Susana Alicia Alaniz-Alvarez, Carlos Ortega-Obregón, Margarita López-Martínez, BenitoNoguez-Alcántara & Jesús Solé-Viñas (2016) New stratigraphic, geochronological, and structuraldata from the southern Guanajuato Mining District, México: implications for the caldera hypothesis,International Geology Review, 58:2, 246-262, DOI: 10.1080/00206814.2015.1072745

To link to this article: http://dx.doi.org/10.1080/00206814.2015.1072745

View supplementary material Published online: 31 Jul 2015.

Submit your article to this journal Article views: 192

View related articles View Crossmark data

Citing articles: 2 View citing articles

New stratigraphic, geochronological, and structural data from the southernGuanajuato Mining District, México: implications for the caldera hypothesisÁngel Francisco Nieto-Samaniegoa, Javier Antonio Báez-Lópeza,b,c, Gilles Levressea, Susana Alicia Alaniz-Alvareza,Carlos Ortega-Obregóna, Margarita López-Martínezd, Benito Noguez-Alcántarac and Jesús Solé-Viñase

aCentro de Geociencias, Universidad Nacional Autónoma de México, Querétaro, México; bPosgrado en Ciencias de la Tierra, Centro deGeociencias, Universidad Nacional Autónoma de México, Querétaro, México; cFresnillo PLC, Metepec, Mexico; dCentro de InvestigaciónCientífica de Educación Superior de Ensenada, Ensenada, México; eInstituto de Geología, Universidad Nacional Autónoma de México, CiudadUniversitaria, México

ABSTRACTThe Cenozoic stratigraphy of the southern Guanajuato Mining District (GMD) was established 40years ago. The existence of a caldera structure that produced the Cenozoic volcanic cover waspostulated and the world-class silver ore deposit of the Oligocene age has been closely related tomagmatism. In this context, we present a new geological map of the southern GMD, U–Pb andAr–Ar ages of the volcanic units, and structural data for the Cenozoic faults. Our results documentthat the volcanic centre was active between ca. 33.5 Ma and ca. 31.3 Ma, coeval with NW–SEnormal faulting. We propose that the Bufa, Calderones, and Cedro formations are stratigraphicunits directly related to the volcanic centre. Although the younger Chichíndaro Rhyolite scarcelycrops out within the study area, it appears to be more extensive outside of the study area, formingpart of the rhyolitic volcanism of the Mesa Central of Mexico. In the study area, the ChichíndaroRhyolite buries major faults, demonstrating that it was emplaced after the peak of faulting. Thetwo main structures are the El Cubo and Veta Madre grabens; also there are several faulted andbrecciated zones where silver–gold mineralization was emplaced. The extension direction chan-ged from NE to NW producing normal faulting, reactivating older structures and allowing dikeintrusion. The extensional phase continued to be active throughout the Oligocene. The age of thevolcanic event and a new K–Ar age of the Veta Madre vein of 29.8 ± 0.8 Ma (K–Ar in adularia)indicate that the hydrothermal event began immediately after the emplacement of the CedroFormation. The emplacement of the Chichíndaro Rhyolite allowed hydrothermal activity to beactive for two million years or more.

ARTICLE HISTORYReceived 27 April 2015Accepted 11 July 2015

KEYWORDSGuanajuato Mining Districtstratigraphy; normal fault-ing; extensional volcanism;Cenozoic caldera; silver-oredeposit

Introduction

The Guanajuato Mining District (GMD) is located in thesouthern limit of the Mesa Central (MC) physiographicprovince, in the southeastern Sierra de Guanajuato, cen-tral México (Figure 1). The stratigraphic column of GMDhas been studied for a century, and the presence of twogroups of rocks is well-established: the older group isMesozoic and was studied in detail in the Sierra deGuanajuato. This group is formed of Jurassic–Cretaceous marine volcanic and sedimentary rocks, aplutonic suite including tonalite and diorite, zones ofbasaltic and doleritic dike swarms, and an outcrop ofultramafic rocks (Echegoyen-Sánchez et al. 1970;Martínez-Reyes 1992; Lapierre et al. 1992; Ortiz-Hernández et al. 2003; Mortensen et al. 2008; Martiniet al. 2011). The second group consists of Cenozoicterrestrial deposits, which unconformably overly

Mesozoic rocks. The Cenozoic volcanic rocks of theGMD correlate with the volcanic cover of the SierraMadre Occidental volcanic province, which is a pile ofsilicic ash flow tuffs and rhyolitic lavas with minoramounts of andesitic lavas with an average thicknessthat exceeds 1 km (McDowell and Clabaugh 1979).Those volcanic rocks cluster into three discrete groups:Eocene, Oligocene, and Miocene ages (Ferrari et al.2007). In the GMD, the Cenozoic rocks have receivedless attention in the geological literature than theMesozoic rocks. A conglomerate at the base of theCenozoic cover is intercalated with a small number oflava flows and pyroclastic deposits. Covering the con-glomerate, there is a group of volcanic rocks, the strati-graphic units of which were established in Echegoyen-Sánchez et al. (1970). The Cenozoic stratigraphy has notbeen modified or refined since the 1970s, and there are

CONTACT: Ángel Francisco Nieto-Samaniego afns@geociencias.unam.mxSupplemental data for this article can be accessed at http://dx.doi.org/10.1080/00206814.2015.1072745.

INTERNATIONAL GEOLOGY REVIEW, 2016VOL. 58, NO. 2, 246–262http://dx.doi.org/10.1080/00206814.2015.1072745

© 2015 Taylor & Francis

only three units with published K–Ar isotopic ages(Nieto-Samaniego et al. 1996; Aranda-Gómez andMcDowell 1998). Two of those ages, published byGross (1975), have large uncertainties because they cor-respond to ignimbrite and rhyolite whole-rock analysis,and the sample locations are unknown.

Based on the location of the volcanic facies and theexistence of a megabreccia with clasts of Mesozoicrocks, Randall et al. (1994) proposed a caldera origin ofthe Cenozoic volcanic rocks of the GMD. This idea wasfollowed by Davis (2005) and Davis et al. (2009), whoargued that ‘the eastern-dipping of beds in the centraland eastern parts of the district, the location of mega-breccia only on the eastern edge, the anticlinal struc-tures on the western side of the district, the thinning of

bed thickness moving from east to west, and the pre-sence of major domal intrusions on the eastern edgeand absence on the western edge of the district supporta trapdoor collapse for the caldera’. According to theseauthors, the source of the Cenozoic volcanic rocks waslocated near El Cubo (Figures 1 and 2). Although theseauthors traced a half-circular structure following theoutcrops of rhyolite domes, this structure does notappear in geologic maps and is not obvious in digitalelevation models. Instead of an ellipsoidal or sub-circu-lar structure, a system of W-dipping faults has beenmapped, resembling a domino style structure(Echegoyen-Sánchez et al. 1970; Martínez-Reyes 1992;Alvarado-Méndez et al. 1998; López-Ojeda et al. 2002;Davis 2005; Davis et al. 2009).

Figure 1. (A) Location of the study area relative to the major neighbouring physiographic provinces: SMOc, Sierra Madre Occidental;SMOr, Sierra Madre Oriental; MC, Mesa Central; FVTM, Transmexican Volcanic Belt. The grey zone indicates the area covered byCenozoic volcanic rocks of the SMOc volcanic province; note that they extend into the MC and to the south of the FVTM. (B) Regionalgeologic map, modified from Nieto-Samaniego et al. (1997); the Oligocene rhyolitic rocks (Tol) correspond to the cover of the SMOcvolcanic province; SLP, San Luis Potosí; SMA, San Miguel de Allende; Ags, Aguascalientes. (C) Digital elevation model showing thelocation of the Sierra de Guanajuato, the Guanajuato Mining District, and the study area; red dashed line shows the area covered bythe caldera-related Guanajuato volcanic group.

INTERNATIONAL GEOLOGY REVIEW 247

Randall et al. (1994) and Davis et al. (2009) do notreport radiometric ages of the volcanic rocks. The agesare needed to support the caldera hypothesis becausethey could indicate that the volcanic rocks were depos-ited during an interval of time that agrees with themagma residence time in caldera structures (Jellinekand DePaolo 2003; Costa 2008) and the time intervalsthat have been documented from other caldera-relateddeposits.

Petrogenetic or volcanologic study of the southernGMD is beyond the scope of this contribution. Our maincontributions are new data on the structural style offaults and the age of Cenozoic volcanic units. The firsttopic is based on a new geologic map scale of 1:20,000,with certain key zones being mapped at a scale of1:5000. During fieldwork, we emphasized acquisition ofstructural data, primarily strike, and dip of faults andbeds. For reconstructing volcanic events, we obtainedthe ages of stratigraphic units using two methods: U–Pblaser ablation analysis of zircons and Ar–Ar in sanidineor plagioclase. Combining these methods, we obtainedinformation about the ages of magma crystallizationand of the emplacement of volcanic rocks that supportthe existence of a volcanic centre in the GMD. With thisstudy, we contribute to understanding the geologicalevents that prepared the area for mineralization underfavourable structural conditions.

Analytical techniques

Petrography

Datable minerals and alterations were documentedthrough field observations, petrography, and electronmicroprobe analyses. Plagioclase and sanidine sepa-rated minerals were verified using scanning electronmicroscopy with energy dispersive X-ray spectroscopyat the Laboratorio de Geofluidos in the Centro deGeociencias, Universidad Nacional Autónoma deMéxico (UNAM), in Juriquilla, Queretaro (CGEO).

Mineral separation

Twelve samples of rocks and veins were analysed. Thesamples were crushed, powdered, and sieved (200 to 50mesh). Heavy mineral fractions with zircons wereobtained by panning. The non-magnetic fraction wasseparated using a Cook isodynamic magnet. The finalmineral separates (zircons, plagioclase, and sanidine)were hand-picked under a binocular microscope.Zircons were mounted in epoxy resin together with astandard (National Institute of Standards andTechnology (NIST)), and subsequently polished. Laserablation target points were selected after cathodolumi-nescence sample recognition in order to identify zirconcores and re-growth zones. Target points of 20–30 were

Figure 2. Geologic map of the study area. The location of the map within the Guanajuato Mining District is shown in Figure 1.Sections A–A’ and B–B’ indicated in the map correspond to Figure 8.

248 Á. F. NIETO-SAMANIEGO ET AL.

selected in the most external zones of zircons for datingthe latest crystallization phases.

U–Pb age dating

All the zircon samples were analysed at the laser ablationsystem facility at Laboratorio de Estudios Isotópicos (LEI)of the Centro de Geociencias, UNAM. The laboratoryconsists of a Resonetics excimer laser ablation worksta-tion (Solari et al. 2010), coupled with a Thermo XII-Seriequadrupole inductively coupled plasma mass spectrome-try (ICPMS). The laser ablation workstation operates a193 nm excimer laser (Lambda Physik, model CoherentLPX 200, Boston, MA, USA), and an optical systemequipped with a long working-distance lens with50–200 μm focus depth. The ablation cell is He-pressur-ized capable of fast signal uptake and washout. The sizeof the drill employed during this work was 30 μm, andthe drill depth during the analysis is about 20–25 μm, fora total mass ablated during each analysis of ~70–80 ng.

Seventeen isotopes were scanned during each ana-lysis. This allows a quantitative measurement of thoseisotopes necessary for U–Pb dating (lead, uranium, andthorium), as well as detailed monitoring of major andtrace elements such as Si, P, Ti, Zr, and rare earth ele-ments, which can yield important information on thepresence of microscopic inclusions into the zircons (e.g.monazite, apatite, or titanite) that can produce erro-neous age results. For each analysis, 25 s of signal back-ground are monitored, followed by 30 s of signal withlaser firing with a frequency of 5 Hz and an energydensity of ~8 J/cm2 on the target. The remnants (25 s)are employed for washout and stage repositioning. Anormal experiment involves the analysis of natural zir-con standards, as well as standard glasses. NIST stan-dard glass analyses are used to recalculate the zircons Uand Th concentrations. Repeated standard measure-ments are in turn used for mass-bias correction, aswell as for calculation of down-hole and drift fractiona-tions. We used the Isoplot 3.71 program (Ludwig 2008)for the zircon selection and calculation of ages anderrors. For details of laboratory and methodology seeSolari et al. (2010) and Solari and Tanner (2011).

40Ar–39Ar age dating

The 40Ar–39Ar experiments were conducted at Centrode Investigación Científica y de Educación Superior deEnsenada (CICESE) in Ensenada, México. Samples andmonitors were irradiated in the U-enriched researchreactor of McMaster University in Hamilton, Ontario,Canada. The mineral fragments were covered with aCd liner to block thermal neutrons. Sanidine Fish

Canyon Tuff (FCT) (28.201 ± 0.046 Ma, Kuiper et al.2008) was used as an irradiation monitor. Upon irradia-tion, sanidine FCT was fused in one step to calculate Jand the samples were step-heated. Samples and moni-tors were heated using a Coherent Innova 300 Ar-ionlaser. The argon isotopes were analysed using a VG-5400 type mass spectrometer. The argon measurementswere preceded by the blank determination of all theargon masses. The argon isotopes were corrected formass discrimination, Ca, K, and Cl neutron-inducedinterference reactions. The constants recommended bySteiger and Jäger (1977) were used in all the calcula-tions, while all the straight lines were calculated withthe equations presented in York et al. (2004).

40K–40Ar age dating

40K–40Ar analysis was conducted at the Instituto deGeología, UNAM. The sample was crushed manuallyusing a mortar and then was washed under acetone.The K content was measured by X-ray fluorescence on50 mg aliquots using a specific regression for measuringK in K–Ar samples (Solé and Enrique 2001). The analy-tical precision was better than 2%. Samples weighingbetween 1 and 2 mg were degassed under highvacuum at ~150°C for 12 h before analysis in order toreduce atmospheric contamination. Argon wasextracted by the total sample fusion using a 50 W CO2

laser defocused to 1–3 mm diameter. The evolvedgasses were mixed with a known amount of 38Ar spikeand purified with a cold finger immersed in liquid nitro-gen and two Società Apparecchi Elettrici e Scientificigetters in a stainless steel extraction line.Measurements were done in a static vacuum modeusing an MM-1200B mass spectrometer using electro-magnetic peak switching controlled by a Hall probe.The analytical precision for 40Ar and 38Ar peak heightswas >0.2%, and >0.6% for 36Ar. The data were calibratedwith the reference materials B4 M muscovite (18.5 Ma)and LP-6 biotite (128 Ma). All ages were calculated usingthe constants recommended by Steiger and Jäger(1977). A detailed description of the procedure andcalculations are given in Solé (2009).

Stratigraphy of the Cenozoic volcanic cover

Typical exposures of six volcanic units above the basalGuanajuato conglomerate (Losero Formation, BufaFormation, Calderones Formation, Cedro Formation,Chinchinadaro Formation, and Cañada La VirgenIgnimbrite) of the southern GMD are shown inFigure 3. The stratigraphic sequence is summarized inFigure 4 and discussed in a greater detail below.

INTERNATIONAL GEOLOGY REVIEW 249

Losero Formation (Plo)

Guiza (1949) named this unit ‘La Bufa sandstone’,Edwards (1955) used the term ‘losero tuff’, andEchegoyen-Sánchez et al. (1970) defined this unit more

formally as the Losero Formation. The Losero Formationis widely distributed in the GMD, and there are goodoutcrops near the El Carrizal fault (Figure 2).

The Losero Formation is a tuffaceous sandstone,which is easy to recognize because it exhibits brown,

Figure 3. Field photographs of the Cenozoic rocks of the southern Guanajuato Mining District. (A) Photograph view to the north; theGuanajuato Conglomerate rests under the Losero Fm. beneath a minor angular unconformity; the Bufa Fm. rests conformably on theLosero Fm. The three units dip NE. (B) Example of sequence repetition due normal faulting: The Bufa Fm. has Calderones Fm. to bothsides; in the right part of the image the contact is concordant and in the left part is a normal fault. (C) Example of normal faultdipping SW, fault displacement produced a clockwise rotation of the volcanic units. (D) Dike of the Cedro Fm. intruding Bufa andCalderose formations. (E) One of the best outcrops of the megabreccia near ‘Tajo de Dolores’, located 1.5 km to the northwest of ElCubo; the lithic fragments are angular to sub-angular, with diameters varying from 1 to 80 cm; clasts are mainly Mesozoic rocks, theGuanajuato Conglomerate and andesite; the matrix is tuffaceous sandy material. (F) Panoramic view of the Chichíndaro Rhyoliteresting on the andesite of the Cedro Fm; the arrow indicates a rhyolitic dome structure; in the centre of the image is a dike feedingthe Cedro Fm.

250 Á. F. NIETO-SAMANIEGO ET AL.

green, and pink colours. This formation contains manyangular to sub-angular lithic fragments and abundantcrystals. In most of the observed outcrops, three hori-zons can be distinguished. The upper and lower hori-zons commonly show inverse grading, whereas themiddle part shows normal grading. The most abundantcrystals are plagioclase and quartz; pyrite is common,and pumice fragments were observed in some samples.Fragments are approximately the size of coarse ash(0.032–2 mm) within a green or a red matrix that con-tains oxidized ferromagnesian minerals with the size offine ash (<0.032 mm). A detailed study of this unit wasconducted by Puy-Alquiza et al. (2014), who identifiedtwo members that are mainly recognizable from theircolour: the lower member is litharenite, ~5 m thick,formed of brownish red sandstone and mudstone; theupper member is arkose, formed of brownish red andgreen sandstone and mudstone ~15 m thick. Puy-Alquiza et al. (2014) proposed that the LoseroFormation is of sedimentary, not volcanic, origin, andthese authors consider that this origin represents thedistal facies of the underlying Guanajuatoconglomerate.

In the studied outcrops, the Losero Formation over-lies the Guanajuato conglomerate by a minor angularunconformity and conformably underlies the BufaFormation (Figure 3A). The thickness varies from 5 to

30 m, with thicker zones located near the El Carrizalfault (Figure 2).

There are no isotopic ages of this unit. The strati-graphic position indicates an Eocene age because it liesabove the Guanajuato conglomerate whose age is49.3 ± 1 Ma (K/Ar, whole rock), obtained from interca-lated lava flow (Aranda-Gómez and McDowell 1998),and below the Bufa Formation with an age of33.53 ± 0.48 Ma (Ar/Ar in K-feldspar, this work,Supplementary Data Table 1; see http://dx.doi.org/10.1080/00206814.2015.1072745) (Figure 4).

Bufa Formation (Pbu)

The name Bufa Rhyolite was proposed by Echegoyen-Sánchez et al. (1970), but other authors have used thename Bufa Formation (Davis 2005; Davis et al. 2009).The best outcrops are located near the town ofCalderones (Figure 2). In the northern and easternparts of the GMD, the thickness of the Bufa Formationdiminishes gradually, and it has not been reported out-side of the GMD. This formation is formed of pyroclasticflows with at least three main horizons.

The basal horizon is formed of highly welded ignim-brites, pink to beige coloured, which developed coolingfractures that sometimes form columnar prisms.Ignimbrites have fiamme structures; the phenocrysts

Figure 4. Stratigraphic column of the southern Guanajuato Mining District. Mesozoic units that crop out near the study area wereomitted. Solid circles indicate the stratigraphic position of the dated samples. s, sandstone; c, conglomerate; vc, volcaniclasticdeposits; wi, welded ignimbrite; bx, breccia; ig, ignimbrite; If, lavas; dm, dome structures. The ages come from: ***Aranda-Gómezand McDowell (1998), **Cerca-Martínez et al. (2000), *Nieto-Samaniego et al. (1996); dates without asterisks are from this study.

INTERNATIONAL GEOLOGY REVIEW 251

are quartz, sanidine, and plagioclase, which reach 10%of the rock volume. The groundmass is fine-grainedwith collapsed pumice and, in general, appears to bere-crystallized forming a cryptocrystalline material. Thishorizon contains lithic fragments that may reach 10 cmin diameter.

The second horizon is brown- to pink-colouredignimbrite. This horizon contains angular to sub-angularlithic fragments of andesite and ignimbrite with dia-meters <1 cm. There are euhedral to subhedral pheno-crysts of quartz, sanidine, plagioclase, and biotite. Thepumice is moderately to lightly collapsed and welded.The groundmass is fine-grained and commonly presentsan incipient argillic alteration.

The third horizon does not appear in all of the out-crops of the Bufa Formation. This horizon is formed ofpumice and ash deposits, and shows normal gradation.The particle size varies from fine to coarse ash. Under apetrographic microscope, it can be observed that weld-ing is poor or absent, crystals show grain-to-grain con-tacts, with microlitres of quartz, plagioclase, andpumice. There are brown to dark lithic fragments; anumber of other fragments are chloritized, and a num-ber exhibit sericite as alteration.

The Bufa Formation concordantly overlies the LoseroFormation (Figure 3A). The thickness was estimated tobe 360 m by Randall et al. (1994); we estimate thethickness to be ~400 m, measured in the outcrop tothe east of the El Coronel hill (Figure 2). Randall et al.(1994) and Davis (2005) considered this unit as an intra-caldera deposit due to its thickness, lithology, andrestricted distribution. Our field observation and mapsof surrounding areas (Martínez-Reyes 1992; Nieto-Samaniego 1992; Cerca-Martínez et al. 2000; Nieto-Samaneigo et al., 2012) confirm that this unit is absentoutside of the GMD.

Gross (1975) reported an age of 37 ± 3 Ma; this age issuspect because it has large error, was obtained from K–Ar analysis of a whole rock sample, and the locality ofthe sample is unknown. We obtained an age of33.53 ± 0.48 Ma for the Bufa Formation (Ar–Ar, plateauage from sanidine, sample G-2, Figure 5A). The datedsample comes from the second horizon (see Figure 2 forthe sample location).

Calderones Formation (Pca)

This name was introduced by Echegoyen-Sánchez et al.(1970). These authors described the unit as ‘sandstone,conglomerate, breccia, and tuff, of andesitic composi-tion, with some intercalations of dacitic tuffs’. Daviset al. (2009) considered the Calderones Formation asan intra-caldera deposit and divided it into four

members. The lower member is a megabreccia asso-ciated with the caldera margins; the second member isa coarse reddish to brown volcanic sandstone; the thirdmember is a fine green volcanic sandstone; and theupper member is a massive pyroclastic flow deposit.

The Calderones Formation outcrops in the southernpart of the GMD. There are many good outcrops wherethe Calderones Formation lies over the Bufa Formation,for example, near the towns Calderones, La Rosa deCastilla, and El Cubo (Figure 2). The CalderonesFormation comprises at least five members. The first isthe megabreccia member that was previously described(Randall et al. 1994; Davis et al. 2009) and consists ofbreccia with clasts that may reach 10 m in diameter to afine conglomerate. The clasts are mainly of Mesozoicrocks, as well as of the Guanajuato conglomerate.

The second member has a thickness of ~50 m and isformed of a pyroclastic deposit that shows cross-cuttingstratification and contains pumice fragments, com-monly at the base of a lapilli horizon. This memberhas a well-defined gradation, containing rounded tosub-rounded lithic fragments that appear to be similarto those in the Bufa Formation; the crystals are plagio-clase, sanidine, and quartz. Good outcrops can beobserved around La Rosa de Castilla (Figure 2).

The third member is a polymictic conglomerate thatcrops out in the northern part of study area. The clastsare angular to subangular and consist of volcanic rocksand lutite. This member also contains conglomerateclasts that are very similar to those in the GuanajuatoConglomerate. The size of the clasts varies from 1 to20 cm and they are supported by a fine-grained matrixthat shows a chloritic alteration. Thickness was esti-mated to be 130 m along the Guanajuato-Santa Rosaroad, which is located to the north, outside of the studyarea (Labarthe-Hernández et al. 1995).

The fourth member is formed of pyroclastic deposits,mainly ignimbrites. The ignimbrites are massive, and nosorting of clasts is observed. There are some air falldeposits, which are sorted with normal gradation. Thismember has a variable thickness, with a maximum esti-mated to be 50 m. The lithic fragments are angular tosub-angular ignimbrites varying from 0.1 to 10 cm.There is pumice supported in a fine-grained matrix;commonly, both pumice and the matrix appear to bechloritized. In general, these rocks contain K-feldspar +plagioclase > quartz. Very good outcrops can beobserved along the road from Calderones to LasGachas (Figure 2).

Around the town Carboneras, there are andesitic lavaflows that in certain places rest on the Bufa Formation,while in other places the flows appear to be intercalatedbetween pyroclastic deposits of the Calderones

252 Á. F. NIETO-SAMANIEGO ET AL.

Formation (Figure 2). We propose these andesitic rocksas a fifth member of the Calderones Formation. The lavaflows are massive, with thicknesses of 0.5–4 m. Thephenocrysts are mainly andesine and biotite; therewere some hornblende crystals with reabsorbed mar-gins. The matrix is formed of andesite and biotite micro-crystals. Chlorite appears as alteration of biotitephenocrysts and in the matrix.

The Calderones Formation rests on the BufaFormation and below the Cedro Formation (Figures3B–D and 4). The estimated thickness is approximately700 m in the El Cubo-Calderones zone. We used twosamples to date this unit; both were collected near theLas Gachas fault (Figure 2). From sample G-5, weobtained a sanidine Ar–Ar plateau age of 31.33 ±0.29 Ma (Figure 5B, Supplementary Data Table 2). Thisdate represents the deposition time of the fourth

member described above. From sample G-1, weobtained an age of 31.84 ± 0.27 (mean age, U–Pb laserablation inductively coupled plasma mass spectrometry(LA-ICPMS) in zircon), which we interpret as the finaltime of crystallization of the zircons in the magmaticreservoir (Figure 6). Although the two ages differslightly, they are indistinguishable considering the ana-lytic errors.

Cedro Formation (Pce)

Guiza (1949) described this unit as ‘Andesite’.Echegoyen-Sánchez et al. (1970) proposed the name‘Cedro Formation’. The rocks of the Cedro Formationcrop out widely in the southern part of the GMD, andandesitic lava flows out of the district have been corre-lated with this formation (Cerca-Martínez et al. 2000).

Figure 5. (A) Dark graph shows 40Ar–39Ar age spectrum for the Bufa Formation sample (G-2). The plateau age (tp) was calculatedwith the weighted mean of fractions g–i. Fraction h was excluded in because it was very close in size to the blank; therefore, for thisfraction mass 36 is unreliable. For comparative purposes, the age obtained from the isochron is shown in the inset. In the dashedgraph, we show high content of 37Ar from Ca relative to 39Ar from K, indicating a probably mix of K-feldspar and plagioclase. (B)Dark graph is the 40Ar–39Ar age spectrum for Calderones Formation sample (G-5). The plateau age (tp) was calculated with theweighted mean of fractions f–h, with 52.3 % of 39Ar released. For comparative purposes, the age obtained from the isochron isshown in the inset. Analytical results are given in Supplementary Data Tables 1 and 2.

INTERNATIONAL GEOLOGY REVIEW 253

Figure 6. U–Pb isotopic ages of sampled lithostratigraphic units. Panels beneath (A) are Tera–Wasserburg diagrams for the U–Pbisotopic compositions of analysed zircons. Panels beneath (B) are age distributions of the analysed zircon ordered by age; rectanglesindicate the zircons used for age calculation; note that zircons with ages near 700 Ma are not shown. Panels beneath (C) are themean weighted ages obtained from the younger zircons, with concordance >20%, which form a coherent group with MSWD near1.0 and with mean weighted age in agreement with the stratigraphic position of the unit. Analytical results are given inSupplementary Data Table 3.

254 Á. F. NIETO-SAMANIEGO ET AL.

The Cedro Formation is formed of andesitic lava flows; itcommonly presents a massive structure, but in someoutcrops noticeable pseudo-stratification is present.Near the El Nayal hill, there are pyroclastic fall deposits.Lavas are grey or reddish-brown when fresh but com-monly appear to be darkish-green due to intense hydro-thermal alteration. The lavas show pilotaxitic texturewith 70–80% plagioclase in the mesostasis of unknowncomposition; occasional isolated crystals of augite,chlorite, epidote, and rare hornblende appear. Thelavas of the Cedro Formation rest on the CalderonesFormation and below the Chichíndaro Rhyolite(Figure 3E). A minimum thickness of 400 m was esti-mated in the El Cubo area. There are many andesiticdikes, and these are very similar to Cedro Formationlavas; the dikes cut through the Bufa, Calderones, andCedro formations but not the overlying units. We con-sider that the dikes feed the andesitic lava flows of theCedro Formation (Figure 3E–F).

Two K–Ar ages have been published by Cerca-Martínez et al. (2000). Both dates correspond to lavaflows outcropping to the southwest of the GMD, out-side of the study area. The first age is 30.6 ± 0.4 Ma(rock matrix), and the other age is 30.7 ± 0.6 Ma (wholerock). In our study area, the age of the Cedro Formationwas determined by dating two samples of andesiticlava. Sample CED-5 B (Figure 2) was collected ~500 mto the southeast of the El Nayal hill, and the obtainedage was 32.53 ± 0.18 Ma (U–Pb LA-ICPMS in zircon)(Figure 6, Supplementary Data Table 3). The other sam-ple (CED-4) was collected ~500 m to the north of theRosa de Castilla (Figure 2), and the obtained age was32.58 ± 0.21 Ma (U–Pb LA-ICPMS in zircon) (Figure 6,Supplementary Data Table 3). These note that theseages are ~1 million years older than those obtainedfrom the underlying Calderones Formation.

The U–Pb LA-ICPMS age method in zircon requiresthat analysed borders are the youngest, clean, withoutinclusions or fractures and thicker than 20 microns. Theobtained ages commonly are slightly older than thecooling ages because they represent the time of crystal-lization, which were formed before the emplacement ofthe volcanic deposit. The stratigraphic positionundoubtedly shows that the Cedro Formation isyounger than the Calderones Formation, for this reasonwe interpret the ca. 32.5 Ma age as the time of antecrystformation. The crystallization of those zircons occurredjust between the emplacement of the Bufa andCalderones formations. We think that the30.6 ± 0.4 Ma K–Ar age of Cerca-Martínez et al. (2000)may be a better age for the Cedro Formation (Figure 4).

Additionally, we dated the sample JBCAL-1 from adike that is located 1 km to the north of the town of

Calderones (Figure 2). The U–Pb LA-ICPMS ages in zir-con vary from ca. 42 to ca. 31 Ma, the younger coherentzircon group has a mean age of 32.20 ± 0.91 Ma(Figure 6). Our interpretation is that this age representsthe final time of antecrystal zircon crystallization.

Chichíndaro rhyolite (Pch)

This name was used by Echegoyen-Sánchez et al. (1970)to refer to ‘rhyolite lavas with interbedded breccia andpyroclastic deposits’. These rocks cover wide extensionsoutside of the study area (Nieto-Samaniego et al. 1996,2007) (Figure 2). The Chichíndaro Rhyolite is mainlyformed of rhyolitic domes and lava flows, which arelocally associated with pyroclastic deposits, brecciasand vitrophyres. Domes commonly have lavas with aflow structure in concentric lobes from the emissioncentre. In the borders of the domes, or intercalatedwith lava flows, there are pyroclastic deposits. It is com-mon to find vitrophyres at the base of these lava flows.Rhyolitic lavas are white or pink and show a porphyritictexture with phenocrysts of quartz and sanidine ≫ oli-goclase. There are lavas with abundant biotite, butbiotite is not present in every case. The matrix is cryp-tocrystalline with some vitric zones. Pyroclastic depositsare mainly small surge-type and fall-ash deposits; it isalso common to find breccias associated with thedomes or lava flows.

The Chichíndaro Rhyolite rests on the CedroFormation (Figure 3E) and below the Cañada La Virgenignimbrite. There are rhyolitic dikes that cut units olderthan the Chichíndaro Rhyolite; these dikes fed conduitsof the domes and lava flows. Rhyolite domes wereemplaced in NW–SE faults (Davis et al. 2009). This situa-tion can be observed in the Rosa de Castilla area whereChichíndaro Rhyolite domes were emplaced along afault parallel to the El Cubo-Villalpando fault. Also, inthe southern part of the La Leona fault, there is a domeof Chichíndaro Rhyolite covering the fault trace(Figure 2). The Veta Madre fault disappears abruptly tothe south of the El Carrizal fault where the ChichindaroRhyolite crops out extensively. The Veta Madre is wellexposed 500 m NE of the intersection of these faults,displacing the Calderones Formation a minimum of150 m and producing bed tilts of 27° to the NE. Burialof the Veta Madre fault below the Chichíndaro Rhyoliteexplains their absence to the south of the fault inter-section (Guiza 1949; Nieto-Samaniego 1992).

The thickness of the Chichíndaro Rhyolite is variableand difficult to estimate because the base is onlyobserved in more eroded exposures. However, it hasbeen estimated to be approximately 400 m for thethicker zones (Nieto-Samaniego 1992).

INTERNATIONAL GEOLOGY REVIEW 255

There are three published ages for the ChichínaroRhyolite, 32 ± 1 Ma (K–Ar, whole rock, Gross 1975) and30.1 ± 0.8 Ma and 30.8 ± 0.8 Ma (K–Ar in sanidine, Nieto-Samaniego et al. 1996). In the present study, we datedthe Chichíndaro Rhyolite from a sample that was col-lected in the type locality Chichíndaro hill (sample RCHI-2), located 1 km to the north of the study area. Theobtained age is 30.36 ± 0.4 Ma (U–Pb LA-ICPMS inzircon) (Figure 6), which represents the final time ofzircon crystallization in the magma reservoir.

Cañada la Virgen ignimbrite (Pov)

This name was proposed by Nieto-Samaniego et al.(1996) for a pyroclastic unit that crops out in the SanMiguel de Allende region, 50 km east of the study area.Within the study area, there are only two outcrops,which are located in the Tacubaya area (Figure 2). Thisunit is formed of highly to moderately welded ignim-brites, with some associated ash-fall deposits. Theignimbrites are massive, with abundant crystals of sani-dine >> plagioclase > quartz. This unit contains pumicefragments no larger than 2 cm, which commonly formfiamme structures; there are also scarce lithic fragmentsof less than 5 cm in size.

The Cañada la Virgen ignimbrite unconformablyoverlies the Chichíndaro Rhyolite, and there are noyounger units in the study area that cover it. An ageof 28.6 ± 0.7 Ma (K–Ar in sanidine) was obtained byNieto-Samaniego et al. (1996). In this study, we collectedsample M-1 and obtained an age of 29.44 ± 0.19 Ma (U–Pb, LA-ICPMS in zircon) (Figure 6).

Structure of the southern GMD

In the study area, bedding dips decrease with time (upthe stratigraphic column). This observation suggestsactive faulting during deposition. In the Guanajuatoconglomerate, there are two preferred orientations ofbeds: dominant 000–020°/25–35° E (strike/dip) and sub-ordinate 330–040°/10°–40° W (Figure 7A). In theCalderones Formation, the beds show one well-definedpreferred orientation with a mean 161°/26° E (strike/dip)NE of Las Gachas fault (Figures 2 and 7B); to the south-east of the Las Gachas fault the Calderones Formationdips SE like the Guanajuato Conglomerate. The otherunits do not show good bedding. Two characteristicsare noticeable from the bed attitudes: (1) beds of theGuanajuato Conglomerate and Calderones Formationshow two opposite dip directions, forming an anti-form-like structure, with the axis in the southwesternpart of the GMD (Figure 8); (2) in the A–A’ and B–B’sections (Figure 8), to the east of the Las Gachas fault,the beds dip NE, indicating a general clockwise rotationalong the sections.

The main structures in the southern GMD are sets ofnormal faults and dikes. The major faults comprise seg-ments that show that incipient or no-linkage systemsare the El Cubo-Villalpando, La Leona, Veta Madre, andLas Gachas faults. All of these faults trend N30°–40° Wand form the El Cubo and Veta Madre grabens (Daviset al. 2009). There is a horst between these grabensshowing some strike-slip offset. We interpret this asevidence of a minor right lateral component in theVeta Madre and La Leona normal faults (Figure 2).There are some direct measurements of displacements

Figure 7. (A) Contours of stratification planes from the Guanajuato Conglomerate. There are two concentrations: one set of planesdip E and the other planes dip W, these groups reflect the antiform structure shown in the sections of Figure 6. (B) Contours ofbedding planes from the Calderones Formation; notice the homogeneous and well-grouped orientation data reflecting a unique axisof rotation associated with the NW-trending fault system.

256 Á. F. NIETO-SAMANIEGO ET AL.

in major faults: Gross (1975) reported a throw of ~200 mfor the El Cubo-Villalpando fault zone and 1400 m ofdip-slip for the Veta Madre fault. Displacements alongthe La Leona and Las Gachas faults has not been mea-sured but considering the topographic escarpments, thethickening and apparent displacements of lithostrati-graphic units, as well as the tilting of bedding, weestimate dip-displacements of ~200 m for the LaLeona fault and ~350 m for the Las Gachas fault(Figure 8). The NE–SW El Carrizal fault corresponds tothe ‘northern fault’ of the La Sauceda graben (Nieto-Samaniego 1992). This is a normal fault, 6 km long, withstrike varying N60°–70° E and dipping 65°–80° SE. Thereare few outcrops of the fault plane and slickenlines insome show a minor left lateral component. The ElCarrizal fault cuts the La Leona, Veta Madre, and LasGachas faults, indicating that it is younger (Figure 2). Wemapped 17 mafic dikes; these are 1–3 km long and2–10 m wide, trending ENE–WSW, nearly parallel tothe El Carrizal fault. Some of these feed the lava flowsof the Cedro Formation (Figure 2).

During fieldwork, a large amount of vein, fault, andslickenline measurements were collected. The vein andfault data show mean orientations that are nearly parallelto the major faults (Figures 2 and 9). Almost all of theobserved faults are normal or oblique, with rare excep-tions including purely lateral faults, which we did notinclude in Figure 9C. A palaeostress tensor that wascalculated using the right dihedron method is shown inFigure 9D (Delvaux and Sperner 2003). The minimumcompressional principal stress is oriented 14°/242° (dip/trend), which agrees with the distribution of slickenlines(Figure 9C), indicating that the main extension wasoriented NE–SW. However, a secondary extension

direction oriented near NW–SE is inferred from the pre-sence of the dike system and the El Carrizal fault(Figure 2). These two extension directions have beenreported in the MC and were interpreted as products ofCenozoic deformation (Nieto-Samaniego et al. 1997,1999).

The general structure of the southern GMD is illu-strated in sections A–A’ and B–B’ (Figure 8). Masterfaults of the El Cubo and Veta Madre half-grabens dipSW, and in both structures beds of the CalderonesFormation dip NE indicating that rotation of the fault-limited blocks responded to fault displacements. Thestrong control of major faults on bed tilting and outcroppatter suggest that they are more probably related tonormal faulting than to a caldera collapse. We proposethat the southern GMD structure is formed of listric SW-dipping normal faults that could be linked at depth(Figures 2 and 8).

To the northeast of the Las Gachas fault, the mean dipof the Calderones Formation is 25° NE, whereasGuanajuato conglomerate beds dip 25°–35° to the E orENE (Figure 7). A similar magnitude of dips indicates thatmuch of the tilting occurred during or after deposition ofthe Calderones Formation. The main GMD faulting epi-sode occurred before eruption of the ca. 29 MaChichíndaro Rhyolite because this buried the major faults.A good example is the Veta Madre fault; the vein-fillminerals were dated by Gross (1975) at 29.2 ± 2.0,28.3 ± 5.0, and 30.7 ± 3.0 Ma (K–Ar in adularia).Therefore, the main fault activity occurred before 30 Ma.We dated an adularia sample from the La Valenciana Minelocated in the central zone of the Veta Madre, not faroutside the mapped area. The obtained age of29.8 ± 0.8 Ma (Supplementary Data Table 4) agrees with

Figure 8. Structural sections of the study area. Location is shown in the geologic map of Figure 2.

INTERNATIONAL GEOLOGY REVIEW 257

ages reported by Gross (1975) and with the field observa-tion of the Chichíndaro Rhyolite filling or burying the VetaMadre fault. Our data document that the peak activity ofthe Veta Madre fault ceased a little before 30 Ma.

Discussion

Cenozoic volcanic activity in the study area lasted ~3 mil-lion years from ca. 33.5 to ca. 29.4 Ma. The Bufa,Calderones, and Cedro formations have similar ages,their spatial distribution is restricted to the GMD, andthey were emplaced coeval with the peak of normalfaulting. These characteristics strongly suggest thatthese rocks were produced by a volcanic centre locatedin the southern part of the district during extension. Incontrast, the Chichindaro Rhyolite has a few outcropswithin the southern GMD. It mainly appears in surround-ing areas and is widely distributed in the MC (Nieto-Samaniego et al. 2007). Furthermore, the ChichindaroRhyolite was emplaced after the main fault activity ofthe southern GMD. The Cañada La Virgen ignimbrite alsoappears mainly outside of the GMD. At the margins ofthe district, these rocks were laterally deposited on

topographic highlands, suggesting that the source ofthe Cañada La Virgen ignimbrite was located outside ofand probably far away from the study area.

We infer that an Early Oligocene volcanic centreexisted in the study area, which erupted the Bufa,Calderones, and Cedro formations, and that magmacompositions evolved from felsic to mafic. Within thesouthern GMD, outcrops of the Chichíndaro Rhyolitecould be related to this volcanic centre, but correspondbetter to the well-documented regional rhyolitic mag-matism (Nieto-Samaniego et al. 1996, 2007) that mostlycrops out beyond the GMD. We argue the Cañada LaVirgen Formation is unrelated to the southern GMDvolcanic centre because it mainly crops far away anddoes not occur within the GMD.

To determine volcanic rock emplacement ages, zirconanalysis concentrated near the margins and we preferrednon-zoned crystals. In this way, younger zircon ages repre-sent the last crystallization in themagma chamber prior toeruption. The upper units most likely have zircons thatregister residence time in the magma chamber, as well aszircon xenocrysts from older rocks. When viewed as awhole, the zircon ages group from ca. 42 to ca. 30 Ma.

Figure 9. Contours of veins (A) and faults (B) measured in the study area. In both, there is a clear concentration corresponding toNW–SE-trending structures. Faults show a secondary concentration with NE–SW trend. In (C), the contours of the slickenlines weremeasured in faults and veins, and in (D), a palaeostress-reduced tensor was calculated from faults in which the sense of movementwas identified. The distribution of slickenlines and the palaeostress indicate the main extension direction oriented NE–SW. Dots arethe poles of planes. Palaeostress inversion was performed using the Win-Tensor 4.04 program (Delvaux and Sperner 2003).

258 Á. F. NIETO-SAMANIEGO ET AL.

We interpret this interval as the duration of the magmaticepisode. Eruption of GMD volcanics took place from ca. 33to ca. 32 Ma, lasting about 1 million years, which agreeswith the interval ages of the deposits from other nearbyMexican calderas, for example, Amazcala (Aguirre-Díazand Martínez-López, 2001), Amealco (Aguirre-Díaz 1996)and Tilzapotla (Morán-Zenteno et al. 2004).

Two directions of extension were identified in thesouthern GMD: one direction is recorded by NW–SE-trending normal faults that were active during eruptionof the Bufa, Calderones, and Cedro formations, indicat-ing the NE–SW extension. The other direction was theNW–SE-directed extension recorded by dikes and nor-mal faults; the second extension began about the sametime as eruptions to form the Cedro Formation andcontinued after Chichíndaro Rhyolite eruptions. Duringthis faulting phase, the NW–SE faults were also reacti-vated (Nieto-Samaniego et al. 1997, 1999). Regionalnormal faulting predates and postdates the volcanicactivity of the southern GMD. The age of faulting inthe Sierra de Guanajuato and Sierra de Codornicesinitiated at ca. 50 Ma and extends after 14 Ma (Nieto-Samaniego, 1992; Nieto-Samaniego et al. 2012; Botero-Santa et al. 2015).

It is important to note that the obtained ages, areasof outcropping, thicknesses, and facies that observedin the Bufa, Calderones, and Cedro formations areconsistent with the existence of a caldera-like volcaniccentre in the southern GMD, which could correspondto that proposed by Randall et al. (1994) and Daviset al. (2009). However, there is no mapped ellipticalcrater-like structure in the southern GMD. The absenceof faults resembling a caldera collapse structure couldbe explained if volcanism occurred during extension toform southern GMD grabens. Rapid extension accom-modated by normal faulting and volcanism couldexplain the absence of a clear caldera structure. TheVeta Madre accommodated 1400 m of dip-slip displa-cement, whereas the Villalpando-El Cubo accommo-dated 230 m. Both structures cut the CedroFormation, unfortunately the age of this unit is notwell constrained, but taking as a proxy the age of theCalderones Formation, fault displacements occurredafter 31.33 ± 0.4 Ma and before the vein fill(29.8 ± 0.8 Ma). Rates of displacement are 0.93 mm/year for the Veta Madre fault and 0.15 mm/year for theEl Cubo-Villalpando fault. The rates of displacement innormal faults are between 0.004 and 1.0 mm/year, forfaults active ~1.5 million years is lesser than 0.1 mm/year (Nicol et al. 1997), showing that faults in the GMDhad a high displacement rates. This structure wasprobably erased by the normal faulting that spannednear 15 million years after the hypothetic caldera

episode. Another possibility, following the idea ofDavis (2005), is that the caldera could be a ‘trap-doorcaldera’ that is associated with the major GMD normalfaults. This possibility is based on the observation thatvolcanic activity overlapped in time and space withthe formation of the El Cubo and Veta Madre grabens.Although now we have more and better geologicalinformation, some problems persist for the calderahypothesis: (1) in the best case, the ages of relatedcaldera units span ~2 million years, (2) the Bufa andCalderonse formations do not crop out outside ofsouthern GMD indicating the existence of the volcaniccentre in this region, but no units has been recognizedas pyroclastic deposits outside the caldera; (3) there isno detailed study about the facies and thickness varia-tions of the volcanic units; and (4) there is no petro-genetic analyses of the volcanic rocks for determiningif they could be related to caldera evolution.

Mineral deposits of the GMD have ages close tothat of the Chichíndaro Rhyolite. K–Ar ages of Gross(1975) have large errors (30.7 ± 3.0, 29.2 ± 2.0, and28.3 ± 5.0 Ma; K–Ar in adularia), and the hydrothermalevent could have spanned between 33.7 and 23.3 Ma.We think that hydrothermal event probably spannedbetween 30.7 and 28.3 Ma. We obtained an age of29.8 ± 0.8 Ma (K–Ar in adularia) with a small error,falling within the interval 30.7–28.3 Ma and coevalwith the Chichíndaro Rhyolite (considering the errorsof isotopic ages). With this information, we proposethat mineralization responsible for the GMD world-class ore deposit was development of a caldera-likevolcanic centre coeval with normal faulting. Thisoccurred from ca. 33 to 32 Ma, produced the volcanicunits and the major faults, and formed many fracturesand breccias. The intense fracture that occurred dur-ing volcanism was essential to prepare rocks formineralization. The Las Torres mine is an emblematicmineralized breccia, consisting of a 30–40 m widezone of mineralized breccia and stockwork, locatedin the hanging-wall of the Veta Madre fault, insidethe volcanic centre and hosted by the BufaFormation (Gross 1975).

GMD hydrothermal activity probably was related tothe latest pulses of the caldera-like event. In addition,regional magmatism to produce the ChichíndaroRhyolite buried the fractured zones and allowed thehydrothermal system to last two or more million yearsafter the end of the caldera-like event.

Conclusions

A refinement of the Cenozoic volcanic stratigraphy of theGMD is presented. Explosive volcanism initiated with the

INTERNATIONAL GEOLOGY REVIEW 259

rhyolitic ignimbrite sequence of the Bufa Formation fol-lowed by the Calderones Formation, formed of threemain pyroclastic deposits of rhyolitic to andesitic compo-sition and one andesitic lava member. The CedroFormation overlies these pyroclastic units, which ismainly effusive, formed of andesite lavas with somemore mafic flows and scarce pyroclastic deposits.

The Bufa, Calderones, and Cedro formations consti-tute the here-named ‘Guanajuato volcanic group’ thatwas produced by a volcanic centre in the study area.From the zircon ages of all of the dated rocks, weinterpret that the magmatic episode initiated near42 Ma and culminates between ca. 33.53 and ca.31.33 Ma, which is the emplacement interval of theGuanajuato volcanic group (Figure 10). The previousage of the Bufa Formation of ca. 37 Ma is discarded;their age is 33.53 ± 0.48 Ma. Overlying the Guanajuatovolcanic group are two volcanic units, the ChichíndaroRhyolite and the Cañada la Virgen ignimbrite, bothwidely spanning outside of the GMD.

Normal faulting predates and postdates theGuanajuato volcanic group, indicating that it respondsto the regional tectonic regime of the MC and that it isnot due the magmatic events that took place in thesouthern GMD. The caldera hypothesis proposed in theliterature is only partially supported by our geochrono-logical data, which document that GMD volcanic activitylasted ~1 million years, which seems to be a long timefor a small caldera.

The absence of a semi-elliptic structure, as it isexpected for a classic caldera, suggests that there wasanother type of volcanic centre. With the available data,we conclude that there was a volcanic centre in thesouthern GMD, which produced the Guanajuato

Volcanic Group. This volcanic centre is part of the exten-sive Oligocene volcanism of the MC.

The geological events that produced the GD world-class ore deposit were characterized by volcanism coe-val with normal faulting to produce intense fracturingand many breccias, setting the stage for the hydrother-mal activity and mineralization. Hydrothermal activitylasted for two million years or more due to the mag-matic episode that formed the Chichíndaro Rhyolite,which buried and trapped the very permeable brecciasand faulted rocks hosting GMD mineral deposits(Figure 10).

Acknowledgements

We acknowledge the assistance by Juan Tomás Vázquez-Ramírez for making thin sections. The palaeostress-reducedtensor was obtained using Win-Tensor, a software devel-oped by Dr Damien Delvaux, Royal Museum for CentralAfrica, Tervuren, Belgium. A. Rosas prepared samples forAr–Ar analysis and M. A. García performed the Ar–Ar massspectrometry.

Disclosure statement

No potential conflict of interest was reported by the authors.

Funding

This research was funded by Fresnillo PLC, Consejo Naional deCiencia y Tecnología [80142] and Programa de Apoyo aProyectos de Investigación e Innovación Tecnológica-UNAM[IN104014].

Figure 10. Summary of Cenozoic events in the study area. A volcanic event took place under NE extension, synchronous withformation of the main NW–SE faults and grabens. The end of the volcanic event coincided with a change in the extension directionfrom NE to NW; NE-tending normal faults formed and the NE-trending dikes which feed the lavas of the Cedros Formation wereemplaced. Hydrothermal activity initiated at the end of this volcanic event and continued more than two million years, sustained byheat associated with eruption of the Chichíndaro Rhyolite.

260 Á. F. NIETO-SAMANIEGO ET AL.

References

Aguirre-Díaz, G.J., 1996, Volcanic stratigraphy of the Amealcocaldera and vicinity, Central Mexican Volcanic Belt: RevistaMexicana de Ciencias Geológicas, v. 13, p. 10–51.

Aguirre-Dı́az, G.J., and López-Martı́nez, M., 2001, The Amazcalacaldera, Queretaro, Mexico. Geology and geochronology:Journal of Volcanology and Geothermal Research, v. 111,p. 203–218. doi:10.1016/S0377-0273(01)00227-X

Alvarado-Méndez, H., López-Ojeda, J.A., and Caballero-Martínez, J.A., 1998, Carta Geológico-Minera GuanajuatoF14C43: Servicio Geológico Mexicano, Secretaría deEconomía, 1 map with text.

Aranda-Gómez, J.J., and McDowell, F., 1998, Paleogene exten-sion in the southern Basin and Range province of Mexico:Syndepositional tilting of Eocene red beds and Oligocenevolcanic rocks in the Guanajuato Mining District:International Geology Review, v. 40, p. 116–134.doi:10.1080/00206819809465201

Botero-Santa, P.A., Alaniz-Álvarez, S.A., Nieto-Samaniego, Á.F.,López Martínez, M., Levresse, G., Xu, S.-S., and Ortega-Obregón, C., 2015, Origen y desarrollo de la cuenca ElBajío en el sector central de la Faja VolcánicaTransmexicana: Revista Mexicana de Ciencias Geológicas,v. 32, p. 84–98.

Cerca-Martínez, L.M., Aguirre-Díaz, G.J., and López-Martínez,M., 2000, The geologic evolution of the southern Sierra deGuanajuato, Mexico: A documented example of the transi-tion from the Sierra Madre Occidental to the MexicanVolcanic Belt: International Geology Review, v. 42, p. 131–151. doi:10.1080/00206810009465073

Costa, F., 2008, Residence times of silicic magmas associatedwith calderas, in Gottsmann, J., and Marti, J., eds.,Developments in volcanology, IAVCEI publications, p. 1–55.

Davis, J.B., 2005, Relation of a caldera in the GuanajuatoMining District, Mexico to the distribution of magmas andmineral deposits [M.Sc. thesis]: Department of GeologicalSciences, The University of Texas at El Paso, 151 p.

Davis, J.B., Clark, K., and Randall, J.A., 2009, Relación de unacaldera con la mineralización en el distrito minero deGuanajuato, in Clark, K.F., Salas Piza, G.A., and CubillasEstrada, R., eds., Geología Económica de México:Asociación de Ingenieros de Minas, Metalurgistas yGeólogos de México, A. C., México, Servicio GeológicoMexicano, p. 584–635.

Delvaux, D., and Sperner, B., 2003, Stress tensor inversion fromfault kinematic indicators and focal mechanism data: TheTENSOR program, in Nieuwland, D., ed., Vol. 212: Newinsights into structural interpretation and modelling, geo-logical society: London, Special Publications, p. 75–100.

Echegoyen-Sánchez, J., Romero-Martínez, S., and Velázquez-Silva, S., 1970, Geología y yacimientos minerales de laparte central del distrito minero de Guanajuato: ConsejoDe Recursos Naturales No Renovables, Boletín, v. 75, 48 p.

Edwards, D.J., 1955, Studies of some early Tertiary red con-glomerates of central Mexico: United States GeologicalSurvey Professional Paper 264-H, p. 153–185.

Ferrari, L., Valencia-Moreno, M., and Bryan, S., 2007,Magmatism and tectonics of the Sierra Madre Occidentaland its relation with the evolution of the western margin ofNorth America, in Alaniz-Álvarez, S.A., and Nieto-Samaniego, Á.F., eds., Geology of México: Celebrating the

centenary of the geological society of México, GeologicalSociety of America Special Paper 422, p. 1–39. doi:10.1130/2007.2422(01)

Gross, W.H., 1975, New ore discovery and source of silver-goldveins, Guanajuato, Mexico: Economic Geology, v. 70, p.1175–1189. doi:10.2113/gsecongeo.70.7.1175

Guiza, R., 1949, Estudio geológico del distrito minero deGuanajuato: Instituto Nacional para la Investigación deRecursos Minerales: Boletín, v. 22, 75 p.

Jellinek, A.M., and DePaolo, D.J., 2003, A model for the originof large silicic magma chambers: Precursors of caldera-forming eruptions: Bulletin of Volcanology, v. 65, p. 363–381. doi:10.1007/s00445-003-0277-y

Kuiper, K.F., Deino, A., Hilgen, F.J., Krijgsman, W., Renne, P.R.,and Wijbrans, J.R., 2008, Synchronizing rock clocks of earthhistory: Science, v. 320, p. 500–504. doi:10.1126/science.1154339

Labarthe-Hernández, G., Jiménez-López, L.S., Tristán-González, M.,Mata-Segura, J.L., Aguillón-Robles, A., Almaguer-Tapia, R., andBarboza-Gudiño, J.R., 1995, Cartografía Geológica 1:25:000 dela Sierra de Guanajuato: Universidad Autónoma de San LuisPotosí, Instituto de Geología, Internal Report, 185 p.

Lapierre, H., Ortiz, L.E., Abouchami, W., Monod, O., Coulon, C.,and Zimmermann, J.L., 1992, A crustal section of an intra-oceanic island arc: The Late Jurassic-Early CretaceousGuanajuato magmatic sequence, central Mexico: Earth andPlanetary Science Letters, v. 108, p. 61–77. doi:10.1016/0012-821X(92)90060-9

López-Ojeda, J.A., Loaeza-García, J.P., and Zárate-Barradas, R.,2002, Carta Geológico-Minera Aldama F14-C53: ServicioGeológico Mexicano, Secretaría de Economía, 1 map withtext.

Ludwig, K.R., 2008, Isoplot/EX version 3.0, A geochronologicaltoolkit for Microsoft Excel: Berkeley, CA, BerkeleyGeochronology Center Special Publication. http://www.bgc.org/isoplot_etc/isoplot.html.

Martínez-Reyes, J., 1992, Mapa geológico de la Sierra deGuanajuato con resumen de la geología de la Sierra deGuanajuato: Universidad Nacional Autónoma de México,Instituto de Geología, Cartas Geológicas y Mineras 8, 1color sheet.

Martini, M., Mori, L., Solari, L., and Centeno-García, E., 2011,Sandstone provenance of the Arperos basin (Sierra deGuanajuato, central Mexico): Late Jurassic-Early cretaceousback-arc spreading as the foundation of the Guerrero ter-rane: The Journal of Geology, v. 119, p. 597–617.doi:10.1086/661989

McDowell, F.W., and Clabaugh, S.E., 1979, Ignimbrites of theSierra Madre Occidental and their relation to the tectonichistory of western Mexico, in Chapin, C.E., and Elston, W.E.,eds., Ash-flow tuffs: Boulder, CO, Geological Society ofAmerica Special Paper, v. 180, p. 113–124.

Morán-Zenteno, D.J., Alba-Aldave, L.A., Solé, J., and Iriondo, A.,2004, A major resurgent caldera in southern Mexico: Thesource of the late Eocene Tilzapotla ignimbrite: Journal ofVolcanology and Geothermal Research, v. 136, p. 97–119.doi:10.1016/j.jvolgeores.2004.04.002

Mortensen, J.K., Hall, B.V., Bissig, T., Friedman, R.M., Danielson,T., Oliver, J., Rhys, D.A., Ross, K.V., and Gabites, J.E., 2008,Age and paleotectonic setting of volcanogenic massivesulfide deposits in the Guerrero Terrane of central Mexico:Constraints from U-Pb Age and Pb isotope studies:

INTERNATIONAL GEOLOGY REVIEW 261

Economic Geology, v. 103, p. 117–140. doi:10.2113/gsecongeo.103.1.117

Nicol, A., Walsh, J.J., Watterson, J., and Underhill, J.R., 1997,Displacement rates of normal faults: Nature, v. 390, p. 157–159. doi:10.1038/36548

Nieto-Samaniego, A.F., 1992, Fallamiento y estratigrafía ceno-zoicos en la porción sudoriental de la Sierra de Guanajuato,Vol. 9: Revista, Instituto de Geología, Universidad NacionalAutónoma de México, p. 146–155.

Nieto-Samaniego, A.F., Alaniz-Álvarez, S.A., and Camprubí, A.,2007, Mesa Central of México: Stratigraphy, structure, andCenozoic tectonic evolution, in Alaniz-Álvarez, S.A., andNieto-Samaniego, Á.F., eds., Geology of México:Celebrating the centenary of the geological society ofMéxico, Geological Society of America Special Paper 422,p. 41–70, doi:10.1130/2007.2422(02)

Nieto-Samaniego, A.F., Alaniz-Alvarez, S.A., and Labarthe-Hernández, G., 1997, La deformación cenozoicaposlaramídica en la parte meridional de la Mesa Central,México: Revista Mexicana de Ciencias Geológicas, v. 14, p.13–25.

Nieto-Samaniego, Á.F., Ferrari, L., Alaniz-Alvarez, S.A., Labarthe-Hernández, G., and Rosas-Elguera, J., 1999, Variation ofCenozoic extension and volcanism across the southernSierra Madre Occidental Volcanic Province, Mexico:Geological Society of America Bulletin, v. 111, p. 347–363.doi:10.1130/0016-7606(1999)111<0347:VOCEAV>2.3.CO;2

Nieto-Samaniego, A.F., Macías-Romo, M.C., and Alaniz-Alvarez,S.A., 1996, Nuevas edades isotópicas de la cubiertavolcánica cenozoica de la parte meridional de la MesaCentral, México: Revista Mexicana de Ciencias Geológicas,v. 13, p. 117–122.

Nieto-Samaniego, A.F., Ojeda-García, A.C., Alaniz-Álvarez, S.A.,and Xu, S.-S., 2012, Geología de la región de Salamanca,Guanajuato, México: Boletín De La Sociedad GeológicaMexicana, v. 62, no. 3, p. 411–425.

Ortiz-Hernández, L.E., Acevedo-Sandoval, O.A., and Flores-Castro, K., 2003, Early Cretaceous intraplate seamounts

from Guanajuato central México: Geochemical and miner-alogical data: Revista Mexicana de Ciencias Geológicas, v.20, no. 1, p. 27–40.

Puy-Alquiza, M.J., Miranda-Aviles, R., Cruz-Cruz, M., Pérez-Arbizu, O., Vega-González, M., and Ana-Zanor, G., 2014,Geochemistry and depositional environment of the LoseroFormation in the Mesa Central, México: Boletín De LaSociedad Geológica Mexicana, v. 66, p. 413–430.

Randall, J.A., Saldaña, A., and Clark, K.F., 1994, Exploration ina volcano-plutonic center at Guanajuato, Mexico:Economic Geology, v. 89, p. 1722–1751. doi:10.2113/gsecongeo.89.8.1722

Solari, L.A., Gómez-Tuena, A., Bernal, J.P., Pérez-Arvizu, O., andTanner, M., 2010, U-Pb Zircon Geochronology with anIntegrated LA-ICP-MS Microanalytical Workstation:Achievements in precision and accuracy: Geostandardsand Geoanalytical Research, v. 34, p. 5–18. doi:10.1111/j.1751-908X.2009.00027.x.

Solari, L.A., and Tanner, M., 2011, UPb age, a fast data reduc-tion script for LA-ICP-MS U-Pb geochronology: RevistaMexicana de Ciencias Geológicas, v. 28, p. 83–91.

Solé, J., 2009, Determination of K-Ar ages in milligram samplesusing an infrared laser for argon extraction: RapidCommunications in Mass Spectrometry, v. 23, p. 3579–3590. doi:10.1002/rcm.4280

Solé, J., and Enrique, P., 2001, X-ray fluorescence analysis forthe determination of potassium in small quantities of sili-cate minerals for K-Ar dating: Analytica Chimica Acta, v. 440,p. 199–205. doi:10.1016/S0003-2670(01)01060-1

Steiger, R.H., and Jäger, E., 1977, Subcommission on geochro-nology: Convention on the use of decay constants in geoand cosmochronology: Earth and Planetary Science Letters,v. 36, p. 359–362. doi:10.1016/0012-821X(77)90060-7

York, D., Evensen, N.M., López-Martínez, M., and DeBasabe-Delgado, J., 2004, Unified equations for the slope,intercept, and standard errors of the best straight line:American Journal of Physics, v. 72, p. 367–375.doi:10.1119/1.1632486

262 Á. F. NIETO-SAMANIEGO ET AL.