Vous êtes sur la page 1sur 23
Journal of Volcanology and Geothermal Research 146 (2005) 284 – 306 www.elsevier.com/locate/jvolgeores Geology,
Journal of Volcanology and Geothermal Research 146 (2005) 284 – 306 www.elsevier.com/locate/jvolgeores Geology,

Journal of Volcanology and Geothermal Research 146 (2005) 284 – 306

Volcanology and Geothermal Research 146 (2005) 284 – 306 www.elsevier.com/locate/jvolgeores Geology, geochronology

www.elsevier.com/locate/jvolgeores

Geology, geochronology and tectonic setting of late Cenozoic volcanism along the southwestern Gulf of Mexico: The Eastern Alkaline Province revisited

Luca Ferrari a, T , Takahiro Tagami b , Mugihiko Eguchi b , Ma. Teresa Orozco-Esquivel a , Chiara M. Petrone c , Jorge Jacobo-Albarra´n d , Margarita Lo´ pez-Martı´nez e

a Centro de Geociencias, Universidad Nacional Auto´noma de Me´xico, Campus Juriquilla, Qro. Apdo. Postal 1-742, Centro, 76000 Quere´taro, Qro., Mexico b Department of Geology and Mineralogy, Division of Earth and Planetary Sciences, Kyoto University, Japan c Dipartimento Scienze della Terra, Universita´ degli Studi di Firenze, Italy d Instituto Mexicano del Petro´leo, Mexico D.F., Mexico e Departamento de Geologı´a, Centro de Investigacio´n Cientı´fica y Educacio´n Superior de Ensenada, Ensenada, Baja California, Mexico

Received 22 March 2004; received in revised form 10 February 2005; accepted 28 February 2005

Abstract

A NNW-trending belt of alkaline mafic volcanic fields parallels the Gulf of Mexico from the U.S. border southward to Veracruz state, in eastern Mexico. Previous studies grouped this volcanism into the so-called bEastern Alkaline Province Q (EAP) and suggested that it resulted from Gulf-parallel extensional faulting migrating from north to south from Oligocene to Present. On the basis of new geologic studies, forty-nine unspiked K–Ar and two 40 Ar– 39 Ar ages, we propose a new geodynamic model for the volcanism along the southwestern Gulf of Mexico. We studied in detail four of the six recognized fields of mafic alkaline volcanism in Veracruz state: 1) The lavas flows of Tlanchinol area (7.3–5.7 Ma), 2) the Alamo monogenetic field and Sierra de Tantima (7.6–6.6 Ma), 3) the Poza Rica and Metlatoyuca lava flows (1.6–1.3 Ma) and 4) the Chiconquiaco–Palma Sola area (6.9–3.2 Ma). Other two mafic volcanic fields may represent the continuation of alkaline volcanism to the southeast: the Middle Miocene lavas at Anegada High, offshore port of Veracruz, and the Middle to Late Miocene volcanism at the Los Tuxtlas. The existence of major Neogene extensional faults parallel to the Gulf of Mexico (i.e., ~N–S to NNW–SSE) proposed in previous works was not confirmed by our geological studies. Elongation of volcanic necks, vent alignment, and faults mapped by subsurface data trend dominantly NE to ENE and NW to NNW. These directions are parallel to transform and normal faults that formed during the Late Jurassic opening of the Gulf of Mexico. Ascent of mafic magmas was likely facilitated and controlled by the existence of these pre-existing basement structures.

T Corresponding author. Tel.: +52 442 238 1104x177; fax: +52 442 238 1129. E-mail address: luca@geociencias.unam.mx (L. Ferrari).

0377-0273/$ - see front matter D 2005 Elsevier B.V. All rights reserved.

doi:10.1016/j.jvolgeores.2005.02.004

L. Ferrari et al. / Journal of Volcanology and Geothermal Research 146 (2005) 284–306

285

Coupled with previous studies, our data demonstrate the occurrence of three magmatic episodes in Veracruz: 1) A Middle Miocene (~15–11 Ma) episode in southern Veracruz (Palma Sola, Anegada, and Los Tuxtlas); 2) A Late Miocene to Early Pliocene (~7.5–3 Ma) pulse of mafic alkaline volcanism throughout the study region; and 3) A Late Pliocene to Quaternary transitional to calc–alkaline volcanism in southern Veracruz (Palma Sola, Los Tuxtlas). Whereas the first and third episodes may be considered part of the subduction-related Trans-Mexican Volcanic Belt, the second pulse of mafic alkaline volcanism has a more complex origin. The absence of significant extensional faulting precludes a rift origin. We favor a model in which a transient thermal anomaly and melting of the mantle was triggered by the tearing and detachment of part of the subducted slab. D 2005 Elsevier B.V. All rights reserved.

Keywords: alkaline volcanism; tectonics; eastern Mexico; Gulf of Mexico; geochronology; late Cenozoic

1. Introduction

In the early days of the plate tectonics theory, the occurrence of alkaline volcanism was generally related to intra-plate tectonic settings distinct from convergent plate boundaries. In recent decades, however, alkaline volcanism has been recognized in virtually every tectonic environment, including many continental volcanic arcs worldwide (see Lange and Carmichael, 1991 , for a review). Alkaline volcanism at convergent plate boundaries has been inferred to be associated with slab-induced asthenospheric corner flow (Tokso¨z and Bird, 1977 ), slab-window formation ( Dickinson and Snyder, 1979; Hole et al., 1995 ), slab roll-back ( Furlong et al., 1982 ), and combination of slab windows and mantle plumes (e.g., Abratis and Worner, 2001 ). Cenozoic alkaline volcanism is widespread in eastern Mexico but its relation with the southern Mexico subduction zone ( Fig. 1 ) is unclear. A roughly north–south belt of Tertiary mafic alkaline volcanic fields runs from the U.S. border to the southern Veracruz State ( Fig. 1 ), intersecting the subduction- related Trans-Mexican Volcanic Belt (TMVB) ( Ferrari et al., 1999 ) in central Veracruz. Spanning over 1500 km in length, this zone of mafic volcanism constitutes a prominent feature in the geology and geomorphol- ogy of eastern Mexico, which otherwise is dominated by Mesozoic to early Tertiary marine and late Tertiary nonmarine sedimentary successions. Robin (1976, 1982) defined the belt of mafic alkaline volcanism as the b Eastern Alkaline Province Q (EAP hereafter) and, based on a number of conventional K–Ar datings and mostly major elements geochemistry, suggested that it represented intraplate-type volcanism migrating from north to south from Oligocene to Present. In the

Robin (1982) model, the EAP would be the result of Gulf-parallel extensional faulting and would be unrelated to the subduction of the Cocos plate; however he reported that during the Pliocene the products of alkaline volcanism alternated with arc- related lavas of the eastern TMVB in the Veracruz region. Subsequent geochemical and isotope studies questioned this model, at least for the southernmost part of the EAP. Besch et al. (1988) and Lo´ pez- Infanzon (1991) show that the Chiconquiaco–Palma Sola volcanic field has a geochemical imprint of fluids from the subducting plate. Go´mez-Tuena et al. (2003) provided a more detailed petrologic study of the three volcanic successions of this area. They interpret the chemical and isotopic characteristics of the Neogene volcanism at Palma Sola as controlled by variation in time of the depth of the subducting Cocos slab. Similarly Nelson et al. (1995) recognized a variable subduction signal for the Los Tuxtlas volcanic field, for which they suggest an analogy with lavas erupted in back-arc settings in Japan and the Andes. Such data would indicate that at least part of the EAP could represent the continuation of the arc volcanism of the TMVB toward the southeast ( Fig. 1 ). If so, the mere existence of a single volcanic province would be under question. However, with the notable exception of the Los Tuxtlas volcanic field ( Nelson and Gonza´lez-Caver, 1992 ), after the Robin (1982) syn- thesis the rest of the EAP has remained largely unstudied geologically and geochronologically. In an attempt to better define the time and space evolution of the mafic alkaline volcanism in eastern Mexico, and to understand its relation with the TMVB, we undertook a geologic, geochronologic and petrologic study of the southern half of this province. In this paper we summarize the geologic

286 L. Ferrari et al. / Journal of Volcanology and Geothermal Research 146 (2005) 284–306

of Volcanology and Geothermal Research 146 (2005) 284–306 Fig. 1. Geodynamic map of Mexico showing the

Fig. 1. Geodynamic map of Mexico showing the alkaline volcanism of eastern Mexico, the Trans-Mexican Volcanic Belt (adapted from Ferrari, 2004), and the present configuration of plates. The Eastern Alkaline Province of Robin (1982), includes the following volcanic fields: Sierra de San Carlos (SC), Sierra de Tamaulipas (ST), Tlanchinol–Tantima–Alamo (TTA), Chiconquiaco–Palma Sola (CP), Anegada High (AH), and Los Tuxtlas (LT). PV= Puerto Vallarta; Gdl =Guadalajara.

setting and present an extensive geochronologic study of the products of mafic volcanism exposed in Veracruz state. Our new fieldwork, 49 unspiked K– Ar ages and two 40 Ar/ 39 Ar ages, indicate that volcanism throughout the southern EAP commenced pene-contemporaneously at ~7 Ma, peaked at differ- ent times, and, locally, continued until the Quaternary. Together with a petrologic study ( Orozco-Esquivel et al., 2003 submitted) these data allow us to propose a new model for the genesis of the late Miocene to Pliocene alkaline volcanism in eastern Mexico.

2. Sampling and analytical procedure

Forty-six lavas were dated in an attempt to elucidate the regional pattern of volcanism and the

main pulses of activity. Five intrusive rocks were also dated to determine the initiation of magmatism in the Palma Sola area. Our study focused on areas without previous reported ages, or where previous ages seemed to be inconsistent with the stratigraphy. Tables 1 and 2 list the sampling localities and the analytical data. Samples were collected from road- cuts or from large blocks broken from lava-flow scarps using a sledgehammer. Most of the samples are aphyric and none contains plagioclase phenoc- rysts. On the basis of visual observation, petro- graphic studies, and whole-rock water content (LOI), the freshest rocks were selected for dating. For each sample, about 100 g of fresh rocks were crushed in a stainless steel mortar and sieved to 30-60 mesh. The sieved samples were washed first with deionized water and then with acetone in an ultrasonic cleaner

 

L. Ferrari et al. / Journal of Volcanology and Geothermal Research 146 (2005) 284–306

287

Table 1 Locations and K–Ar ages of samples dated in this study

 

Sample number

Site

 

Latitude (N)

Longitude (W)

Altitude (m)

K–Ar age (Ma) with error

Poza Rica–Metlatoyuca flows

 

EAP 1

3 km W of Poza Rica airport Plateau b La MesaQ, road from Santa Cruz Road Lazaro Cardenas–Mecapalapa

20.595

97.475

58

1.53 F 0.03 1.62 F 0.05 1.31 F 0.03 1.43 F 0.03 1.64 F 0.06 1.53 F 0.08 1.61 F 0.10 1.60 F 0.06 1.39 F 0.14 1.40 F 0.05

EAP 2

20.508

97.556

232

EAP 4

20.512

97.822

389

EAP 6

W of

Xicotepec

20.315

97.869

562

EAP 8

Lazaro Cardenas plateau E of Emiliano Zapata

20.464

97.710

362

EAP 9

20.554

97.519

123

EAP

11

Plateau b La Plateau b La

Pita Q

Pita Q

20.902

97.861

116

EAP

12

20.856

97.862

127

EAP

14

Plateau b CacahuatengoQ Plateau b MetlatoyucaQ

20.815

97.902

131

EAP

15

20.807

97.854

227

Alamo volcanic field and Sierra de Tantima

 

EAP

10

Cerro

Sombrerete

20.902

97.813

147

7.56 F 0.15 7.02 F 0.16 6.66 F 0.12 6.69 F 0.41 6.74 F 0.14 7.33 F 0.13 7.11 F 0.16 7.32 F 0.15 7.12 F 0.14 7.12 F 0.14 9.04 F 0.16 6.91 F 0.11 6.57 F 0.12 6.75 F 0.09

EAP

16

Cerro

Cacalote

21.086

97.826

303

EAP 17

Neck near La Guasima

21.110

97.833

308

EAP 18

Neck south of Tepetzintla Neck near Tierra Blanca

21.152

97.838

404

EAP 19

21.154

97.920

308

EAP

20

Cerro

El Espinal

21.056

97.982

149

EAP 21

Cerro

Tlacolula

near Zapotal Espinal

21.040

97.971

156

EAP

22

Cerro

El Can˜o´ n Moralillo

21.050

97.913

163

EAP

23

Cerro

21.189

97.814

239

EAP

24

Neck W

of San

Felipe

21.162

97.792

202

EAP

25

Cerro El Pelo´n San Juan Otontepec, Sierra Tantima S of Tantima village, Sierra Tantima S of Tantima village, Sierra Tantima

21.200

97.856

365

EAP 26

21.223

97.944

473

EAP 27

21.324

97.833

650

EAP 28

21.324

97.833

560

Tlanchinol flows

 

EAP

30

Plateau

Huautla

21.069

98.253

470

2.82 F 0.16 5.72 F 0.13 7.33 F 0.13 7.30 F 0.13

EAP

31

Plateau

Zihuapiltepetl

21.075

98.538

715

EAP

33

SW of Huejuetla S of Tlanchinol

21.023

98.610

1200

EAP 36

20.941

98.687

1473

Chiconquiaco–Palma Sola area

 

EAP 39

Alto de Tı´o Diego Mafafas–Tepetlan Mafafas–Tepetlan Alto Lucero–Enrı´quez

19.620

96.758

1134

2.04 F 0.04 3.18 F 0.06 3.25 F 0.06 1.97 F 0.04 3.22 F 0.06 14.65 F 0.32 3.37 F 0.06 3.38 F 0.06 15.62 F 0.50 6.93 F 0.16 3.85 F 0.07 3.50 F 0.07 4.03 F 0.07 3.53 F 0.05 3.27 F 0.04 7.48 F 0.13

EAP

41

19.667

96.772

1426

EAP

42

19.668

96.775

1371

EAP

43

19.660

96.739

1270

EAP

44

El Madron˜o South of Plan de las Hayas

19.700

96.674

1207

EAP 45

19.735

96.667

1152

EAP 46

Paz

de

Enrı´quez

19.811

96.824

1611

EAP 47

Paz

de

Enrı´quez

19.843

96.816

1326

EAP

48

Junique La Esperanza Plan de Las Hayas Rı´o Vado El Vencedor Paso del Toro Rancho NuevoJuan Martı´n Cerro Cantera (Metates)

19.839

96.703

367

EAP 49

19.777

96.658

760

EAP 50

19.767

96.621

815

EAP 51

19.766

96.559

585

EAP

52

19.934

96.664

418

EAP 53

19.922

96.608

231

EAP 54

19.882

96.590

586

EAP 55

19.670

96.415

141

(continued on next page)

288

L. Ferrari et al. / Journal of Volcanology and Geothermal Research 146 (2005) 284–306

 

Table 1 (continued )

 

Sample number

Site

Latitude (N)

Longitude (W)

Altitude (m)

K–Ar age (Ma) with error

Chiconquiaco–Palma Sola area

 

EAP 56

Mesa de 24 El Limo´ n Rı´o Actopan Plan del Rı´o

19.712

96.547

564

3.51 F 0.06 11.07 F 0.20 1.92 F 0.18 2.23 F 0.05

EAP

57

19.646

96.540

429

EAP

58

19.497

96.583

222

EAP 59

19.402

96.647

342

to remove adhered fine grains. Phenocrysts were removed from the fraction using a Frantz isodynamic magnetic separator and handpicking to minimize the possible influence of excess argon ( Takaoka, 1989; Matsumoto et al., 1989 ). An aliquot of each sample was ground further by an automatic agate mortar for potassium analysis. Potassium concentrations were measured by flame emission photometry, with lithium internal standard and peak integration method ( Matsumoto, 1989 ). Analysis was performed at the Kyoto University geochronology laboratory using a FP-33D flame emission photometer, made by Hekisa Kagaku Co., Japan. About 100 mg of sample was used for each analysis. See Matsumoto (1989) for the details of handling procedures. Precision and accuracy of the potassium determi- nations was established through the analysis, under identical analytical conditions, of Geological Survey of Japan (GSJ) standards JA-2 and JB-3 ( Ando et al., 1989; Matsumoto, 1989 ). The estimated precision and accuracy of the sample data is thus ~1% (1 sigma), which is propagated to determine the analytical uncertainty of the K–Ar age. Argon isotopic measurements were performed by peak-height comparison method ( Matsumoto et al., 1989; Sudo et al., 1996 ). Analysis was made at the Kyoto University geochronology laboratory with a VG 3600 mass spectrometer operated in the static mode, connected to extraction and purification lines made by Ayumi Co. Ltd., Japan. The extraction and analysis system as well as the techniques used for argon isotopic measurements are described by Sudo et al. (1996) . The standard air, calibrated using Sori 93 biotite standard ( Sudo et al., 1998 ), and hot blank were measured periodically during the experiments for system sensitivity, mass discrimination and back- ground corrections. To obtain the accurate content of radiogenic 40 Ar, we applied a mass-fractionation

correction procedure ( Matsumoto and Kobayashi, 1995 ) in cases where the uncorrected value deviated from the actual value by N 1%. If single-stage fractionation is assumed, the difference between corrected and uncorrected ages of a sample is less than 1% when the air contamination is less than 16%. Sample C-1 was analyzed by the conventional K–Ar method in 1990 in the Instituto Mexicano del Petroleo, Mexico city by J. Jacobo-Albarra´n and the result of the experiment is reported in Table 3 . All K–Ar ages were calculated using the isotopic abundances and decay constants for 40 K recommen- ded by the IUGS Subcommission on Geochronology ( Steiger and Ja¨ger, 1977 ; k e = 0.581 10 10 y 1 , k h = 4.962 10 10 y 1 , 40 K/ K = 1.167 10 4 ). Samples PS-99-21 and LH 1718 were dated by the 40 Ar– 39 Ar method at CICESE’s geochronology labo- ratory using a MS-10 mass spectrometer; details on the methodology are given in Ferrari et al. (2002) . The samples were analyzed in duplicate using the step- heating technique. Five to six fractions were collected between 950 and 1350 8 C. The results are given in Table 4 and Fig. 6 . For sample PS-99-21, a plagioclase concentrate was analyzed. Due to prob- lems with the data-acquisition system, the first two fractions collected on sample PS-99-21 (labeled 1 and 2 in the age spectra) are not considered reliable (see Fig. 6 ). Nonetheless, the results of the duplicate experiments yielded reproducible results with inte- grated ages of 12.5 F 1.6 Ma and 12.1 F 0.6 Ma. The slightly b UQ -shaped age spectrum suggests the pres- ence of excess argon, which is confirmed by the isochron age of 10.9 F 0.8 Ma calculated from the 36 Ar/ 40 Ar vs. 39 Ar/ 40 Ar correlation diagram with the combined fractions of the duplicate experiments, excluding fractions 1 and 2. We then take the isochron age as our best estimate for the age of sample PS-99- 21. For sample LH 1718, two experiments were performed using a biotite concentrate, and these

 

L. Ferrari et al. / Journal of Volcanology and Geothermal Research 146 (2005) 284–306

 

289

Table 2 Analytical data of K–Ar ages obtained with the unspiked sensitivity method at the Kyoto University geochronology laboratory

Sample

Weight

2 O

K

38 Ar/ 36 Ar

40 Ar/ 36 Ar

40 Ar/ 36 Ar initial

40 Ar rad (10 8 cm 3 STP/g)

40 Ar atm (%)

K–Ar age

number

(g)

(wt.%)

(Ma)

Poza Rica and Metlatoyuca flows

 

EAP

1-3

3.8634

0.87

0.1866 F 0.0006 0.1880 F 0.0007 0.1855 F 0.0008 0.1866 F 0.0007 0.1883 F 0.0006 0.1870 F 0.0008 0.1873 F 0.0006 0.1859 F 0.0006 0.1880 F 0.0006 0.1878 F 0.0006

561.1 F 0.9 380.2 F 0.6 426.2 F 0.7 455.2 F 0.7 367.8 F 0.6 349.1 F 0.6 329.4 F 0.5 352.2 F 0.6 319.3 F 0.5 362.8 F 0.6

294.6 F 1.9 298.8 F 2.2 291.2 F 2.5 294.6 F 2.2 299.8 F 1.9 295.8 F 2.6 296.9 F 1.9 292.5 F 1.9 299.0 F 1.9 298.3 F 1.9

4.290 F 0.072 6.485 F 0.206 4.199 F 0.101 4.595 F 0.094 6.647 F 0.217 3.966 F 0.209 4.723 F 0.294 4.599 F 0.166 3.903 F 0.383 3.865 F 0.132

52.5

1.53 F 0.03 1.62 F 0.05 1.31 F 0.03 1.43 F 0.03 1.64 F 0.06 1.53 F 0.08 1.61 F 0.10 1.60 F 0.06 1.39 F 0.14 1.40 F 0.05

EAP

2-2

1.5552

1.24

78.6

EAP

4-3

3.0549

0.99

68.3

EAP

6-2

3.0000

1.00

64.7

EAP

8-2

1.5060

1.25

81.5

EAP

9-2

1.5237

0.80

84.7

EAP

11

0.7585

0.91

90.1

EAP

12-2

1.6327

0.89

83.0

EAP

14-2

1.4849

0.87

93.6

EAP

15-2

2.2867

0.86

82.2

Alamo volcanic field and Sierra de Tantima

 

EAP

10

0.7454

1.42

0.1864 F 0.0006 0.1861 F 0.0006 0.1875 F 0.0009 0.1878 F 0.0006 0.1856 F 0.0006 0.1881 F 0.0007 0.1878 F 0.0007 0.1880 F 0.0006 0.1860 F 0.0006 0.1868 F 0.0006 0.1868 F 0.0006 0.1877 F 0.0006

582.3 F 0.9 443.2 F 0.7 877.7 F 2.3 332.6 F 0.5 468.2 F 0.7 739.4 F 1.2 486.1 F 0.8 514.9 F 0.8 519.4 F 0.8 537.6 F 0.9 419.5 F 0.7 416.1 F 0.7

294.1 F 1.9 293.1 F 1.9

34.791 F 0.573 28.734 F 0.575 32.670 F 0.492 42.453 F 2.559 20.310 F 0.378 41.570 F 0.625 30.764 F 0.611 37.887 F 0.666 27.030 F 0.468 38.445 F 0.658 45.800 F 0.997 46.584 F 1.045

50.5

7.56 F 0.15 7.02 F 0.16 6.66 F 0.12 6.69 F 0.41 6.74 F 0.14 7.33 F 0.13 7.11 F 0.16 7.32 F 0.15 7.12 F 0.14 7.12 F 0.14 8.97 F 0.21 9.12 F 0.22 9.04 F 0.16 6.93 F 0.16 6.89 F 0.16 6.91 F 0.11 6.57 F 0.12 6.77 F 0.14 6.73 F 0.12 6.75 F 0.09

EAP

16

0.7704

1.27

66.1

EAP

17

0.7899

1.52

295.5

33.7

EAP

18-3

0.1642

1.96

298.1 F 2.0 291.6 F 1.9

89.6

EAP

19-2

1.5342

0.93

62.3

EAP

20-2

0.7748

1.75

295.5

40.0

EAP

21

0.7526

1.34

298.2 F 2.4 298.9 F 1.9 292.8 F 1.9 295.2 F 1.9 295.1 F 1.9 297.9 F 1.9

61.3

EAP

22-2

0.7708

1.60

58.0

EAP

23

0.7639

1.17

56.4

EAP

24-2

0.7416

1.67

54.9

EAP

25-2

0.5004

1.58

70.3

EAP

25-3

0.5096

1.58

71.6

Weighted mean

 

EAP

26-2

0.7524

1.05

0.1878 F 0.0006 0.1876 F 0.0008

458.1 F 0.7 532.5 F 0.9

296.3 F 2.3 299.2 F 3.6

23.509 F 0.477 23.379 F 0.488

62.9

EAP

26-3

0.8062

1.05

54.3

Weighted mean

 

EAP

27-2

1.2206

1.06

0.1860 F 0.0009 0.1860 F 0.0006 0.1836 F 0.0008

674.3 F 1.3 555.0 F 0.9 634.6 F 1.1

295.5

22.409 F 0.338 34.877 F 0.613 34.669 F 0.524

43.8

EAP

28-2

0.8074

1.59

290.662 F 2.6

50.5

EAP

28-3

0.8014

1.59

295.5

44.7

Weighted mean

 

Tlanchinol flows

 

EAP

30-2

0.5021

0.96

0.1866 F 0.0006 0.1875 F 0.0006 0.1869 F 0.0006 0.1869 F 0.0009

332.0 F 0.7 441.2 F 0.7 1391.2 F 2.2 788.0 F 1.4

294.6 F 1.9 297.4 F 1.9

8.761 F 0.486 18.335 F 0.373 27.943 F 0.419 39.249 F 0.590

88.7

2.82 F 0.16 5.72 F 0.13 7.33 F 0.13 7.30 F 0.13

EAP

31

1.5292

0.99

67.4

EAP

33

1.5183

1.18

295.5

21.2

EAP

36-2

0.7924

1.66

295.5

37.5

Chiconquiaco–Palma Sola area Intrusive rocks

 

EAP45

0.5311

1.19

0.1856 F 0.0008 0.1859 F 0.0006 0.1867 F 0.0011

517.5 F 1.4 367.0 F 0.6 1093.4 F 2.6

291.6 F 2.6 292.4 F 1.9

56.648 F 1.084 64.052 F 1.934 76.256 F 1.146

56.4

14.65 F 0.32 15.62 F 0.50 11.07 F 0.20

EAP48

0.2580

1.26

79.7

EAP57

0.5114

2.13

295.5

27.0

Chiconquiaco plateau

 

EAP39

2.2640

3.49

0.1844 F 0.0006 0.1842 F 0.0009 0.1865 F 0.0014 0.1849 F 0.0006

1393.7 F 4.2 1247.5 F 3.1 1090.4 F 2.3 548.9 F 0.9

295.5

22.951 F 0.345 26.760 F 0.402 21.654 F 0.325 14.346 F 0.245

21.2

2.04 F 0.04 3.18 F 0.06 3.25 F 0.06 1.97 F 0.04

EAP41

1.5412

2.61

295.5

23.7

EAP42

1.5096

2.06

295.5

27.1

EAP43

1.5580

2.20

289.4 F 2.1

52.7

(continued on next page)

290

L. Ferrari et al. / Journal of Volcanology and Geothermal Research 146 (2005) 284–306

 

Table 2 (continued )

 

Sample

Weight

K

2 O

38 Ar/ 36 Ar

40 Ar/ 36 Ar

40 Ar/ 36 Ar initial

40 Ar rad (10 8 cm 3 STP/g)

40 Ar atm (%)

K–Ar age

number

(g)

(wt.%)

 

(Ma)

Chiconquiaco plateau

 

EAP44

1.5655

1.26

0.1854 F 0.0013 0.1859 F 0.0014 0.1837 F 0.0015 0.1867 F 0.0006 0.1828 F 0.0014 0.1832 F 0.0007 0.1869 F 0.0020 0.1846 F 0.0016 0.1868 F 0.0012

881.2 F 2.4 737.1 F 1.4 1055.9 F 2.1 442.1 F 0.7 1197.8 F 2.2 505.5 F 0.8 1129.2 F 3.9 888.4 F 2.4 881.3 F 1.6

295.5

13.165 F 0.198 13.817 F 0.208 17.456 F 0.262 32.435 F 0.669 20.167 F 0.303 13.534 F 0.244 26.353 F 0.397 11.975 F 0.180 12.019 F 0.181

33.5

3.22 F 0.06 3.37 F 0.06 3.38 F 0.06 6.93 F 0.16 3.85 F 0.07 3.50 F 0.07 4.03 F 0.07 3.52 F 0.06 3.54 F 0.06 3.53 F 0.05 3.24 F 0.06 3.30 F 0.06 3.27 F 0.04 7.48 F 0.13 3.51 F 0.06 1.92 F 0.18 2.23 F 0.05

EAP46

1.5550

1.27

295.5

40.1

EAP47

1.5525

1.60

295.5

28.0

EAP49

0.7649

1.45

294.9 F 2

66.7

EAP50

1.5932

1.62

295.5

24.7

EAP51

1.5079

1.20

284.3 F 2.2

56.2

EAP52

0.7760

2.02

295.5

26.2

EAP53

1.5378

1.05

295.5

33.3

EAP53

2.2857

1.05

295.5

33.5

Weighted

mean

EAP54

2.3118

0.86

0.1855 F 0.0008 0.1874 F 0.0007

797.2 F 2.5 777.4 F 1.7

295.5

9.006 F 0.136 9.185 F 0.138

37.1

EAP54

3.0552

0.86

295.5

38.0

Weighted

mean

EAP55

0.7567

2.63

0.1850 F 0.0006 0.1850 F 0.0016 0.1867 F 0.0006 0.1841 F 0.0006

821.1 F 1.3 922.7 F 2.5 316.6 F 0.7 425.9 F 0.7

295.5

63.674 F 0.957 14.685 F 0.221 4.155 F 0.392 20.239 F 0.415

36.0

EAP56

1.5621

1.30

295.5

32.0

EAP58

0.7976

0.67

294.8 F 1.9 286.9 F 1.9

93.1

EAP59

1.5277

2.82

67.4

Analytical error reported as 1r . bWeightQ indicates the amount of rock powder used for 40 Ar measurements.

yielded reproducible results, with statistically undis- tinguishable plateau ages defined by more than 90% of the 39 Ar released. The 37 Ar Ca / 39 Ar K diagram mimics the b U Q -shaped age spectra. Our best age estimate for sample LH 1718 is 4.0 F 0.1 Ma, taken from the isochron age calculated with the combined fractions of the two experiments performed with this sample.

3. Geologic setting and geochronological results

In this section, we provide a geologic description of the volcanic fields that compose the southern EAP within the context of the ages obtained in this work as well as those published previously. The new geo- chronological data are presented in Tables 1 and 2 . Major-element analyses were used to characterize the chemical composition of the dated lavas ( Fig. 2 ). The volcanic fields are described below from north to south.

3.1. Sierra de Tantima and Alamo monogenetic volcanic field

Sierra de Tantima is a 19-km long, 5-km wide, and 1320 m-high mountain with a marked NE alignment that rises from the coastal plain of the Gulf of Mexico ( Fig. 3 ). At its center, it consists of a 700 m-thick succession of mafic lavas flows with negligible dip that cover early Tertiary sandstones and shales. The flows are typically 2–10 m thick but, toward the top of the succession, they can attain a thickness of up to 20 m. Lavas are aphyric to microporphyritic, olivine, clinopyroxene and plagioclase being the most abun- dant minerals. Compositionally, they range from basanite to hawaiite ( Fig. 2 ). No ages were previously available for this volcanic edifice, so we dated three samples (EAP 26, 27 and 28) by the K–Ar unspiked method. Ages range from 6.91 F 0.11 to 6.57 F 0.12 Ma (Table 2 ). Sample EAP 26 (6.91 F 0.11 Ma) comes from the lower part of the succession in the southwestern part of Sierra de Tantima. The other two

Table 3 Analytical data of K–Ar age obtained for core N-1 at well C-1 at Instituto Mexicano del Petro´leo

Location

Average % K

40 Ar* (mol/g)

40 K (mol/g)

40 Ar*/ 40 K

% Rad

Age (Ma)

Offshore Villa Zempoala

3.25

4.14 10 11

9.73 10 8

4.25 E 4

49

7.3 F 0.73

The well is located east of Villa Zempoala ( Fig. 5) at 96.3208 W, 19.3728 N.

L. Ferrari et al. / Journal of Volcanology and Geothermal Research 146 (2005) 284–306

Table 4 40 Ar– 39 Ar results for samples from the Chiconquiaco–Palma area

291

Sample

t i (Ma)

t p (Ma)

t c (Ma)

( 40 Ar/ 36 Ar) i MSWD/n J 10 3

Temp % 39 Ar t (Ma)

37 Ar Ca /

% 40 Ar atm

 

(8C)

39

Ar K

PS 99-21

12.5 F 1.6

10 F 2

310 F 14

0.02/ 5

4.97 F 0.03

950 z 21.3

14 F 5 12 F 6 11 F 2 12 F 2 15 F 3 14 F 2 11.8 F 1.2

9.4

84.9

plagioclase

1000 z 12.9

5.8

70.7

1150

21.4

2.8

73.6

1250

28.7

3.7

75.4

1350

15.7

9.8

87.7

PS 99-21

12.1 F 0.6

11.0 F 0.9

304 F 6

0.04/ 5

4.97 F 0.03

850

13.0

9.6

88.9

plagioclase

1000

18.7

7.8

68.8

1150

23.0

11.3 F 1.0

2.9

66.6

1250

29.5

11.8 F 0.8

3.8

68.0

1350

15.8

12.7 F 1.7

9.4

82.4

 

10.9 F 0.8 y 305 F 5

0.07/ 8

LH 1718

4.5 F 0.2

4.3 F 0.2

4.2 F 0.2

304 F 4

0.06/ 6

2.07 F 0.04

850

1.8

10 F 4

2.0

94.0

biotite

1050

35.2

4.5 F 0.2 § 0.09

66.1

1125

19.9

4.3 F 0.3 § 0.06

40.8

1175

22.1

4.3 F 0.3 § 0.04

27.7

1225

19.7

4.3 F 0.3 § 0.24

38.1

1350

1.3

11 F 5 8.8 F 1.6

6.3

96.2

LH 1718

4.3 F 0.2

4.1 F 0.1

4.0 F 0.1

308 F 14

0.08/ 6

2.07 F 0.04

900

3.5

1.4

92.8

biotite

1050

35.3

4.2 F 0.2 § 0.07

62.3

1125

23.8

4.1 F 0.2 § 0.06

36.7

1175

32.1

4.0 F 0.2 § 0.04

25.1

1225

4.7

4.7 F 1.1 § 0.9

52.8

1350

0.6

6 F 9

1.3

95.4

 

4.0 F 0.1 y 307 F 3

0.21/ 12

PS 99-21: Candelaria gabbro (96.5218 W, 19.7928 N); LH 1718: diorite recovered from at about 1700 m below the surface in a well near Plan de las Hayas (Fig. 5). Analyses performed at CICESE’s geochronology laboratory, Mexico. All errors are 1r . The abbreviations are: plg = plagioclase; bio = biotite; t i =integrated age; t p = plateau age with the error in J included; t c =isochron age calculated from the 36 Ar/ 40 Ar vs 39 Ar/ 40 Ar correlation diagram; y =isochron age calculated with the combined fractions of the duplicate experiments performed; z = fraction ignored in the isochron calculation (see text); § fractions used to calculate t p . Best age estimate is given in bold typeface. The decay constants used, are those recommended by the IUGS (Steiger and Ja¨ger, 1977). Formulae given in York et al. (2004) was used to obtain the best straight line for the isochron calculation and its parameters.

samples (EAP 27 and 28) are stratigraphically higher and yielded slightly younger ages ( Table 2 ). The narrow age range is consistent with the absence of erosional discontinuity and/or paleosoils within the lava flows. The Alamo volcanic field is composed of at least 40 monogenetic volcanoes that surround Sierra de Tantima (Fig. 3 and Table 5 ). Because of erosion, lava flows are rarely preserved and a neck of massive lava (volcanic vent) is the only evidence of the volcanic structure. These volcanic necks have elevations ranging from 100 to 300 m. Modal composition of these rocks is very similar to the Sierra de Tantima samples. However, they show a greater variation in chemical composition, ranging from basanite to phonotephrite ( Fig. 2 ). Samples from ten necks

yielded ages that cluster in two time intervals:

6.66 F 0.12 to 6.74 F 14 Ma (samples EAP 17–19) and 7.02 F 0.16 to 7.56 F 0.15 Ma (samples EAP 10, EAP 16 and EAP 20–24) ( Table 2 ). A neck close to the southern edge of Sierra de Tantima yielded an older age of 9.04 F 0.16 Ma (weighted mean of duplicate experiments, sample EAP 25 in Table 2 ). No radiometric ages were available previously for this region. With the exception of the 9 Ma old neck our ages clearly point to a relatively short volcanic pulse spanning less than 900 ka. Initially, the volcanic activity was weak and scattered, resulting in the development of the Alamo monogenetic field. Later, the output rate of volcanism apparently increased and was able to sustain also more voluminous fissure eruptions that eventually built Sierra de Tantima.

292 L. Ferrari et al. / Journal of Volcanology and Geothermal Research 146 (2005) 284–306

10 5 8 4 6 1 3 4 2 2 0 40 45 50 55
10
5
8
4
6
1
3
4
2
2
0
40
45
50
55
Alkaline
Subalkaline
Na 2 O + K 2 O wt.%

SiO 2 wt%

Quaternary

Poza Rica flowsSubalkaline Na 2 O + K 2 O wt.% SiO 2 wt% Quaternary Late Pliocene Chiconquiaco

Late Pliocene

Chiconquiaco plateauwt.% SiO 2 wt% Quaternary Poza Rica flows Late Pliocene Late Miocene – early Pliocen e

Late Miocene – early Pliocene

Chiconquiaco plateauChiconquiaco plateau Late Miocene – early Pliocen e Late Miocene Alamo volcanic field Sierra de Tantima

Late Miocene

Alamo volcanic field– early Pliocen e Chiconquiaco plateau Late Miocene Sierra de Tantima Tlanchinol flows Fig. 2. Total

Sierra de Tantimae Chiconquiaco plateau Late Miocene Alamo volcanic field Tlanchinol flows Fig. 2. Total alkali–silica

Tlanchinol flowsplateau Late Miocene Alamo volcanic field Sierra de Tantima Fig. 2. Total alkali–silica classification for the

Fig. 2. Total alkali–silica classification for the dated rocks (after Le Bas et al., 1986). Line dividing subalkaline and alkaline fields from Irvine and Baragar (1971). Labels in fields are 1: Basanite; 2: Basalt; 3: Hawaiite/potassic trachybasalt; 4: Mugearite/shoshonite. The Late Pliocene samples, with Na 2 O 2 b K 2 O, belong to the shoshonitic series (potassic trachybasalt and shoshonite). Major element analyses performed by XRF by R. Lozano at Instituto de Geologı´a, UNAM, Mexico City.

by R. Lozano at Instituto de Geologı´a, UNAM, Mexico City. Fig. 3. Geologic map of the

Fig. 3. Geologic map of the Tlanchinol–Tantima–Alamo area showing the location and age of previous and new isotopically dated rocks.

 

L. Ferrari et al. / Journal of Volcanology and Geothermal Research 146 (2005) 284–306

 

293

Table 5 Vent elongation in the Alamo volcanic field

 

Name of vent

Latitude (N)

Longitude (W)

Sample number

Age (Ma)

Error

Elongation

Cerro Las Borrachas Cerro Quebracho

21.742

97.074

320

21.443

97.792

320

Cerro

Pelo´ n

21.388

97.743

58

Cerro El Borracho Cerro El Trueno Ojo de Brea Cerro El Pelo´ n

21.385

97.750

49

21.371

97.762

323

21.308

97.716

312

21.200

97.856

EAP 25

9.04

0.16

17

Neck Cerro Moralillo

21.189

97.814

EAP 23

7.12

0.14

318

W

of San Felipe

21.162

97.792

EAP 24

7.12

0.14

Not detectable

Near Tierra Blanca

21.154

97.920

EAP 19

6.74

0.14

Not detectable

South of Tepetzintla

21.152

97.838

EAP 18

6.69

0.41

43

N

of Loma El Repartidero

21.150

97.929

Not detectable

Near La Guasima Cerro Tepenecuile W Loma El Repartidero Cerro Cacalote Cerro Tepenecuile E Near Tenango Near La Mesilla Cerro Artı´culo Cerro La Cuatro Cerro El Espinal Cerro el Can˜o´n Cerro Chapopotal Cerro Tlacolula Cerro El Tepetate Cerro La Mesa S of Cerro Viejo Cerro del Plumaje Cerro Xococatl Cerro Ayacaxtle Cerro Tepecxtla Cerro La Cruz Cerro Aguacatepec Cerro Tepoxteco Cerro Ixcacuatitla Cerro Sombrerete Cerro Tzohuacale Cerro Mirador

21.110

97.833

EAP 17

6.66

0.12

Not detectable

21.090

97.901

22

21.089

97.933

Not detectable

21.086

97.826

EAP 16

7.02

0.16

41

21.085

97.905

40

21.083

97.841

43

21.080

97.895

Not detectable

21.075

97.785

315

21.063

97.707

338

21.056

97.982

EAP 20

7.33

0.13

40

21.050

97.913

EAP 22

7.32

0.15

55

21.045

97.712

335

21.040

97.971

EAP 21

7.11

0.16

45

21.040

97.898

Not detectable

21.033

97.920

Not detectable

21.003

97.733

345

20.983

97.774

Not detectable

20.974

97.975

41

20.962

98.015

43

20.946

98.038

30

20.937

97.700

Not detectable

20.936

97.929

45

20.922

98.066

Not detectable

20.908

98.032

45

20.902

97.813

EAP 10

7.56

0.15

48

20.896

98.045

25

20.892

97.705

38

3.2. Tlanchinol flows

To the west of the Alamo volcanic field, the Sierra Madre Oriental (SMOr) consists of a thick marine succession of Mesozoic age involved in regional NNW-trending folds and thrusts of Laramide age ( Fig. 1 ). Several massive lava flows are exposed on the eastern slopes of the SMOr between Tlanchinol and Huejutla ( Fig. 3 ). In the Tlanchinol area, these lavas were studied by Robin and Bobier (1975) who identified at least 22 basaltic flows emplaced in four

phases of activity. Cantagrel and Robin (1979) later dated two samples near Molango at 6.5 and 7.4 Ma, as well as a sample to the northeast of Tlanchinol at 7.3 Ma ( Fig. 3 ). The volcanic succession is thick as 250 m and the lavas are aphyric to microporphyritic with olivine and pyroxene phenocrysts. They range in composition from alkali basalt to hawaiite ( Fig. 2 ). In the highest part of the SMOr, between Tlanchinol and Molango, the flows are deeply eroded and in part covered by early Pliocene silicic ash and pumice flow deposits

294 L. Ferrari et al. / Journal of Volcanology and Geothermal Research 146 (2005) 284–306

that become more abundant toward the southeast ( Fig. 3 ). To the east, the mafic flows probably filled paleo- valleys but now constitute WSW–ENE elongated ridges because of their resistance to erosion ( Fig. 3 ). In the Huautla area, several flows coalesced to form large b mesas Q with a thickness of several tens of meters. In the Tlanchinol area we dated two samples (EAP 33 and 36) from the lower part of the succession, yielding nearly identical ages of 7.30 F 0.13 and 7.33 F 0.13 Ma, confirming the previous age of Cantagrel and Robin (1979) . We obtained an age of 5.72 F 0.13 Ma for a third sample (EAP 31), exposed at lower elevation and closer to Huejutla, ( Table 2 ). Consisting with its more youthful morphology, the Huautla flow yielded a younger age of 2.82 F 0.16 Ma (EAP 30). This latter age is comparable with the

Atotonilco basaltic succession exposed to the south of Zacualtipan ( Fig. 3 ) for which Cantagrel and Robin (1979) obtained ages of 2.6 and 2.4 Ma (south of the limit of Fig. 3 ). Our new ages indicate that the beginning of mafic volcanism in the Tlanchinol area is contemporaneous with the early activity in the Alamo volcanic field.

3.3. Poza Rica flows

Several massive mafic lava flows cover the eastern slope of the SMOr west of Poza Rica (Fig. 4 ). Our geological studies indicate that these lavas flowed for over 90 km from the front of the SMOr west of Huachinango to the coastal area of Poza Rica, about 2200 m topographically lower. Locally, the lavas

Rica, about 2200 m topographically lower. Locally, the lavas Fig. 4. Geologic map of the Tulancingo–Poza

Fig. 4. Geologic map of the Tulancingo–Poza Rica area showing the location and age of previous and new isotopically dated rocks.

L. Ferrari et al. / Journal of Volcanology and Geothermal Research 146 (2005) 284–306

295

filled paleo-valleys, reaching a thickness of several hundreds of meters. Although partly eroded, the flows are almost continuous from Huachinango to Poza Rica. Two other large flows are exposed northwest of Poza Rica in the Metlatoyuca area ( Fig. 4 ). In general, all these flows are better preserved than those in the Tlanchinol area. Lavas are variably porphyritic with olivine and clinopyroxene as phenocrysts. They show a small compositional range that span across the basalt and hawaiite fields ( Fig. 2 ). No ages were available in the literature for these rocks. We have dated six samples (EAP 1, 2, 4, 6, 8 and 9) at different elevations of the flows. All data cluster in a narrow age range comprised between 1.64 F 0.06 and 1.31 F 0.03 Ma ( Table 2 ). Four samples (EAP 11, 12, 14 and 15) from the Metla- toyuca flows were also dated. The lower flow yielded two ages, indistinguishable within the error, of ~1.6 Ma. The upper flow produced also two nearly identical ages of ~1.4 Ma ( Table 2 ). In light of these results, the Metlatoyuca flows may represent part of the Poza Rica flows that branched off toward the north once the flows arrived in the coastal plain ( Fig. 4 ). The Poza Rica flows are probably equivalent to the alkaline basalts described by Nelson and Lighthart (1997) in the Tulancingo canyon ( Fig. 4 ) at the base of the Quaternary Las Navajas volcano. This correlation is suggested by stratigraphic position and petrographic similarity between the two successions. As a whole, the flows in the Poza Rica area represent a major pulse of alkaline volcanism in the eastern part of the TMVB.

3.4. Chiconquiaco–Palma Sola area

Cenozoic volcanism in the Chiconquiaco–Palma Sola region spans a longer time interval than the areas to the north. The geology of the region is also more complex. The region is located in a structural high (the so-called b Teziutla´n massif Q ) where the Paleozoic basement (polydeformed schists, volcano–sedimen- tary and intrusive rocks) is uplifted along a roughly E–W-trending structure that breaks the continuity of the coastal basins ( Lo´ pez-Infanzo´ n, 1991 ) and divide the Tampico–Misantla basin from the Veracruz basin to the south ( Fig. 1 ). In the study area, these basement rocks are found west of Altotonga ( Fig. 5 ) and yielded K–Ar ages of 269–252 Ma ( Lo´ pez-Infanzo´ n, 1991 )

whereas 40 Ar/ 39 Ar experiments produced ages of 268 and 281 Ma ( Iriondo et al., 2003 ). The basement rocks are covered by the marine Mesozoic succession of the SMOr and by Cenozoic siliciclastic rocks and igneous successions ( Fig. 5 ). The Cenozoic magmatic activity in the region consists of four groups of rocks: 1) middle to late Miocene intrusive bodies of gabbroic to dioritic composition, mainly exposed along the coast in the Misantla and Palma Sola areas; 2) a latest Miocene to early Pliocene alkaline basaltic plateau centered in the Chiconquiaco area; 3) latest Pliocene shoshonitic lava flows of the Alto Lucero–Actopan area; and 4) Late Pleistocene to Holocene cinder cones with associated lava flows, located mostly to the south of the Chiconquiaco plateau ( Fig. 5 ). Lo´ pez-Infanzo´ n (1991) also distinguishes a unit of lahars and basalts between group 1 and 2. However, our field work rather suggest that this unit consists of debris and other erosional deposits, which locally may include part of the overlying basalts. The intrusive rocks often have a microporphyritic or microcrystalline texture indicative of a shallow depth of intrusion. They commonly display sulfur mineralization and chlorite alteration and are locally cut by mafic dikes. Previous workers dated these rocks by conventional K–Ar methods. Cantagrel and Robin (1979) obtained an age of 17 Ma for the Laguna Verde microdiorite; Negendank et al. (1985) report ages of 12.3 and 12.9 Ma for the Candelaria gabbro (no errors and analytical data provided); and Lo´ pez-Infanzo´ n (1991) dated three plutons exposed near Tenochtitlan and Junique at 13.0 F 1.0, 9.0 F 0.7, and 6.2 F 0.6 Ma ( Fig. 5 ). With the purpose of confirming this rather large age interval, we dated some of the plutonic bodies previously studied by these workers. We were unable to find a sample of the Laguna Verde microdiorite fresh enough for dating. These rocks, also called b green formation Q by Cantagrel and Robin (1979) , are actually intensely and pervasively chloritized; thus, we consider the previously reported age of 17 Ma as unreliable. For the Junique gabbro, previously dated at 6.2 Ma by Lo´ pez-Infanzo´ n (1991) , we obtained an age of 15.62 F 0.5 Ma (sample EAP 48; Table 2 ), which appears more consistent with the stratigraphy and other ages of the intrusive bodies. A plagioclase concentrate from the Candelaria gabbro was dated by

296 L. Ferrari et al. / Journal of Volcanology and Geothermal Research 146 (2005) 284–306

of Volcanology and Geothermal Research 146 (2005) 284–306 Fig. 5. Geologic map of the Chiconquiaco–Palma Sola

Fig. 5. Geologic map of the Chiconquiaco–Palma Sola area showing the location and age of previous and new isotopically dated rocks.

40 Ar/ 39 Ar. The integrated ages obtained are in agree- ment with the published K–Ar ages ( Negendank et al., 1985 ). However, we prefer to take the isochron age of 10.9 F 0.8 Ma as our best estimate for this sample, because of the possibility of excess argon as discussed earlier. In addition, we dated two other samples (EAP 45 and 57) from hypoabyssal bodies exposed imme- diately below the overlying plateau basalts near Plan de las Hayas and El Limo´n ( Fig. 5 ). These rocks yielded ages of 14.65 F 0.32 Ma and 11.07 F 0.2 Ma respectively. In summary, the most reliable ages for the intrusive rocks of Palma Sola range between 15.6 and 10.9 Ma. The only exception would be a tonalite

near the Tenochtitlan village ( Fig. 5 ), for which Lo´ pez-Infanzo´ n (1991) reports an age of 9.0 Ma age. This first phase of magmatism extended also in the Gulf of Mexico as suggested by the recent identification of a small eruptive center located offshore west of Santa Ana (Santa Ana High, Fig. 5 ), for which seismic reflection data suggest an age just younger than middle Miocene (Tim Wawrzyniec, written communication, 2003). An erosional unconformity and/or several tens of meters of clastic deposits ( b lahars and basalts unit Q of Lo´ pez-Infanzo´ n, 1991 ) separate the intrusive rocks from the fissural lava flows comprising the Chicon-

L. Ferrari et al. / Journal of Volcanology and Geothermal Research 146 (2005) 284–306

297

quiaco plateau, presently covering about 1700 km 2 . The northern side of the plateau is highly dissected whereas to the south the presence of capping Quaternary volcanic rocks has impeded strong ero- sion. North of Chiconquiaco, the succession is up to 800 m thick, although it thins rapidly to the east, being only some tens of meters in the coastal areas. Individual flows are 10 to 50 m thick and are generally separated by red soils or thin alluvial deposits. Compositionally, the lavas span a large range from basanite to benmoreite mostly with an intra-plate type affinity ( Fig. 2 ). Some rocks of this succession, however, show a moderate subduction signal, and their Pb isotope signature suggests the involvement of melts from subducted sediments in their genesis ( Go´mez-Tuena et al., 2003 ). South of Palma Sola, a group of dacitic domes (Cerro Cantera, Cerro Metates and associated domes) appear older than the plateau as they diverted the basaltic flows ( Fig. 5 ). According to Go´mez-Tuena et al. (2003) , melt derived from the subducted MORB is involved in magma genesis; Cerro Cantera displays the highest adakitic signature of the area. Cantagrel and Robin (1979) dated the dacitic dome of Cerro Cantera at 6.5 F 0.2 Ma and one plateau basalt at 3.1 F 0.3 Ma. Lo´ pez-Infanzo´ n (1991) dated sixteen samples from several sections on the Chiconquiaco plateau. Fourteen of them yielded ages ranging between 6.0 F 0.6 and 2.2 F 0.2 Ma; two older ages are discussed below. Our sampling strategy was designed to better con- strain the reported age span of this volcanic episode and to compare with some of the previously published ages. A sample from Cerro Cantera yielded an age of 7.48 F 0.13 Ma (EAP 55, Table 1 ), somewhat older than the Cantagrel and Robin (1979) age. The base of the plateau, sampled to the NE of Plan de las Hayas, yielded an age of 6.93 F 0.16 Ma (EAP 49, Table 2 ), which is similar to that obtained by Lo´ pez-Infanzo´ n (1991) for one basal lava southeast of Tenochtitlan ( Fig. 5 ). Most of the plateau succession, however, seems to have formed in a narrower age range. In fact, eleven samples taken at different stratigraphic levels within the succession yielded ages between 4.0 and 3.2 Ma (EAP 41, 42, 44–47, 50 and 56; Table 2 ). In particular, two samples collected on elevation differ- ence of nearly 300 m (EAP 46 and 47) along the road

from Chiconquiaco to Misantla yielded ages of about 3.4 Ma, indistinguishable within 1 j error ( Table 2 ). Our datings also include isolated mafic flows towards the coast in the El Bejuco area (EAP 52–54; Fig. 5 ), confirming that these rocks are part of the Chic- onquiaco plateau. Dioritic intrusives are also reported at depth in boreholes drilled by PEMEX (Mexican Oil Company) in the region, but doubts exist on whether they are part of the basement complex or belong to a younger magmatic episode. For this reason, we dated two samples from these wells. Biotite from a microdioritic dike from a well east of Villa Zempoala (on the southeast corner of Fig. 5 ) has been dated by conventional K–Ar at 7.3 F 0.7 Ma (Table 3 ), which is consistent with the beginning of the mafic volcanism of the Chiconquiaco plateau. A diorite core (LH 1718) recovered at about 1700 m below the surface in a well near Plan de las Hayas (Fig. 5 ) yielded an isochron age of 4.0 F 0.1 by the 40 Ar/ 39 Ar method on a biotite concentrate ( Table 4 ). Considering the good agreement of the two experi- ments and the nearly flat age spectrum obtained ( Fig. 6 ), we interpret this age as a crystallization age corresponding to the main phase of the Chiconquiaco plateau volcanism. These two ages of hypoabyssal bodies indicate the existence of intrusive equivalent of the plateau at depth. Moreover, the presence of an extensive, but unexposed, mafic intrusive complex would also explain the prominent positive Bouguer anomaly that this area displays in the gravimetric map of Mexico ( De la Fuente et al., 1994 ). We regard with caution two ~9 Ma ages reported by Lo´ pez-Infanzo´ n (1991) for samples from the lower part of the plateau. Sample LI 203, is reported as a porphyritic trachydacite, a composition more akin to the underlying rock unit. On the other hand, this sample, with a 9.4 F 0.5 Ma age, is reported to have been collected several tens of meters above two other samples that yielded ages of 5.6 and 6.0 Ma, casting doubts on its age and/or location. The other sample (LI 100), according to Lo´ pez-Infanzo´ n (1991) , is a hawaiite collected below a lahar and an unconformity, thus it does not belong to the plateau. Therefore, our new data suggest that the volcanic activity that built the Chiconquiaco plateau began at around 7.0 Ma (sample EAP 49, Table 2 ), although the main pulse of volcanism seems to have occurred at the beginning of the Pliocene.

298 L. Ferrari et al. / Journal of Volcanology and Geothermal Research 146 (2005) 284–306

of Volcanology and Geothermal Research 146 (2005) 284–306 Fig. 6. 4 0 Ar– 3 9 Ar

Fig. 6. 40 Ar– 39 Ar age spectra and 37 Ar Ca / 39 Ar K diagram obtained for the samples LH 1718 and PS-99-21. Within the spectra, the errors are given at the level of one standard deviation and are represented by the vertical width of the boxes. The analyses were performed by Margarita Lo´pez-Martı´nez at CICESE geochronological laboratory at Ensenada, Baja California, with the procedures described in Ferrari et al. (2002).

In the area between Alto Lucero and south of Actopan ( Fig. 5 ), several lava flows advanced to the southeast, discordantly overlying the Chiconquiaco plateau succession. These lavas are highly potassic and can be classified as shoshonites ( Fig. 2 ). In the Alto Lucero area, we dated two samples (EAP 39 and 43) from lavas flows without visible vent exposed at the top of the Chiconquiaco plateau. These samples yielded nearly identical ages of 1.97 F 0.04 and 2.04 F 0.04 Ma ( Table 2 ). For a

third sample (EAP 58) from a flow near Actopan, a similar age of 1.92 F 0.18 Ma was obtained. Finally, a lava flow at Plan del Rio, located about 10 km south of Actopan (just south of the limit of Fig. 5 ) yielded an age of 2.24 F 0.05 (EAP 59). These ages are comparable with the 2.2 Ma age obtained by Lo´ pez-Infanzo´ n (1991) for a flow south of Chic- onquiaco and the 1.9 Ma age reported by Cantagrel and Robin (1979) for a flow 5 km to the east of EAP 59.

L. Ferrari et al. / Journal of Volcanology and Geothermal Research 146 (2005) 284–306

299

The youngest activity in the Chiconquiaco–Palma Sola region is represented by at least 20 cinder cones and associated lava flows that in some cases display a very young morphology. The vents are aligned in a WSW–ENE direction between Perote and Palma Sola ( Fig. 5 ) and their lavas have basaltic to trachybasaltic composition ( Siebert and Carrasco-Nu´ n˜ez, 2002 ). In the valley between Naolinco and Actopan ( Fig. 5 ), a welded ash flow tuff (not reproducible at the map scale of Fig. 5 ) separates these rocks from the underlying Chiconquiaco plateau succession. Siebert and Carrasco-Nu´ n˜ez (2002) dated several of these flows by the 14 C method and obtained ages between 42,000 and 840 yr BP.

3.5. Neogene alkaline volcanism in southern Veracruz

The mafic volcanism of Chiconquiaco–Palma Sola extends to the southeast into the submarine volcanism at the Anegada High and the Los Tuxtlas volcanic field. These two volcanic fields also bound the

Veracruz basin to the east ( Jennette et al., 2003 ) (Fig. 7 ). We describe briefly the geology of these two volcanic areas, because they appear strictly related to the Chiconquiaco–Palma Sola volcanic field.

3.5.1. Anegada volcanic center The Anegada High is a positive structure where the pre-Cenozoic basement is uplifted with a NNW trend. The presence of submarine mafic volcanism in the southeastern part of the Anegada high was initially suggested by Moore and Castillo (1974) , mostly on aeromagnetic evidence and later corroborated by two exploratory wells drilled by PEMEX offshore of the Veracruz port (Anegada 2 and 3; Fig. 7 ). We obtained several cores of these wells but the samples proved to be too altered to be analyzed and dated. In thin section, the samples show porphyritic structure with olivine, clinopyroxene, and plagioclase phenocrysts. Microprobe analyses of the clinopyroxenes indicate an OIB affinity in the Ti vs. Ca + Na and Ti +Cr vs. Ca diagrams, with values mostly overlapping the alkaline

vs. Ca diagrams, with values mostly overlapping the alkaline Fig. 7. Regional geologic map of eastern

Fig. 7. Regional geologic map of eastern Mexico showing the main Neogene volcanic episodes discussed in the text. Mid-late Miocene adakitic TMVB as recognized by Go´mez-Tuena et al. (2003). Apan = Apan volcanic field; CG =Cerro Grande; Popo= Popocate´petl; Izta = Iztaccihuatl; Chich = Sierra Chichinautzin volcanic field; Pico= Pico de Orizaba; Cofre =Cofre de Perote. Crustal thickness (in kilometers) is from the gravimetric study of Urrutia-Fucugauchi and Flores-Ruiz (1996). Boundary of the Tampico–Misantla and Veracruz basins from Jennette et al.

(2003).

300 L. Ferrari et al. / Journal of Volcanology and Geothermal Research 146 (2005) 284–306

products of the Los Tuxtlas volcanics ( Jacobo- Albarra´n et al., 1994 ). The seismic stratigraphy calibrated at the Anegada 1 well indicates that the volcanism occurred between 16 and 7 Ma ( Jennette et al., 2003 ), coeval with the early phase of magmatism in the Palma Sola area and with the offshore volcanism at Santa Ana High ( Fig. 5 ).

3.5.2. Los Tuxtlas The Los Tuxtlas volcanic field covers an area of over 2200 km 2 in southernmost Veracruz State. A synthesis of previous studies ( Nelson and Gonza´lez- Caver, 1992; Nelson et al., 1995; Jacobo-Albarra´n, 1997 ) indicates that volcanism may be divided into two main episodes. An older episode began in late Miocene (6.9 Ma, Nelson and Gonza´lez-Caver, 1992 ; 7.9 F 0.7 Ma, Jacobo-Albarra´n, 1997 ) and locally continued until ~1 Ma, producing hyper- sthene-normative alkali-basalts, hawaiites, mugearites and benmoreites now exposed in remnants of large cones, a plateau, and over 200 cinder cones. A younger episode of calk–alkaline to transitional affinity began around 3.3 Ma. Most of the products of the younger episode were emplaced in five calderas located in the southeastern part of the field and other centers in the northwest of the field. Although the oldest dated rocks are ~7 Ma, seismic and borehole data indicate that volcanism in the Los Tuxtlas has influenced the sedimentation of the Veracruz basin since Middle Miocene ( Jennette et al., 2003 ). Therefore, the initial activity of the Los Tuxtlas volcanism appears to be concurrent with the first volcanic episode in the Palma Sola area as well as that of the Anegada High.

4. Tectonic setting

From a tectonic point of view, the studied volcanic fields are located in, or bound, two Cenozoic coastal basins of the western Gulf of Mexico that are well studied for their hydrocarbon contents: the Tampico–Misantla basin and the Veracruz basin ( Figs. 1 and 7 ). The Tlanchinol and Poza Rica flows were emitted from fissures along the front of the SMOr, coincident with the western boundary of the Tampico–Misantla basin. Sierra de Tantima and the Alamo volcanic field are located

inside this basin and mostly overlie its filling. Neogene volcanism of the Chiconquiaco–Palma Sola, Anegada High, and Los Tuxtlas areas con- stitute the northern and eastern boundary of the Veracruz basin respectively. Here, we discuss the relation between the mafic alkaline volcanism and the Neogene tectonics of the western Gulf of Mexico.

4.1. Northern Veracruz and Hidalgo

Previous workers ( Robin, 1976, 1982; Robin and Tournon, 1978 ) associated the alkaline volcanism of northern Veracruz and Hidalgo (Tlanchinol, Tantima and Alamo) with NNW-trending extensional faulting affecting most of the SMOr border. However, no structural information was provided by these authors, and in the geologic sections of Robin (1982) most of the normal faults are drawn as inferred and do not cut the surface. Although the dense vegetation precludes a detailed structural study, our studies indicate that Late Tertiary normal faults are not common in and around most of the studied volcanic fields. Published geo- logic maps ( Suter, 1990; Consejo de Recursos Minerales, 1997 ) also show the border of the SMOr as an Early Tertiary fold and thrust belt, only locally affected by minor Late Tertiary normal faulting. In his detailed geological survey of the Tlanchinol area, Ochoa-Camarillo (1997) explicitly rejects the exis- tence of the extensional fault system postulated by Robin (1982) . Moreover, the subsurface geology of the Tampico–Misantla basin indicates that the com- pressive regime that built the SMOr lasted until late Eocene times, but with only very minor deformation affecting post Eocene strata (Eguiluz de Antun˜ano et al., 2000 ). Nevertheless, the emplacement of the relatively primitive alkaline magmas of Tlanchinol, Tantima, and Alamo is likely to have occurred through some pre-existing crustal discontinuities. The flows of Tlanchinol appear spatially associated with a major basement fault system of probable Late Jurassic age. In fact, the Proterozoic gneisses of the Huiznopala Formation crop out at ~500 m elevation just 10 km to the north of Molango (Fig. 3 ) ( Ochoa-Camarillo, 1997; Lawlor et al., 1999 ), and disappear toward the north-east due to the NNE Ixtlapala–Huiznopala normal fault, which has a minimum displacement of

L. Ferrari et al. / Journal of Volcanology and Geothermal Research 146 (2005) 284–306

301

1600 m ( Ochoa-Camarillo, 1997 ). Further to the east, Precambrian rocks correlative with the Huiznopala gneisses were cored at a depth of 2600 m in the coastal plain of the Gulf of Mexico ( Suter, 1990 ). These normal faults developed in Late Jurassic as a consequence of the southeast motion of the Yucatan

block during the initial opening of the Gulf of Mexico ( Pindell and Kennan, 2001 ). The Sierra de Tantima and Alamo volcanic field lie

in a plain without any visible evidence of faulting.

Robin (1982) suggested that the peculiar NE align- ment of Sierra de Tantima was due to the accumu-

lation of lava flows in a paleo-valley whose sides were later completely eroded. This explanation seems highly unlikely, because the Sierra rises about 1300

m above the surrounding plain, plus a distance of over

35 km separates it from the other flows descending from the SMOr ( Fig. 3 ). In addition, lava flows that built the Sierra do not show a NE dip, which would be expected if they originated from the NNW-trending front of the SMOr. Consequently, we suggest that the NE elongation of Sierra de Tantima is the result of

eruption along a NE-trending fissure vent. This view

is corroborated by the analysis of vent elongation and

dike orientation in the Alamo volcanic field. In our

field work, we observe that many volcanic necks show a preferential elongation. In four cases, the feeding dikes were also exposed, with a direction parallel to the neck elongation. As summarized in Table 5 and Fig. 8 , there are two main elongation

directions: necks in the southwestern part of the field

as well as Sierra de Tantima trend NE–SW, whereas

necks in the northeastern part of the field are elongated in a NNW–SSE direction. The elongation of a cone is generally considered to express the underlying feeding fracture (e.g., Tibaldi, 1995 ). In this context, the directions observed in the field may reflect the presence of buried crustal structures. Recent paleo-reconstructions of the opening of the Gulf of Mexico ( Pindell and Kennan, 2001 ) show that Late Jurassic to Early Cretaceous paleotransforms that allowed the southeastern motion of the Yucatan block were oriented NNW, whereas normal faults bounding continental blocks were oriented NE or ENE. We conclude that the mafic volcanism in the Tampico– Misantla basin (Tlanchinol, Alamo, Tantima) was fed through old basement faults that were not necessarily structurally active at the time of volcanism.

necessarily structurally active at the time of volcanism. Fig. 8. Elongation direction (azimuth) of volcanic vents

Fig. 8. Elongation direction (azimuth) of volcanic vents of the Alamo Volcanic field (Table 5). Inset shows the possible orientation of the principal underlying basement structures.

4.2. Southern Veracruz

The tectonics of the Veracruz basin is much more complicated than those of basins to the North. The mafic volcanism of this region (Palma Sola, Anegada, and Los Tuxtlas) was included by Robin (1982) in the EAP and considered as extension-related intraplate volcanism. More recent studies by Besch et al. (1988) , Nelson et al. (1995) , and Go´mez-Tuena et al. (2003) , however, demonstrate that, unlike the volcanic fields to the north, the volcanism in the Palma Sola and Los Tuxtlas areas shows a variable signature from fluids and melts from the subducted plate. In the Chicon- quiaco–Palma Sola area, we could not find any clear evidence of an episode of extensional faulting concurrent with the alkaline volcanism. The main tectonic feature of the region seems to be an old ENE- trending structure, as substantiated by the following evidences. A 60-km long alignment of Quaternary cinder cones trends ENE from the Cofre de Perote area to the Candelaria area, in the coast ( Fig. 5 ). Along this alignment, near site EAP 50 (Plan de las Hayas, Fig. 5 ), we measured several dikes feeding the

302 L. Ferrari et al. / Journal of Volcanology and Geothermal Research 146 (2005) 284–306

early Pliocene Chiconquiaco plateau, which also strikes ENE. Finally, at the eastern end of the alignment lies the middle Miocene offshore volcanic center of the Santa Ana High (see Previous section). This volcanic center is located at the eastern end of an ENE-trending left-lateral, strike-slip fault active until the end of Middle Miocene (Tim Wawrzyniec, written communication, 2003). Therefore, most of the volcan- ism in the Chiconquiaco–Palma Sola region seems to be controlled by a major ENE-trending basement fault system that acted as a preferential magma pathway since Middle Miocene. In the Los Tuxtlas volcanic field, a statistical analysis of the distribution of cinder cones and polygenetic centers shows the existence of a 65-km long alignment with a 120 8 (ESE) orientation, expressed by at least 42 vents ( Jacobo-Albarra´n et al., 1992; Jacobo-Albarra´n, 1997 ). An array of secondary alignments and satellite lineaments also indicates that the main alignment could be the expression of movements along a right-lateral base- ment structure ( Jacobo-Albarra´n, 1997 ). The age of most recent activity of this fault system, however, cannot be established from the field geology. Additional information on the tectonics of the region is provided by the subsurface geologic studies in the Veracruz basin. Interpretation of extensive seismic and borehole data indicate that contractile and strike-slip deformations characterize the basin during Miocene and Pliocene times ( Jennette et al., 2003 ). In particular, volcanic activity at both the Anegada High and the Los Tuxtlas field was controlled by a NW–SE trending fault system known as the b Anton Lizardo trend Q , which has accommo- dated a right-lateral, strike-slip deformation since middle Miocene times ( Jennette et al., 2003 ) and still shows strike-slip seismic activity ( Sua´rez, 2000 ). Jennette et al. (2003) also recognize a major change in tectonics and sedimentation at about 7 Ma. At this time, the uplift of the northern part of the basin (Palma Sola area) triggered a major reorganization in the regional sedimentation pattern. Intrabasinal faults were reactivated mostly as strike-slip faults and broad folds that affected the center of the basin. In summary, the occurrence of mafic volcanism in the Veracruz basin seems to have been guided mostly by pre- existing strike-slip faults. Some of these faults may have been active concurrently with the volcanism.

5. Discussion and conclusions

5.1. Regional evolution of volcanism

The geologic and geochronologic data presented in this work, when integrated with previous studies, shed light on the Neogene evolution of volcanism in eastern Mexico ( Fig. 7 ). Our synthesis of the available data recognizes three main episodes of magmatism east of Mexico City. The onset of the TMVB magmatism can be placed in the Middle Miocene ( Ferrari et al., 1999 ). At this time, an E–W belt of mostly intermediate calc–alkaline centers developed. This belt includes polygenetic volcanoes of the Apan volcanic field ( Garcı´a-Palomo et al., 2002 ), the huge Cerro Grande volcanic complex ( Go´mez-Tuena and Carrasco-Nu´ n˜ez, 2000 ), and the intrusive and sub- volcanic bodies of Palma Sola ( Go´mez-Tuena et al., 2003 ; this work) (Fig. 7 ). According to Go´mez-Tuena et al. (2003) , most of these rocks have an adakitic signature (