Journal of Petrology | Volume 40 | Number 4 | Pages 511-524 | 1999
© Oxford University Press 1999
Inherited Palaeozoic and Mesozoic Rb–Sr Isotopic Signatures in Neogene Calc-alkaline Volcanics, Alborán Volcanic Province, SE Spain
1 Geological Institute, Copenhagen University Østervoldgade 10, 1350K Copenhagen, Denmark
2 Baker Atlas Geoscience (Dk) Jorcks Passage A-4, 1162K Copenhagen, Denmark
3 Institute for Study of the Earth's Interior, Okayama University Misasa, Tottori-Ken 682-01, Japan
Received November 9, 1997; Revised typescript accepted August 18, 1998
 |
ABSTRACT |
|---|
 TOP ABSTRACT IntroductionGeological SettingHernandez Pyroxene Andesite...Hernandez Rb-Sr Isotopic...Plate Tectonic Setting of...ConclusionsReferences |
|---|
A pyroxene andesite unit within the post-Alpine Alborán volcanic province has a Neogene extrusion age; however, its Rb–Sr isotopic relations define a regression line of 509 ± 62(2
) Ma (Early Palaeozoic). There are two concordant data point clusters on the regression line, one of which is well constrained, defining a secondary regression line of 202 ± 30(2) Ma (Early Mesozoic). Considering the mineralogy of the andesites—plagioclase, Ca-poor and Ca-rich pyroxene, and Ti-magnetite—and the presence of restitic aggregates comprising these same four minerals, recent dehydration melting experiments suggest an origin by anatexis of an amphibolite-dominated source rock complex. Inherited zircon ion-microprobe ages in the range of 500–1800 Ma, an Sm–Nd isochron age of 1.5 ± 0.4(2) Ga, TCHURNd crustal derivation ages from 
0.75 to 1.05 Ga and 
Nd(0) values of –4 to –7 support a complex petrogenesis, involving large-scale reworking of older material. 87Sr/86Sr vs 1/Sr and 143Nd/144Nd vs 1/Nd indicate a heterogeneous source rock complex showing two-component mixing. The data favour volcano-sedimentary source rock complex parent material which at 500 Ma underwent a diagenetic or hydrothermal event, which regionally reset Rb–Sr isotope systematics. Subsequently, at 200 Ma the complex went through local diagenetic or hydrothermal re-equilibration, which created domains with slightly different 87Sr/86Sr ratios, before undergoing Alpine high-grade metamorphism and subsequent anatexis. Roughly coeval, restite-rich cordierite dacites show similar, 200 Ma, high-age Rb–Sr isotopic relations, which are interpreted as the age of diagenesis of its sedimentary parent material. This is supported by inherited zircon ion-microprobe ages of 300–400 Ma upwards. Also for these rocks 87Sr/86Sr vs 1/Sr shows linear trends, which are explained analogically by sedimentary component mixing in the parent material of the anatectic source rock complex rather than by magmatic stage mixing or contamination. A sinking slab model is suggested for the regional setting of the crustal anatectic regime, melting being supported by fast uplift (of isotherms) and diapiric underplating by high-temperature asthenospheric mantle. KEY WORDS: inherited isochrons; crustal anatexis; restite–melt repartition; post-collisional magmatism
|
Introduction |
|---|
|
TOPABSTRACTIntroductionGeological SettingHernandez Pyroxene Andesite...Hernandez Rb-Sr Isotopic...Plate Tectonic Setting of...ConclusionsReferences |
|---|
Older Rb–Sr isochron or pseudo-isochron relations among whole-rock samples from younger magmatic complexes have been interpreted either as artefacts produced by magma mixing or contamination (e.g. Faure et al., 1974
; Munksgaard, 1984), or as true isochron relations inherited from source rock complexes (e.g. Munksgaard, 1984). An 87Sr/86Sr vs 1/Sr plot has been suggested as a simple means of discriminating between the two interpretations: a linear relation between the data points would be indicative of a mixing relationship (e.g. Faure, 1986). This criterion has been designed for ‘classical’ magmatic systems involving two magmas, or one magma and a solid contaminant. However, many calc-alkaline rock complexes represent a different type of system. These rock complexes may have been formed from crystal-rich magmas that were produced by crustal anatexis and whose variation is given by source rock complex variation and restite–melt repartitioning (e.g. Zeck, 1968, 1970, 1972, 1992; White & Chappell, 1977; Scambos et al., 1986; Chappell et al., 1987; Burnham, 1992; Chappell, 1996). In these cases, mixing of source rock complex components and of melt and restite material are inherent features of the primary magmagenetic process, and 87Sr/86Sr vs 1/Sr and similar criteria are less effective in excluding the inherited isochron option. The present paper presents data on some rock units from the Neogene Alborán volcanic province in southern Spain, which seem to offer a better case than most for source rock complex inherited Rb–Sr isotopic relations. |
Geological Setting |
|---|
|
TOPABSTRACTIntroductionGeological SettingHernandez Pyroxene Andesite...Hernandez Rb-Sr Isotopic...Plate Tectonic Setting of...ConclusionsReferences |
|---|
The Alborán volcanic province comprises post-collisional volcanic complexes and minor leucogranitic intrusions within the framework of the Alpine Betic–Rif orogen. Magmatism is of Neogene age and is restricted to a SW–NE trending belt running from North Africa via the island of Alborán to SE Spain, a distance of 500–600 km (Fig. 1). The bulk of the volcanic rocks consist of pyroxene andesites, biotite–hornblende andesites, hornblende–biotite dacites and biotite rhyolites of (high-K) calc-alkaline character ranging in age from
17 to 5 Ma (Burdigalian–Messinian), with a peak at 12–10 Ma (e.g. Fúster, 1956; Fúster et al., 1965; López Ruiz & Rodríguez Badiola, 1980; Bellon et al., 1983; Bordet, 1985; Di Battistini et al., 1987; Serrano, 1992). Restite-rich almandine-bearing cordierite dacites form a spectacular, though not voluminous element in the volcanic spectrum (Zeck, 1968, 1970, 1972, 1992; Munksgaard, 1984; De Larouzière et al., 1988). Ultrapotassic rocks and alkali basalts are volumetrically unimportant and are generally younger, Messinian to Pliocene or Quaternary (e.g. Nobel et al., 1981; Bellon et al., 1983). There is an indication of regional zoning (Fig. 1), with calc-alkaline complexes predominantly located in the central part of the belt, and the ultrapotassic and alkaline rocks occurring at its northern and southern extremities. These differences in age and regional distribution indicate two separate magmatic regimes.
|
The origin of the Neogene Alborán volcanic province is in dispute. Araña & Vegas (1974)
and Torres-Roldán et al., (1986) suggested magma genesis by melting of oceanic crust in one or even two coeval subduction zones. Hernandez et al. (1987), De Larouzière et al. (1988), and Martín Escorza & López-Ruiz (1988) suggested magma generation by regional shear heating in large shear zones. Zeck (1968, 1970, 1992) proposed magma production by crustal anatexis within the Betic lithosphere in a high-temperature regime supported by uplift of isotherms and diapiric underplating by high-temperature sub-lithospheric mantle (Zeck, 1996a). The present paper concentrates on a pyroxene andesite unit from the Las Negras area in the central part of the Cabo de Gata area (Fig. 1). It discusses specifically its Rb–Sr isotopic relations within the context of a wide range of mineralogic, geochemical and isotopic data. A comparison is made with a number of roughly coeval occurrences of cordierite dacites within the same volcanic belt that show similar Rb–Sr isotopic relations.
|
Hernández Pyroxene Andesite Formation |
|---|
|
TOPABSTRACTIntroductionGeological SettingHernandez Pyroxene Andesite...Hernandez Rb-Sr Isotopic...Plate Tectonic Setting of...ConclusionsReferences |
|---|
The pyroxene andesites—here called the Hernández pyroxene andesite formation—form a well-defined lithostratigraphic unit (Cela, 1968
; Bordet, 1985). It occurs within an area of 20–25 km2, ranging in thickness from 10 to 200 m, and is both underlain and overlain by volcaniclastic sedimentary rocks; the underlying horizon is locally calcareous and fossiliferous. Unlike many of the volcanic rocks above and below, the pyroxene andesites are usually not altered. The rocks consist of a black to very dark grey glassy matrix containing crystals of green or brown pyroxene and white plagioclase up to several millimetres in size, both in isolated euhedral crystals and in small crystal aggregates. Most exposures are of block and ash facies, with smaller intrusions of massive rocks, indicative of a rather viscous magma; there are no differences in lithology, mineralogy or chemistry between these two volcanic facies. Hernández formation appears as a continuous eruption unit; there are no epiclastic intercalations or oxidized horizons that could suggest important breaks in eruption activity.
Petrography and mineralogy
In thin section Hernández pyroxene andesite matrix (30–40 vol. %) is shown to consist of glass containing euhedral, 5–20 µm crystals of plagioclase, Ca-poor pyroxene, Ca-rich pyroxene and Ti-magnetite. Euhedral, 30 µm–1 mm (micro-)phenocrysts and 0.5–3 mm aggregates of the same four minerals are enclosed within the matrix. Glass and minerals were analysed by microprobe; representative glass compositions are indicated in Fig. 3, and mineral compositions are given in the text and Fig. 3.
|
Most crystalline material reflects a common calc-alkaline magmatic development. Plagioclase crystals show a fine euhedral oscillatory zoning (Fig. 2d and e) within the range of An80–75; 5–10 µm wide rims are more sodic (
An50). Ca-poor pyroxene crystals show euhedral zoning (Fig. 2f) in the range En60–66Fs40–30Wo0–4); 5–20 µm rims are more Fe- and Ca-rich (En50Fs40Wo10). Ca-rich pyroxene shows euhedral zoning (En40–45Fs20–15Wo40) and Fe-rich and Ca-poor rim and microlite compositions (En45Fs25Wo30). Crystals of Ti-magnetite vary from 10 to 300 µm and range in composition from Mgt65Usp35 to Mgt40Usp60.
|
There are also indications of ultra-metamorphism. These are best demonstrated in the common occurrence of crystal aggregates consisting of 2–10 crystals of
0.1–1 mm in size, which have the same compositions as indicated above for the isolated crystals. However, the internal texture of the aggregates is characterized by anhedral crystal outlines (Fig. 2g), indicating that the crystals did not grow freely from a melt phase, but competed for space; some interstitial glass pools may be present. Externally, towards the matrix, euhedral crystal outlines are maintained. Some plagioclase crystals, both within the aggregates and isolated within the matrix, show large anhedral cores, which are densely clouded by glass inclusions rimmed by thin calcic-sodic rims (Fig. 2a–c), indicative of an early, pre-eruptional stage of incongruent melting.
Geochemistry, Sr and Nd isotopes
Analytical data for Hernández pyroxene andesite formation are given in Tables 1 and 2. Sample size of laboratory crushed material ranged from 5 to 8 kg; this represents 15–40 dm3 of outcrop rock, which in the field was cleaned of weathering and other alteration effects by hammer chipping and broken into fragments of a size suitable for the jaw-crusher. Its calc-alkaline character is demonstrated in current classification diagrams such as the AFM diagram (Irvine & Baragar, 1971). Whole-rock major element compositions are compared with mineral and glass compositions in Harker diagrams (Fig. 3). 87Sr/86Sr ratios range from 0.7096 to 0.7125. Figure 4 shows the data in a Nicolaysen diagram defining a correlation line indicative of an age of 509 ± 62(2) Ma [mean square weighted deviation (MSWD) = 14.64; initial ratio (IR) = 0.7056 ± 6]. Uneven distribution of data points along the regression line defines two data point clusters, each with four samples; one sample (96Z53) plots between the two clusters. The lower of these data point clusters defines a secondary regression line of 202 ± 30(2) Ma (MSWD = 0.86; IR = 0.7082 ± 3). The data cluster higher on the primary regression line suggests a similar slope, but has little Rb/Sr variation and consequently a poorly constrained age indication [153 ± 146(2) Ma; MSWD = 4.41; IR = 0.7104 ± 19]. 143Nd/144Nd ratios range from 0.51228 to 0.51244. Figure 5 shows the Hernández formation data in a 143Nd/144Nd–147Sm/144Nd diagram defining a 1.5 ± 0.4(2) Ga (MSWD = 1.14) regression line.
|
|
|
|
|
Hernández Rb–Sr Isotopic Relations |
|---|
|
TOPABSTRACTIntroductionGeological SettingHernandez Pyroxene Andesite...Hernandez Rb-Sr Isotopic...Plate Tectonic Setting of...ConclusionsReferences |
|---|
The variation in 87Sr/86Sr ratios (Fig. 4) shown by Hernández pyroxene andesites, which have an extrusion age of
10 Ma (see above), can hardly be explained as a result of magmatic fractionation processes, as these would result in an 10 Ma isochron line, that is, nearly constant Sr-isotopic ratios. Not only is the variation much larger than that, it is also systematic, defining a 509 ± 62(2) Ma (Cambrian–Ordovician) isochron regression line. Two concordant data point clusters define two subsets of data, one of which constrains a 202 ± 30(2) Ma (Late Triassic–Early Jurassic) regression line. The petrogenetic significance of these secondary regression lines is greatly enhanced by statistically parallel age lines for three roughly coeval occurrences of cordierite dacites from the same volcanic belt (Munksgaard, 1984) (see below and Fig. 4). Two alternative explanations are available for the high-age isochron relations: (1) they reflect isotopic relations inherited by anatectic magma from its source rock complex, or (2) they are isochron artefacts produced by mixing of genetically unrelated reservoirs in the magmatic stage. The inherited isochron hypothesis requires a magma generation process in which either heterogeneous source rock complexes were partly melted and the resulting crystal-rich magma escaped subsequent homogenization and restite–melt repartition, or a closely similar process in which anatectic crystal-rich magmas from the same environment were formed essentially separately and subsequently mingled, preserving their original restite–melt parageneses. As success of arrestingthese processes depends on the sampling scale, relatively large samples were taken (see above).
Below, a number of arguments will be given that support the inherited isochron hypothesis. Many geochemical and isotopic relations are consistent with both genetic models, as anatexis of a heterogeneous source rock complex and magma mixing or magma contamination by a solid contaminant may produce similar, and thus non-discrimatory plot patterns. Therefore evidence from a wider disciplinary spectrum is necessary to resolve the question.
Inherited isochron hypothesis
During anatexis, restite material coexisting with the melt may comprise both crystalline material from the source rock that did not take part in the melting process and crystals formed by incongruent melting of source rock minerals. The high-temperature anatectic regime will promote local isotopic equilibrium between restite and melt; this implies that restite–melt repartitioning, which has been suggested as a major diversifying mechanism for magmatic complexes in anatectic settings (e.g. White & Chappell, 1977; Chappell, 1996), cannot have produced the Sr-isotopic variation seen in the Hernández formation (Fig. 4). However, on a somewhat larger scale, Sr isotopic variation within the crystal–melt magma body is controlled by Sr-isotopic inhomogeneity of the source rocks, which is mainly dependent on variation in Rb/Sr and Sr isotopic ratios within the source rock complex at the time of melting. Whole-rock samples from igneous complexes derived from such restite–melt magmas should reflect this Rb–Sr isotopic variation, provided that effective restite–melt repartitioning on a larger than sample scale has not taken place during subsequent magma migration. The resulting intrusive/eruptive magmatic rock complex may thus reflect the isotopic relations that the high-grade metamorphic source rock complex would have shown had this not been partially melted.
Dehydration melting experiments
Calculations based on known dehydration reactions of hornblende and biotite (Clemens & Vielzeuf, 1987) show that under fluid-absent conditions common amphibolitic rocks may produce 50–70 vol. % melt at temperatures of 900–1000°C and P = 5 kbar. Experimental dehydration melting of amphibolitic rocks of intermediate and mafic compositions (e.g. Rushmer, 1991; Wyllie & Wolf, 1993; Wolf & Wyllie, 1994; Patiño Douce & Beard, 1995; Rapp & Watson, 1995) confirms these high melt percentages. Patiño Douce & Beard (1995) and Rapp & Watson (1995) reported that, at 5 kbar and 1000°C, amphibolitic compositions (hbl, plg, qu ± bi) may yield 30–50 vol. % melt. Melt forms at the expense of hornblende (biotite), quartz and plagioclase, and the complementary restite parageneses comprise mainly plagioclase, Ca-rich pyroxene, Ca-poor pyroxene, ilmenite and magnetite. Melt compositions for starting materials with 60 wt % SiO2 and 0.5 wt % K2O (‘quartz amphibolite’, Patiño Douce & Watson, 1995) are characterized by 75 wt % SiO2, 14 wt % Al2O3, 3 wt % K2O, 3 wt % CaO and 3 wt % FeO + MgO. Glass matrix in Hernández andesites has similar compositions (Fig. 3), but with somewhat lower CaO and Al2O3 contents. However, as the anatectic magma is suggested to have entrained its restite material, and as also crystalline material has precipitated from the melt, neither whole-rock compositions nor glass compositions are identical to those of primary melts.
The abundant plagioclase–Ca-rich pyroxene–Ca-poor pyroxene–Ti-magnetite crystal aggregates within Hernández andesites are mineralogically identical to the restite material produced in the dehydration melting experiments described above. Also, their anhedral, metamorphic texture (Fig. 2g) suggests a restite origin rather than an origin by phenocryst accumulation or synneusis. The magmatic textural aspects that are also present in these aggregates—such as euhedral zoning and habit relative to surrounding matrix, and intergranular glass pockets—agree well with the transitional metamorphic–magmatic, or ultra-metamorphic, character of the anatectic environment.
Geophysical and geological framework
Seismic studies by Torné & Banda (1992) indicate a present thickness of 15–30 km for the Betic crust. As volcanism, at 15–5 Ma, post-dates main extensional crustal thinning (Zeck, 1996a, 1997), such crustal thickness constrains pressures of melting in the lower part of the crust to the range of 4–9 kbar. As the region is characterized by high heat flow since the late decompressional stage related to the extensional tectonics (Balanyá et al., 1993; Tubía, 1994; Zeck, 1996a, 1997), temperatures of 1000°C in the lower crust represent a feasible estimate. If melting was realized in a contact anatectic setting, as will be argued below, temperatures of 1000°C are also conceivable in higher parts of the crustal section. Thus, pressure and temperature conditions required for the dehydration melting discussed above are realistic during the relevant time span for the basement underlying the volcanics.
The crustal lithospheric section contains a pile of Alpine orogenic nappes each with a pre-Permo-Triassic basement and an Early Mesozoic cover (e.g. Nijhuis, 1964; Egeler & Simon, 1969; Priem et al., 1969; De Jong, 1991, 1993; Puga et al., 1995). The basement consists mainly of garnet-mica schists and quartzites, the cover sequence mainly of mica schists or phyllites, carbonates and mafic rocks. Moreover, the deeper parts of the crustal lithosphere may contain tectonically inducted Tethyan oceanic material, which is not represented in the Betic nappe complexes exposed at present (Zeck, 1996a, 1997). Thus, the crustal lithosphere underlying the Neogene volcanics has suitable source rocks for producing 200 and 500 Ma inherited ages in anatexites, as is suggested by the inherited isochron model.
Regional isotopic variation
Rb–Sr isotopic heterogeneity of the Hernández formation is shown in Fig. 4 and was discussed above. Comparison of sample locations for the two data point clusters shows a systematic regional distribution difference. Figure 6a shows a geographic grid for the area where the Hernández formation is exposed, with samples numbered from 1 (lowest) to 9 (highest) according to their position on the 87Sr/86Sr regression array of Fig. 6b (see Fig. 4); also indicated are TCHURNd model ages for each sample. Figure 6a shows that the nine analysed Hernández samples occur within an 6 km long, SW–NE trending area with higher 87Sr/86Sr and Rb/Sr ratios and TCHURNd model ages in a domain occupying the NE part of the area (samples 96Z54, -55, -56 and -59) and corresponding lower values in a domain in the SW part of the area (samples 96Z51, -52, -57 and -58). Major and trace element variations mimic this domain pattern, with, for example, higher SiO2, K2O, Rb, Ce, Nd, Sm and Zr, and lower FeO*, MgO, Sr and V in the NE domain. For some parameters sample 96Z53 is similar to the NE grouping, but for others it is intermediate between the two groups (see Figs 4 and 6b). Sm–Nd isotopic relations also reflect this two-domain pattern (Fig. 5), with the SW domain showing systematically higher 143Nd/144Nd ratios than the NE domain, which here includes sample 96Z53.
|
The individual character of the two domains within Hernández formation is brought out perhaps most clearly in 87Sr/86Sr vs 1/Sr and 143Nd/144Nd vs 1/Nd diagrams (Figs 7 and 8). Both figures show Hernández formation data points for each domain in a well-constrained linear arrangement (sample 96Z53 has a separate position in Fig. 7, as in Figs 4 and 6b), indicating two-component mixing within each domain (see Faure, 1986
). The nature of the two-component mixing will be discussed below.
|
|
U–Pb and Sm–Nd age support
Involvement of older crustal rocks in the petrogenesis of Neogene Hernández pyroxene andesites, which is suggested by the
500 Ma Rb–Sr isotopic regression line (Fig. 4), is supported by inherited zircon ion-microprobe ages ranging from 500 Ma to 1800 Ma (Zeck et al., 1998). The inherited zircon is in rounded grains consistent with a sedimentary derivation.
TCHURNd crustal derivation ages are also high: 760–1070 Ma (Table 2, Fig. 6a), again supporting involvement of older crustal material. Further interpretation of these model ages appears difficult, as they would indicate average crustal derivation ages for variable batches of supracrustal parent material of the source rock complex based on comparison with the CHUR model mantle evolution line. Sm–Nd isotopic relations (Fig. 5) for the whole data population (nine samples) define a line of 1.5 ± 0.4(2) Ma (MSWD = 1.14). As for the Rb–Sr isotopic diagram (Fig. 4), the Sm–Nd isotopic data for each of the two domains suggest lower slope resetting lines, but the large scatter does not permit a meaningful age estimate.
Coeval anatectic cordierite dacites
Early Palaeozoic and Early Mesozoic Rb–Sr isochron relations have been reported earlier for four occurrences of Neogene peraluminous cordierite dacites from the Alborán volcanic province (Munksgaard, 1984; locations are given in Fig. 1): Cerro Hoyazo, Vera (30 km NE of C° Hoyazo), Mazarrón and Mar Menor (both 70 km N of C° Hoyazo). These occurrences (Zeck, 1968, 1970, 1992; Munksgaard, 1984) form relatively small bodies and consist mainly of magmatic phases (chiefly glass, plagioclase, cordierite and biotite) and abundant restite material (sillimanite, cordierite, almandine, biotite, quartz, plagioclase, andalusite, spinel and graphite); muscovite and K-feldspar are notably absent, and individual restite rock inclusions have either quartz or plagioclase, never the combination. The striking character of the restite material in this case leaves little doubt of an anatectic origin with source rock complexes predominantly of pelitic to quartzo-feldspathic clastic derivation (Zeck, 1968, 1970, 1992). Oxygen isotope compositions of 13.0–16.2%° 
18O for whole-rock Hoyazo dacite and its predominant, graphite-bearing almandine–biotite–fibrolite restite rock inclusions (Munksgaard, 1984) effectively support the model.
Rb–Sr isotopic relations for the four cordierite dacites were given by Munksgaard (1984) and are included in the Nicolaysen diagram of Fig. 4. Three of the occurrences show statistically indistinguishable Late Triassic to Early Jurassic isochron ages: 210 ± 17 Ma (2.40), 199 ± 58 Ma (1.84) and 173 ± 23 Ma (5.54), (2 errors; MSWD values in parentheses) and have initial Sr isotopic ratios of 0.7100 ± 3, 0.7125 ± 12 and 0.7118 ± 6, respectively; variation in initial Sr isotopic ratios is somewhat larger than the quoted errors and may reflect differences in sedimentary provenance for Mazarrón and Mar Menor vs C° Hoyazo. The fourth occurrence, Vera, shows a correlation line indicative of a Late Cambrian age of 535 ± 22 Ma (8.42) and an initial ratio of 0.7061 ± 3. The isochron ages are interpreted as the age of diagenesis of the original sedimentary material, as the diagenetic stage, being characterized by effective, large-scale fluid circulation, offers the possibility of wholesale isotopic equilibration and establishment of a constant Sr-isotope ratio throughout large bodies of sediment (e.g. Roddick & Compston, 1977; Zeck & Hansen, 1988). Zircon ion-microprobe dating of Hoyazo whole rock and its restite inclusions (Williams & Zeck, in preparation) revealing inherited zircon with ages of 300–400 Ma upwards, effectively supports this model.
The exposure size of each of the four cordierite dacite occurrences is of the order of 1 km2 without signs of lithological heterogeneity. However, the existence of the high-age isochron lines (Fig. 4) implies that effective homogenization did not occur within these relatively small magmatic bodies. This is in agreement with the inherited isochron hypothesis outlined above, which implies that homogenization and restite–melt repartition within the heterogeneous anatectic magmatic mass, showing a variation of Sr contents and isotopic ratios, did not take place. These relations are illustrated in the 87Sr/86Sr vs 1/Sr diagram of Fig. 7. For each of the four cordierite dacites the diagram shows a linear relation between samples showing higher 87Sr/86Sr and lower Sr, and samples with lower 87Sr/86Sr and higher Sr. Source rocks derived from clastic material consisting of mixtures of clay and sand (layers) will show variable Rb and Sr contents—according to Turekian & Wedepohl (1961) clays have typically Rb/Sr 0.5 (140/300 ppm), quartz-rich sandstones have Rb/Sr 3 (60/20 ppm), whereas carbonates, which might have been present in minor amounts, have Rb/Sr 0.005 (3/610 ppm). Given the time that, according to the anatectic model, elapsed between Late Triassic to Early Jurassic diagenesis and Neogene anatexis, concomitant differences in 87Sr/86Sr ratio would develop, and the relations shown in Fig. 7 may be produced in the anatectic magma and the resulting volcanic rock.
Anatectic model for Hernández formation
Compared with the cordierite dacites, the development of a petrogenetic model for Hernández pyroxene andesite formation involves more uncertainties because its mineralogy and chemistry are less revealing. Above, on the basis of a comparison with experimental data, a predominantly amphibolitic source rock complex was suggested for the Alpine anatectic event producing Hernández magma. A variety of isotopic and chemical parameters presented above suggest that the source rock complex was inhomogeneous, including two domains, each of which may be described by two-component mixing (Figs 4–8).
Abundant hornblende in the anatectic source rock complex would suggest a major mafic component. The presence of inherited sedimentary zircons indicates terrigenous clastic material in the parent material. As the source rock complex is part of the Alpine nappe pile, tectonically produced lithologic heterogeneities cannot be excluded. Proterozoic ages for inherited zircons and whole-rock Sm–Nd isotopic relations indicate considerable involvement of older rock material.
The inferred source complex parent material characteristics—in part of sedimentary and in part of mafic igneous character, showing Proterozoic inheritance ages and the 500 Ma Rb–Sr event—are perhaps best met by a submarine volcano-sedimentary complex, which at 500 Ma underwent large-scale Sr isotopic equilibration of diagenetic or water–rock interaction character (possibly related to the roughly coeval diagenetic event suggested for Vera cordierite dacite protolith). The suggested 1–2 km scale domainal character of Hernández source rock complex and its isotopic mixing systematics (Figs 7 and 8) may reflect sedimentary component mixing controlled by variations in volcanic or sedimentary input and sedimentary provenance.
The secondary regression lines on the 500 Ma Hernández isochron line (Fig. 4) suggest an 200 Ma event during which Rb–Sr isotope systematics of the protolith for the Alpine anatectic source rock complex were reset within both compositional domains. As for the cordierite dacites, which show evidence for a coeval event, the 200 Ma recrystallization event may have been of diagenetic or hydrothermal character.
|
Plate Tectonic Setting of the Crustal Anatexis |
|---|
|
TOPABSTRACTIntroductionGeological SettingHernandez Pyroxene Andesite...Hernandez Rb-Sr Isotopic...Plate Tectonic Setting of...ConclusionsReferences |
|---|
A sinking slab model has recently been suggested to explain the later stages of the Alpine Betic–Rif orogeny (Zeck, 1996a
, 1996b, 1997). According to the model, orogeny started in a subduction zone system dipping NW under eastward drifting Iberia, active since 80? Ma. Initially the system processed Mesozoic Tethyan oceanic lithosphere; subduction reached its final stage when continental lithosphere from the Tethys realm (Betic–Ligurian lithosphere) became involved (Zeck, in press). After cessation of subduction, the subducted slab steepened, broke off shortly before 22 Ma (Zeck, 1996a) and sank rapidly into the sub-lithospheric mantle (Fig. 9). Sinking of the slab caused diapiric in-flow of high-temperature asthenospheric material into the widening gap above the sinking slab. This, and rapid uplift (of isotherms), caused a high-temperature regime in the overlying Betic lithosphere, creating the framework required for the anatectic–inherited-isochron model.
|
Two options are available for the anatectic setting (Zeck, 1968
, 1970, 1972): (1) regional anatexis controlled by a regional thermal diffusion regime, and (2) contact anatexis, where the crucial heat transfer is by rising magma bodies. The present paper suggests roughly coeval anatexis in rock complexes of very different lithology and age, and thus different melting temperatures at different levels in the crustal section. These relations favour a contact anatectic setting where the nature of the produced secondary anatectic magma series may vary widely and is basically determined by the nature of the rock complex that happens to be intruded by the primary magma. The relatively small time gap between slab break-off (shortly before 22 Ma; Zeck, 1996a) and oldest magmatism (20 Ma, Bellon et al., 1983; Zeck et al., 1989) also favours a contact anatectic setting, as a regional thermal diffusion model would require a longer delay. |
Conclusions |
|---|
|
TOPABSTRACTIntroductionGeological SettingHernandez Pyroxene Andesite...Hernandez Rb-Sr Isotopic...Plate Tectonic Setting of...ConclusionsReferences |
|---|
High-age Rb–Sr isotopic relations in five well-defined volcanic rock formations within the Neogene Alborán volcanic province are interpreted as source rock inherited. The hypothesis implies the formation of crystal-rich anatectic magmas which did not go through a stage of effective restite–melt repartitioning. Chemical and isotopic variations in rocks derived from such magmas reflect the relations that the anatectic source rock complex would have shown had it not been partially melted.
Very different lithologies—cordierite dacites and pyroxene andesites—show statistically parallel, high-age regression lines in an 87Sr/86Sr-87Rb/86Sr diagram (Fig. 4). Two trends and ages are present: 500 Ma and 200 Ma, with initial Sr isotopic ratios varying widely, from 0.7055 to 0.7125. Such relations do not favour an interpretation of these regression lines as artefacts produced by mixing of genetically unrelated reservoirs, which is the alternative genetic option, as this would require too closely tuned compositional variations in a large number of mixing reservoirs. Four (five) pairs and two pairs of reservoirs, respectively, have to vary in conjunction to produce the two sets of statistically parallel lines (Fig. 4), and this seems highly improbable. In contrast, the inherited isochron model explains the parallellism of the isochron lines as a result of the geologically well-established common time-stratigraphy of the involved volcanic rocks and their source rock complexes in the underlying basement.
The other features discussed above are all consistent with the anatectic model. The timing and tectonic setting of magma production, melting processes and melting conditions, as well as ages and lithologies of source rocks, are all embraced by the geodynamic framework provided by the sinking slab model, but do not specifically disprove a magmatic stage mixing model. An example of such non-unique information is given by the linear data point relations shown in 87Sr/86Sr vs 1/Sr and 143Nd/144Nd vs 1/Nd diagrams (Figs 7 and 8). The diagrams indicate two-component mixing (e.g. Faure, 1986), although without discriminating between magma stage ‘mixing’ (magma mixing or magma contamination with a solid contaminant) or supracrustal stage mélanges, involving, for example, clastic sediments, clay and sand, and mafic volcanics, in the parent material of the anatectic source rock complex, as cause for the data plot patterns.
|
Acknowledgements |
|---|
The paper benefited considerably from detailed and constructive reviews by John W. Sheraton (Canberra) and Charles R. Bacon (Menlo Park). We thank Carlos Sanz de Galdeano (Granada), Poul Andriessen (Amsterdam), Henri Maluski (Montpellier), Hermelindo Castro Nogueira and José S. Guirado Romero (Agencia de Medio Ambiente, Junta de Andalucía, Almería) for pleasant co-operation, and Eirik Krogstad and Stefan Bernstein (Copenhagen) for comments on the manuscript. This work was supported by Danish Research Council (SNF) grant 11–0287 and Carlsberg Foundation grant 95–7865 to H.P.Z.
|
FOOTNOTES |
|---|
Dedicated to Professor Ikuo Kushiro on the occasion of his retirement 31 March 1999 as director of the Institute for Study of the Earth's Interior, Okayama University at Misasa. 
* Corresponding author. Present address: Institute for Study of the Earth's Interior, Okayama University at Misasa, Tottori-Ken 682-01, Japan Telephone: 81 858 43 3876. Fax: 81 858 43 2184. email: zeck{at}misasa.okayama-u.ac.jp
|
References |
|---|
|
TOPABSTRACTIntroductionGeological SettingHernandez Pyroxene Andesite...Hernandez Rb-Sr Isotopic...Plate Tectonic Setting of...ConclusionsReferences |
|---|
Araña V., Vegas R. Plate tectonics and volcanism in the Gibraltar arc. Tectonophysics (1974) 24:197–212.[Web of Science]
Balanyá J. C., Azañon J. M., Sánchez-Gómez M., García-Dueñas V. Pervasive ductile extension, isothermal decompression and thinning of the Jubrique unit in the Paleogene (Alpujarride Complex, W Betics, Spain). Comptes Rendus des Séances de l'Académie des Sciences, Paris, Série II (1993) 304:929–932.
Bellon H., Bordet P., Montenat C. Chronologie du magmatisme neogène des Cordillères bétiques (Espagne méridionale). Bulletin de la Société Géologique de France (1983) 25:205–217.[Web of Science]
Blanco M. J., Spakman W. The P-wave velocity structure of the mantle below the Iberian peninsula: evidence for subducted lithosphere below southern Spain. Tectonophysics (1993) 221:13–34.[Web of Science]
Bordet P. Le volcanisme miocène des Sierras de Gata et de Carboneras. Institut Géologique Albert de Lapparent, Documents et Travaux (1985) 8:1–70.
Burnham C. W. Calculated melt and restite compositions of some Australian granites. Transactions of the Royal Society of Edinburgh, Earth Science Section (1992) 83:387–397.
Cela S. V. Estudio petrológico de las sucesiones volcánicas del sector central de la formación del Cabo de Gata (Almería). Estudios Geológicos (1968) 24:1–38.
Chappell B. W. Magma mixing and the production of compositional variation within granite suites: evidence from the granites of southeastern Australia. Journal of Petrology (1996) 37:449–470.
Chappell B. W., White A. J. R., Wyborn D. The importance of residual source material (restite) in granite petrogenesis. Journal of Petrology (1987) 28:1111–1138.
Clemens J. D., Vielzeuf D. Constraints on melting and magma production in the crust. Earth and Planetary Science Letters (1987) 86:287–306.[Web of Science]
Comas M. C., García-Dueñas V., Jurado M. J. Neogene tectonic evolution of the Alborán sea from MCS data. Geo-Marine Letters (1992) 12:157–164.
De Jong K. Tectono-metamorphic studies and radiometric dating in the Betic Cordilleras (SE Spain)—with implications for the dynamics of extension and compression in the western Mediterranean area. (1991) Christian University, Amstelveen. 1–204. Ph.D. Thesis.
De Jong K. The tectono-metamorphic and chronologic development of the Betic Zone (SE Spain) with implications for the geodynamic evolution of the western Mediterranean area. Verhandelingen der Koninklijke Nederlandse Akademie van Wetenschappen (1993) 96:295–333.
De Larouzière F. D., Bolze J., Bordet P., Hernandez J., Montenat C., Ott d'Estevou P. The Betic segment of the lithospheric Trans-Alboran shear zone during the Late Miocene. Tectonophysics (1988) 152:41–52.[Web of Science]
Di Battistini G., Toscani L., Iaccarino S., Villa I. M. K/Ar ages and the geological setting of calc-alkaline volcanic rocks from Sierra de Gata, SE Spain. Neues Jahrbuch für Mineralogie, Monatshefte (1987) 1987:369–383.
Egeler C. G., Simon O. J. Sur la tectonique de la Zone Bétique (Cordillères Bétiques, Espagne). Verhandelingen der Koninklijke Nederlandse Akademie van Wetenschappen, Afdeling Natuurkunde, 1st serie (1969) XXV(3):1–90.
Faure G. Principles of Isotope Geology (1986) New York: John Wiley.
Faure G., Bowman J. R., Elliot D. H., Jones L. M. Strontium isotope composition and petrogenesis of the Kirkpatrick basalt, Queen Alexandra Range, Antarctica. Contributions to Mineralogy and Petrology (1974) 48:153–169.[Web of Science]
Fúster J. M. Las erupciones delleníticas del Terciario superior de la fosa de Vera (provincia de Almería). Boletin de la Real Sociedad Española de Historia Natural (1956) 54:53–88.
Fúster J. M., Aguilar M. J., García A. Las sucesiones volcánicas en la zona del Pozo de los Frailes dentro del vulcanismo cenozoico del Cabo de Gata (Almería). Estudios Geológicos (1965) 21:199–222.
Govindaraju K. 1994 compilation of working values and sample description for 383 geostandards. Geostandards Newsletter (1994) 18(special issue):1–158.
Hernandez J., De Larouzière F. D., Bolze J., Bordet P. Le magmatisme néogène bético-rifain et le couloir de décrochement trans-Alboran. Bulletin de la Société Géologique de France (1987) 8(III):257–267.[Medline]
Irvine T. N., Baragar W. R. A. A guide to the chemical classification of the common volcanic rocks. Canadian Journal of Earth Science (1971) 8:523–548.
López Ruiz J., Rodríguez Badiola E. La región volcánica de sureste de España. Estudios Geológicos (1980) 36:5–63.
Martín Escorza C., López-Ruiz J. Un modelo geodinámico para el volcanismo neogeno del sureste iberico. Estudios Geológicos (1988) 44:243–251.
Munksgaard N. C. High 18O and possible pre-eruptional Rb—Sr isochrons in cordierite-bearing volcanics from SE Spain. Contributions to Mineralogy and Petrology (1984) 87:351–358.[Web of Science]
Nijhuis H. J. Plurifacial Alpine metamorphism in the south-eastern Sierra de los Filabres south of Lubrín, SE Spain (1964) Proefschrift, Amsterdam University. 1–151.
Nobel F. A., Andriessen P. A. M., Hebeda E. H., Priem H. N. A., Rondeel H. E. Isotopic dating of the post-Alpine Neogene volcanism in the Betic Cordilleras, southern Spain. Geologie en Mijnbouw (1981) 60:209–214.
Patiño Douce A. E., Beard J. S. Dehydration-melting of biotite gneiss and quartz amphibolite from 3 to 15 kbar. Journal of Petrology (1995) 36:707–738.
Puga E., Diaz de Federico A., Demant A. The eclogized pillows of the Betic Ophiolite Association: relics of the Tethys ocean floor incorporated in the Alpine chain after subduction. Terra Nova (1995) 7:31–43.[Web of Science]
Priem H. N. A., Boelrijk N. A. I. M., Hebeda E. H., Verschuere R. H. Isotopic age determinations on tourmaline granite-gneisses and a metagranite in the eastern Betic Cordilleras (south-eastern Sierra de Los Filabres), S.E. Spain. Geologie en Mijnbouw (1969) 45:184–187.
Rapp R. P., Watson E. B. Dehydration melting of metabasalt at 8–32 kbar: implications for continental growth and crust-mantle recycling. Journal of Petrology (1995) 36:891–931.
Roddick J. C., Compston W. Strontium isotopic equilibration, a solution to a paradox. Earth and Planetary Science Letters (1977) 34:238–246.[Web of Science]
Rushmer T. Partial melting of two amphibolites: contrasting experimental results under fluid absent conditions. Contributions to Mineralogy and Petrology (1991) 107:41–59.[Web of Science]
Scambos T. A., Loiselle M. C., Wones D. R. The Centre Pond pluton: the restite part of the story (phase separation and melt evolution in granitoid genesis). American Journal of Science (1986) 286:241–280.
Serrano F. Biostratigraphic control of Neogene volcanism in Sierra de Gata (south-east Spain). Geologie en Mijnbouw (1992) 71:3–14.
Shibata T. The lead, strontium and neodymium isotopic studies on the Quaternary volcanic rocks from northeastern Japan: implications for the slab–mantle interaction in subduction zones. (1992) Okayama: Okayama University. 1–124. Ph.D, Thesis.
Torné M., Banda E. Crustal thinning from the Betic Cordillera to the Alboran Sea. Geo-Marine Letters (1992) 12:76–81.
Torres-Roldán R. L., Polli G., Peccerillo A. An Early Miocene arc-tholeiitic magmatic dyke event from the Alborán Sea—evidence for pre-collisional subduction and back-arc crustal extension in the westernmost Mediterranean. Geologische Rundschau (1986) 75:219–234.
Tubía J. M. The Ronda peridotites (Los Reales nappe): an example of the relationship between lithosphere thickening by oblique tectonics and late extensional deformation within the Betic Cordillera (Spain). Tectonophysics (1994) 238:381–398.[Web of Science]
Turekian K. K., Wedepohl K. H. Distribution of elements in some major units of the earth's crust. Geological Society of America Bulletin (1961) 72:175–192.
White A. J. R., Chappell B. W. Ultrametamorphism and granitoid genesis. Tectonophysics (1977) 43:7–22.[Web of Science]
Wolf M. B., Wyllie P. J. Dehydration melting of amphibolite at 10 kbar: the effects of temperature and time. Contributions to Mineralogy and Petrology (1994) 115:369–383.[Web of Science]
Wyllie P. J., Wolf M. B. Amphibolite dehydration-melting: sorting out the solidus. Geological Society, London, Special Publications (1993) 76:405–416.
Yashikawa M., Nakamura E. Precise isotopic determination of trace Sr in magnesium-rich samples. Journal of Mineralogy, Petrology and Economic Geology (1993) 88:548–561.
Zeck H. P. Anatectic origin and further petrogenesis of almandine-bearing biotite–cordierite–labradorite-dacite with many inclusions of restite and basaltoid material, Cerro del Hoyazo, SE Spain (with data on the adjoining part of the Sierra Alhamilla) (1968) Proefschrift, Amsterdam University. 1–168.
Zeck H. P. An erupted migmatite from Cerro del Hoyazo, SE Spain. Contributions to Mineralogy and Petrology (1970) 26:225–246.[Web of Science]
Zeck H. P. A migmatitic lava from SE Spain. Bulletin Volcanologique (1972) 36:301–302.
Zeck H. P. Restite–melt and mafic–felsic magma mixing and mingling in an S-type dacite, Cerro del Hoyazo, SE Spain. Transactions of the Royal Society of Edinburgh, Earth Science Section (1992) 83:139–144.
Zeck H. P. Betic-Rif orogeny: subduction of Mesozoic Tethys lithosphere under E-ward drifting Iberia, slab detachment shortly before 22 Ma, and subsequent uplift and extensional tectonics. Tectonophysics (1996a) 254:1–16.[Web of Science]
Zeck H. P. Alpine kinematic evolution in the W Mediterranean—mosaic of local extensional tectonic regimes following upon regional W-ward directed subduction of Betic-Ligurian lithosphere. In: The Mediterranean Basins: Tertiary Extension within the Alpine Orogen (1996b) Paris: Cergy-Pontoise University. 34. Workshop, 1996.
Zeck H. P. Mantle peridotites outlining the Gibraltar Arc—centrifugal extensional allochthons derived from the earlier Alpine, westward subducted nappe pile. Tectonophysics (1997) 281:195–207.[Web of Science]
Zeck H. P. Alpine plate kinematics in the western Mediterranean—a W-ward directed subduction regime followed by slab roll-back and slab detachment. In: Geological Society, London, Special Publications (1999) in press.
Zeck H. P., Hansen B. T. Rb–Sr mineral ages for the Grenvillian metamorphic development of spilites from the Dalsland Supracrustal Group, SW Sweden. Geologische Rundschau (1988) 77:683–692.
Zeck H. P., Albat F., Hansen B. T., Torres-Roldán R. L., García-Casco A. Alpine tourmaline-bearing muscovite leucogranites, intrusion age and petrogenesis, Betic Cordilleras, SE Spain. Neues Jahrbuch für Mineralogie, Monatshefte (1989) 1989:513–520.
Zeck H. P., Kristensen A. B., Williams I. S. Post-collisional volcanism in a sinking slab setting—crustal anatectic origin of pyroxene-andesitic magma, Caldear Volcanic Group, Neogene Alborán volcanic province, SE Spain. Lithos (1998) 45:499–522.[Web of Science]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||