Journal of Petrology Advance Access originally published online on June 18, 2008
Journal of Petrology 2008 49(7):1365-1396; doi:10.1093/petrology/egn029
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Multiple Pulses of the Mantle Plume: Evidence from Tertiary Icelandic Lavas
The Pheasant Memorial Laboratory for Geochemistry & Cosmochemistry, Institute for Study of the Earth's Interior, Okayama University, Misasa, 682-0193, Japan
RECEIVED MARCH 17, 2007; ACCEPTED MAY 26, 2008
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONGEOLOGICAL SETTING AND SAMPLESANALYTICAL METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
We present major and trace element concentrations and Sr–Nd–Hf–Pb isotope data for the c. 13–2 Ma Tertiary lavas from eastern Iceland. Our new geochemical results, together with published geological, geochronological, geochemical and geophysical data, are used to evaluate temporal changes in mantle sources contributing to the Tertiary Icelandic magmatism and the relative roles of these sources in magma productivity. The trace element and radiogenic isotopic compositions clearly distinguish three distinct end-member components in the Tertiary magmatism. Temporal variations in lava geochemistry can be attributed to changes in the relative contributions of these three end-member components to the erupted magmas and correlated with temporal variations in magma productivity. The extrusion of lavas with geochemically and isotopically enriched compositions was particularly pronounced at
13–12 and 8–7 Ma, coincident in time with higher magma productivity. However, the geochemical characteristics of the lavas are different during these two periods: the 13–12 Ma lavas have more radiogenic 176Hf/ 177Hf and less radiogenic 206Pb/ 204Pb than those erupted from 8 to 7 Ma. The eruption of relatively depleted lavas, at around 10 Ma and younger than 6·5 Ma, is coincident with lower magma productivity. The correlation between the composition and productivity of the Tertiary lavas from eastern Iceland is probably due to periodic changes in the involvement of the enriched end-member component, followed by a gradation to periods dominated by the signature of the depleted end-member component and lower magma productivity, at an approximate frequency of 5 Myr. KEY WORDS: mantle plume; magma productivity; mantle source; temporal variation; trace element and isotope geochemistry
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTING AND SAMPLESANALYTICAL METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Hotspot magmatism is thought to be the surface manifestation of partial melting in a mantle plume (Morgan, 1971
). However, we still know little about how mantle plumes evolve, and how their evolution contributes to secular changes in long-lived hotspot magmatism. To understand the evolution of long-lived magmatism caused by mantle plumes it is necessary to determine how the compositions and volume of lavas, the products of such magmatism, vary on time scales of millions to tens of millions of years. The North Atlantic province is an excellent target for such studies because the volcanism related to the Iceland mantle plume began at c. 60 Ma and continues to the present day, and has produced voluminous igneous rocks, predominantly basaltic lavas, widely distributed in this region including subaerial Iceland (e.g. Saunders et al., 1997). Iceland is located at the junction between the Kolbeinsey Ridge to the north and the Reykjanes Ridge to the south, and its high magma production rate has been attributed to the interaction between the Iceland mantle plume and the Mid-Atlantic Ridge (MAR), starting at 27 Ma (e.g. Óskarsson et al., 1985).
Numerous studies of the petrology and geochemistry of Icelandic basalts have been undertaken during the past three decades in an attempt to solve outstanding problems. One of the central subjects of debate is the origin of geochemical diversity in the postglacial lavas, which is probably due to changes in contributions from several different mantle end-member components (e.g. Zindler et al., 1979; Hemond et al., 1993; Chauvel & Hémond, 2000; Skovgaard et al., 2000; Stracke et al., 2003b). In particular, recent geochemical studies based on comprehensive and precise isotope measurements have attempted to resolve the spatial distribution of different mantle components within the Iceland mantle plume and local mixing trends between them during processes of melt generation and transport (e.g. Thirlwall et al., 2004; Kokfelt et al., 2006).
Although the geochemical variability of the postglacial lavas implies lateral heterogeneity within the mantle, studies of the older Tertiary lavas show that Icelandic magmatism has also varied compositionally during the last 16 Myr, revealing temporal variations in magma source composition (e.g. O’Nions & Pankhurst, 1973; Schilling et al., 1982; Hanan & Schilling, 1997). Hanan & Schilling (1997) first suggested, on the basis of Pb isotope data, that temporal geochemical variations in the Tertiary lavas could be attributed to secular changes in the contributions of three distinct end-member components to the magmatism. They also found a correlation between the composition and productivity of the Tertiary lavas, leading them to a ‘blob’-like plume model for the evolution of this magmatism. Additionally, geophysical and palaeoceanographic observations also indicate that there were temporal fluctuations in magma production rate in and around Iceland during the Neogene (e.g. Vogt, 1971; Wright & Miller, 1996). In particular, recent numerical modeling and palaeoceanographic studies have provided detail regarding temporal fluctuations in magma production rate, allowing us to compare precisely the composition of lavas with magma productivity (Jones et al., 2002; Poore et al., 2006).
In this study, we elaborate on the temporal geochemical variation in the Tertiary Icelandic magmatism based on a comprehensive analytical dataset including major and trace element concentrations and Sr–Nd–Hf–Pb isotope ratios of lavas collected along a palaeomagnetic traverse in eastern Iceland. We also provide important constraints on the geochemical characteristics of the end-member source components and evaluate the role of these components in magma productivity. These evaluations in turn allow us to speculate on models for the evolution of Icelandic magmatism.
|
GEOLOGICAL SETTING AND SAMPLES |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTING AND SAMPLESANALYTICAL METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
The surface exposure of the Icelandic crust is dominantly composed of basaltic lavas (c. 90 vol. %) with subordinate amounts of felsic and intermediate rocks (c. 10 vol. %) (e.g. Pálmason & Sæmundsson, 1974
). The youngest rocks are exposed in the Neovolcanic Zone, which is enclosed by the Pleistocene and Tertiary formations (e.g. Pálmason & Sæmundsson, 1974). The Tertiary formations are well exposed because they are deeply dissected by glacial erosion. According to previous geological investigations (Walker, 1959, 1964), the dissected fields formed by the Tertiary volcanic rocks can be subdivided into three main geological features: (1) volcanic centers; (2) lava piles (flood basalt successions); (3) swarms of dikes that intrude or constitute feeders for the above. The lava piles are volumetrically dominant in the Tertiary formations. The volcanic centers (e.g. Thingmuli; Carmichael, 1964), characteristically have suites of igneous rocks ranging in composition from basaltic to intermediate (andesitic) and felsic (dacitic and rhyolitic).
In eastern Iceland, Tertiary lavas ranging in age from 13 to 2 Ma are exposed on the plateau. The stratigraphy of this region is well established, and the succession is estimated to be c. 10 km in thickness and composed of c. 1000 individual flows (Dagley et al., 1967; Watkins & Walker, 1977). The lava piles generally have westerly dips of 5–10° towards the current rifting axes. Consequently, the oldest sequence is exposed on the east coast (Gerpir) and the younger sequence occurs in the west (south of Nordurdalur). The 114 lavas investigated in this study were sampled from successions along a paleomagnetic traverse (labeled A–V in Fig. 1) (Dagley et al., 1967). We have divided the locations of samples into eight separate areas (I–VIII), which correspond to each of the paleomagnetic sections. The locations of samples along the coast of Lagarfljót lake are grouped into an additional area, described here as the Lagarfljót area. This area would correspond sequentially to area VI (or Dagley's N–O–P profiles) (c. 6·5 Ma) (McDougall et al., 1976a). The ages assigned to the samples are derived from the magnetostratigraphy and published K–Ar and 40Ar/39Ar ages obtained for the lavas of this region (Fig. 1).
|
In general, the lava samples are fine-grained and virtually aphyric or sparsely phyric (generally <10 vol. %) with assemblages of phenocrysts consisting of plagioclase and subordinate amounts of olivine and clinopyroxene. Some of the lavas (I1415 and I1602) exhibit porphyritic textures with plagioclase being the main phenocryst phase (26–35 vol. %). The groundmasses generally have intergranular textures, and consist of plagioclase, clinopyroxene, an opaque mineral and glass.
|
ANALYTICAL METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTING AND SAMPLESANALYTICAL METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
All analyses were performed at the Pheasant Memorial Laboratory, Institute for Study of the Earth's Interior, Okayama University (Nakamura et al., 2003
). Samples were crushed with a jaw crusher to coarse chips of 3–5 mm in diameter, and then chips without weathered crusts were carefully selected. These unweathered chips were rinsed several times with deionized water in an ultrasonic bath until the water was clear after rinsing. Chips were then dried for 10 h in an oven at 100°C. The dried chips were pulverized into powders using an alumina puck mill.
Concentrations of major elements, Ni and Cr were determined by X-ray fluorescence spectrometry (XRF) with a Philips PW2400 instrument, using lithium tetraborate glass beads (1 : 10 ratios of sample and flux) (Takei, 2002). Water content (H2O+) was obtained by gravimetric methods, and FeO content was determined by titration (Yokoyama & Nakamura, 2002). Trace element compositions were determined by inductively coupled plasma mass spectrometry (ICP-MS) using two systems (Yokogawa PMS2000 and Agilent 7500cs), following the method described by Makishima & Nakamura (1997, 2006), Makishima et al. (1997, 1999), Yokoyama et al. (1999), Moriguti et al. (2004) and Lu et al. (2007a). Data are listed in Table 1. Whole-rock analyses, except for the measurements of water content, were duplicated using two aliquots of powdered samples. Analytical errors (1
) for analyses are within 1% for major elements and 3–5% for trace elements, respectively (Table 1).
|
Sr–Nd–Pb isotopic compositions were analyzed by thermal ionization mass spectrometry (TIMS; Finnigan MAT 261 and 262) in static multi-collection mode, following the methods of Yoshikawa & Nakamura (1993
) for Sr, and Nakamura et al. (2003) for Nd, and the normal double-spike (DS) method for Pb described by Kuritani & Nakamura (2003). Hf isotopes were measured on unspiked samples by multiple-collector (MC)-ICP-MS (Finnigan Neptune) using the method described by Lu et al. (2007b). Sr, Nd and Hf isotope ratios were normalized to 86Sr/88Sr = 0·1194, 146Nd/144Nd = 0·7219, and 179Hf/177Hf = 0·7325, respectively, to correct for isotopic fractionation during analysis. Most of the Sr–Nd–Hf isotope measurements were performed on unleached powders. For some samples, Sr–Nd–Hf isotope measurements were made for both acid-leached and unleached sample powders. Leaching was performed in hot 3M HCl (110°C) for 1 h, and the residual powders were multiply rinsed with water prior to dissolution. In most case, differences between leached and unleached results are within the analytical uncertainties, indicating that there is little or no significant surface contamination (Table 2). All Pb isotope analyses were performed on leached powders. Leaching for Pb isotope analysis was performed in hot 1M HCl (110°C) for 20 min, and then the residues were rinsed several times with water prior to acid digestion. After leaching, 30–80% of the Pb was leached out of the samples. Total procedural blanks (including contamination during pulverizing) for Sr, Nd, Hf and Pb were commonly less than 40, 15, 10, and 30 pg, respectively, and are considered to be negligible with respect to the abundances of the elements in the dissolved samples (more than 30000, 200, 100, and 150 ng for Sr, Nd, Hf, and Pb, respectively). 87Sr/86Sr ratios of NBS SRM 987 yielded an average of 0·710190 ± 30 (2, n = 106), and the average of 143Nd/144Nd ratios of La Jolla is 0·511863 ± 20 (2, n = 109), respectively. NBS SRM 987 and La Jolla data were normalized to 87Sr/86Sr = 0·710240 and 143Nd/144Nd = 0·511860, respectively, and these normalization factors are applied to the sample data to facilitate comparison between the datasets. The 176Hf/177Hf ratios of the JMC475 and JMC14375 Hf standards yielded averages of 0·282150 ± 6 (2, n = 9) and 0·282187 ± 8 (2, n = 24), respectively, during the course of analysis. To facilitate comparison, all Hf isotope ratios have been normalized to a JMC475 176Hf/177Hf = 0·282160. The NBS SRM981 Pb standard yielded an average (n = 14) of 206Pb/204Pb = 16·9422 ± 14 (2), 207Pb/204Pb = 15·5000 ± 13 (2) and 208Pb/204Pb = 36·7262 ± 44 (2), respectively, during the same analytical campaign; these values agree well with those obtained by Kuritani & Nakamura (2003) (206Pb/204Pb = 16·9424, 207Pb/204Pb = 15·5003, 208Pb/204Pb = 36·7266) and are comparable to those from other studies (e.g. Baker et al., 2004). Typical analytical reproducibilities (2) for the samples are about 40 ppm for Sr and Nd, and 30 ppm for Hf isotope analyses, and those for 206Pb/204Pb, 207Pb/204Pb and 208Pb/204Pb are less than 150 ppm. All isotope ratios presented in Table 2 are not age-corrected. Age correction does not significantly change the Sr–Nd–Hf isotope data, and thus we have used the measured Sr–Nd–Hf isotope data throughout. The maximum age correction for Pb isotope ratios is –0·054 (2900 ppm) for 206Pb/204Pb, –0·003 (160 ppm) for 207Pb/204Pb and –0·057 (1500 ppm) for 208Pb/204Pb, respectively, calculated using Pb–Th–U concentrations and the inferred age of the samples. These correction factors are clearly larger than the analytical errors, except for 207Pb/204Pb. Thus, we have used the age-corrected Pb isotope data throughout.
|
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTING AND SAMPLESANALYTICAL METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Major and trace element compositions
Using the alkali–silica classification (Irvine & Baragar, 1971
) the Tertiary lavas from eastern Iceland are mostly tholeiitic basalts and basaltic andesites. They are also classified as tholeiitic series lavas (no normative nepheline) based on their CIPW normative composition (Yoder & Tilley, 1962), except for two samples, IB1609 and IB1610, which have very minor (<1· 4%) amounts of normative nepheline (Fe3+/Fetotal molar ratio = 0·12 is used in the calculation). The major element compositions of these samples fall within the ranges reported for Tertiary lavas from Iceland (Wood, 1976, 1978; Hardarson et al., 1997). Figure 2 shows variation diagrams for selected major and minor elements plotted against MgO content. The MgO contents range from 2·6 to 8·6 wt %, and show broad positive correlations with Al2O3, CaO, Cr and Ni, and negative correlations with SiO2 and Na2O. The compositional trends of the lavas from the Thingmuli volcanic center, which are believed to lie on a single liquid line of descent related by crystal fractionation (Carmichael, 1964), are also shown for comparison. The Thingmuli trends generally track the data of this study, suggesting that the eastern Iceland Tertiary lavas are related, to a first order, by fractional crystallization. Samples with MgO of >7 wt % show distinct trends: the older lavas (13–11·5 and 9–7 Ma periods) have lower SiO2, CaO and Cr, and higher Al2O3, Na2O and Ni contents at a given MgO content than those of the younger lavas (5–2 Ma). It should be noted that there is almost no sign of accumulation of phenocrysts in most lava samples based on petrographic observations.
|
Primitive mantle normalized trace element patterns of the samples are illustrated together with those of the postglacial lavas in Fig. 3. Variations in the Tertiary lavas are smaller than those of the postglacial lavas and the patterns are similar to those of the postglacial tholeiitic lavas. In most of the Tertiary lavas, Ba is slightly enriched compared with its neighbouring elements Rb and Th [2 x BaN/(RbN + ThN) = 1·33 ± 0·30 (1
) (excluding two lavas showing plagioclase accumulation); subscript N denotes primitive mantle normalization], a feature shared with the postglacial lavas. On the other hand, the Tertiary lavas (with MgO >7 wt %) have no apparent positive Sr anomaly, which is normally observed in the trace element patterns of Icelandic basalts [2 x SrN/(PrN + NdN) = 1· 05 ± 0·16 (1)]. The differentiated lavas (MgO <7 wt %) show negative Sr anomalies [2 x SrN/(PrN + NdN) = 0·72 ± 0·21 (1)], indicating fractionation of plagioclase. Temporal variations in trace element compositions are described in a subsequent section.
|
Sr–Nd–Hf–Pb isotope compositions
The 87Sr/86Sr and
Nd values of the lavas analyzed in this study range from 0·70313 to 0·70349 and from 6·7 to 8·8, respectively; the data define a broad negative correlation in Sr–Nd isotope space (Fig. 4). Our data fall in the intermediate part of the Sr–Nd isotope array defined by the postglacial basalts. Marked differences in Sr–Nd isotope compositions within the Tertiary lavas from eastern Iceland are observed between lavas older than and younger than c. 6·5 Ma: lavas older than c. 6·5 Ma have higher 87Sr/86Sr and lower Nd than lavas younger than c. 6·5 Ma. The Hf values of the Tertiary lavas in this study range between 12·9 and 16·5, within the range of the postglacial lavas (Hf = 11·2–19·5). The Tertiary Icelandic lavas do not define a single linear trend on the Nd–Hf isotope correlation diagram (Fig. 5). The lavas older than 12 Ma have higher Hf values at a given Nd than the others, and most lavas of this older period plot above the oceanic basalt regression line of Vervoort et al. (1999) (mantle array in Fig. 5). Most of the lavas younger than 12 Ma plot on and around this line. The Tertiary lavas from eastern Iceland cover a range in Pb isotope composition (age corrected), from 18·10 to 18·67 for 206Pb/204Pb, 15·44 to 15·50 for 207Pb/204Pb and 37·85 to 38·35 for 208Pb/204Pb, all within the range for the postglacial lavas (206Pb/204Pb = 17·92–19·30, 207Pb/204Pb = 15·41–15·56, and 208Pb/204Pb = 37·54–38·93) (Fig. 6). The variations in Pb isotope ratios of the Tertiary lavas are consistent with the range previously reported for lavas from eastern Iceland (Hanan & Schilling, 1997).
|
|
|
Temporal variations in trace element and isotopic compositions
Trace element and Sr–Nd–Hf–Pb isotope ratios of the Tertiary lavas from eastern Iceland show systematic temporal fluctuations; the data from this study have been supplemented with additional data obtained by previous workers (Fig. 7). The (La/Sm)n, Ba/Nb and Sr–Nd–Hf–Pb isotope ratios of the Tertiary lavas show the following relationships: (1) (La/Sm)n shows a constant broad range from 13 to 6·5 Ma [(La/Sm)n = 1·3–2·0] with lower ratios (
1· 2) at around 10 Ma, followed by an abrupt decrease at 6·5 Ma (0·9–1· 5); (2) Ba/Nb shows a broad trend of decrease from 13 to 2 Ma with two small positive peaks around 9–8 and 6·5 Ma; (3) 87Sr/86Sr and Nd display abrupt changes at c. 6·5 Ma (lower 87Sr/86Sr and higher Nd at <6·5 Ma); (4) Hf broadly decreases from 13 to 7·5 Ma followed by an increase from 7·5 to 6·5 Ma, with no systematic change after 6·5 Ma; (5) 206Pb/204Pb and 208Pb/204Pb broadly increase from 13 to 8–7 Ma, and then decrease to 6·5 Ma; there are no systematic temporal changes in the lavas younger than 6·5 Ma; (6) 207Pb/204Pb also shows a broad increase from 13 to 7·5 Ma with lower ratios (15·46) around 10 Ma, and a decrease from 7· 5 to 6·5 Ma. There is also no systematic temporal variation in the lavas younger than 6·5 Ma, which show a wide range of variation from 15·44 to 15·49.
|
Regression analysis of Pb isotope trends
The Pb isotope data for the lavas from each area (I–VIII and Lagarfljót) (Fig. 1) are plotted separately on Pb isotope diagrams in Fig. 8. The aim here is to see the Pb isotope compositions sequentially and evaluate the secular changes in trends using two-dimensional isotope representations. The 13–11 Ma lavas (areas I–III) generally have lower 206Pb/204Pb values, and higher 207Pb/204Pb and 208Pb/204Pb at a given 206Pb/204Pb, and thus have higher
7/4 and 8/4 [ units indicate vertical deviations from the Northern Hemisphere Reference Line or NHRL (Hart, 1984)]. At 10 Ma (lavas from area IV), the Pb isotope compositions change to 7/4 and 8/4 slightly lower than those of the older lavas. The 10–7· 5 Ma lavas (IV and V), therefore, form slightly steeper trends relative to the area I–III lavas. The 7·5–6·5 Ma lavas (VI and Lagarfljót) have similar slopes to those of the trends for the 13–11 Ma lavas. After 6·5 Ma, lavas from areas VII and VIII show steeper trends than those for lavas older than 6·5 Ma.
|
At first glance, there is a distinction in slopes between the trends formed by the older lavas and those younger than 6·5 Ma. To evaluate this inference more rigorously, we apply the F-test to the residual variances of the best-fit regression lines in the Pb isotope plots. This test demonstrates whether the Pb isotope trend can be grouped into several distinct trends or not. First, the Pb isotope population is divided into two subpopulations, older and younger than 6·5 Ma, respectively. In applying the F-test, we frame the null hypothesis that the two Pb isotope regression trends can be regarded as the same. If the residual variances of the individual regressions are lower than that of the regression by pooling all data, the two regression lines are different from each other. Table 3 shows the results of the F-test performed on the Tertiary Icelandic lavas analyzed in this study. Comparison of the combined residual sum of squares for individual regressions with a pooled regression results in the statistic F value of 33·7 and 10·9 for 206Pb/204Pb–207Pb/204Pb and 206Pb/204Pb–208Pb/204Pb relationships, respectively. These values are much greater than the critical F value at 1% (F = 4·80) and even 0·1% (F = 7·36) significance levels. The probabilities of F values of 33·7 and 10·9 are 4 x 10–10 and 5 x 10–3%, respectively. Therefore, the null hypothesis that the two subpopulations yield the same regression line can be rejected at much better than 99% confidence, providing strong confirmation that there are at least two distinct Pb isotope trends in the Tertiary lavas rather than a single trend.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTING AND SAMPLESANALYTICAL METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Source characteristics
Statistical examination of the mixing end-member components
To assess the number of end-member components in the source of the Tertiary Icelandic lavas, we used principal component analysis (PCA) on the Pb isotope data. These data have the useful property that the three isotope ratios (206Pb/204Pb, 207Pb/204Pb and 208Pb/204Pb) have an identical denominator isotope (i.e. 204Pb), and thus binary mixing between two end-members is expected to form linear arrays in Pb–Pb isotope space; deviations of the data from such an array imply a contribution from a third end-member component. The calculated eigenvectors in the PCA may thus provide useful information for identifying the number of end-member mixing components in the source of the basalts. In contrast, PCA results for multiple isotope systems often indicate the spurious influence of a third or fourth end-member component because of differences in the denominator elements, as pointed out by Blichert-Toft et al. (2005
) and Debaille et al. (2006). Figure 9 shows the PCA output for the Pb isotope data. The three eigenvectors (v1, v2, and v3) account for 86·86%, 12·05% and 1· 09%, respectively, of the variance in the dataset. Because the first two principal eigenvectors (v1 and v2) represent 98·91% of the total variance, the contribution of v3 is probably negligible. The result of this PCA justifies the use of a mixing model involving three end-member components. Pb isotope data for lavas from eastern Iceland reported by Hanan & Schilling (1997) were also plotted using the projection vectors of our dataset (Fig. 9). Some of their data deviate greatly from the first principal eigenvector towards the 207Pb/204Pb axis, presumably because of some analytical problem (see below).
|
We also applied PCA to the Pb isotope data for the postglacial Icelandic lavas and Reykjanes and Kolbeinsey Ridge basalts combined with those of the Tertiary lavas from this study (Fig. 10). First, we performed PCA on the Pb isotope data obtained by the double-spike (DS) method (Baker et al., 2004
; Thirlwall et al., 2004). Then data obtained by conventional TIMS and Tl-corrected MC-ICP-MS methods were also projected using the projection vectors of the DS-corrected Pb isotope data. Our dataset is also shown plotted onto the plane containing the two calculated eigenvectors. The contribution of the first principal eigenvector, v1, increases significantly (to 98·51%), because the variation in 206Pb/204Pb (17·9–19·3) relative to 207Pb/204Pb and 208Pb/204Pb of the postglacial lavas is larger than that in the Tertiary lavas (18·1–18·7) (Fig. 6). The second and third eigenvectors (v2 and v3) account for 1·13 and 0·36%, respectively, of the variance. The cumulative contribution of v1 and v2 yields over 99·5% of the data variance. Therefore, the PCA results provide a strong confirmation that the Pb isotope variations in the subaerial lavas from Iceland (Tertiary and postglacial) and adjacent mid-ocean ridges can be well approximated by a single plane in 206Pb/204Pb–207Pb/204Pb-208Pb/204Pb three-dimensional space, thus confirming mixing between three mantle end-member components. The projection of the data produced by the conventional and, to a lesser extent, the Tl-doping method show some scatter and some of the data deviate considerably from the v1– v2 plane towards the 207Pb/204Pb axis (Fig. 10b and c). This may be due to an inadequate mass fractionation correction to the Pb isotope data measured by the conventional and Tl-doping methods (Baker et al., 2004, 2005; Thirlwall et al., 2004).
|
The marked difference in the Pb isotope arrays can be seen in plots of v1 vs v2 (Figs 9a and 10a). The v1 eigenvector lies roughly along the radiogenic extension of the Pb isotope trend, and could be considered to largely reflect the involvement of an end-member with the most radiogenic Pb isotope composition. The v2 eigenvector defines the offset among the data arrays and varies mostly in the direction of 207Pb/204Pb (Figs 9c and 10c); v2 thus appears to be dominantly controlled by the involvement of an end-member with higher
7/4.
Geochemical characteristics of the end-member components
The three mantle end-member components required to explain the isotopic compositions of the Icelandic basalts are termed here E-1 (enriched-1), E-2 (enriched-2), and D (depleted), respectively; these are analogous to the Pb isotope end-members ‘e’, ‘p’ and ‘d’ proposed by Hanan & Schilling (1997).
The E-1 end-member component largely contributes to the 12–13 Ma lavas and has lower 206Pb/204Pb and higher 7/4 and 8/4 than the E-2 end-member component (Figs 6 and 11). Hf isotope compositions also discriminate between the E-1 and E-2 end-member components: E-1 has more radiogenic 176Hf/177Hf (Hf) than E-2 (Fig. 11). The Sr–Nd–Pb isotope signature of the E-1 end-member component is characteristic of an Enriched Mantle 1 (EM-1) type mantle source (see Zindler & Hart, 1986). Thirlwall et al. (2004) also reported postglacial Icelandic basalts characterized by higher 7/4 (3·5), 8/4 (62) and 87Sr/86Sr (0·7034), and lower Nd (7·9), which appear to have a greater contribution from the E-1 end-member component. This end-member component is also considered to contribute largely to the postglacial off-axis central volcano Öræfajökull, whose magmas are characterized by high 87Sr/86Sr (0·7037) and low Nd (7–6) (Prestvik et al., 2001; Thirlwall et al., 2004; Kokfelt et al., 2006). The >12 Ma lavas also have higher Ba/Nb, Pb/Nb and Hf for a given Nd (i.e. high Hf) (Figs 5 and 11). These features are commonly observed in hotspot lavas belonging to the typical EM-1 group; for example, Hawaii (Blichert-Toft et al., 1999) and Pitcairn (Eisele et al., 2002). Higher Hf is attributed to higher Lu/Hf for a given Sm/Nd in the source of the E-1 end-member component. A plausible explanation for the higher Hf is selective fractionation of specific Hf-rich minerals, such as zircon or rutile, in the original source material. Selective fractionation of appropriate minerals could take place during the sedimentary cycle between sand and clay fractions (Patchett et al., 1984; Plank & Langmuir, 1998) or be associated with crustal differentiation by anatexis during high-grade metamorphic reactions (Schmitz et al., 2004). Therefore, two possible candidates for the origin of the E-1 end-member component can be proposed: pelagic sediments and lower crustal mafic granulites. Both candidates are able to explain the Sr–Nd–Pb isotope characteristics of the E-1 end-member component based on the parent/daughter element ratios in these isotope systems observed in appropriate crustal lithologies (e.g. Rudnick & Goldstein, 1990; Rudnick & Fountain, 1995; Plank & Langmuir, 1998; Eisele et al., 2002). Higher Ba/Nb and Pb/Nb are also diagnostic of continental-derived materials in the source of the E-1 end-member, because both sediments and continental crustal materials show strong enrichments in Ba and Pb and depletion in Nb (e.g. Rudnick & Fountain, 1995; Plank & Langmuir, 1998; Eisele et al., 2002). Hanan et al. (2004) have recently proposed that lower continental crust is the most plausible candidate for the EM-1 or Dupal-type source component in Southeast Indian Ridge basalts with high Hf. Thus one candidate for the origin of this end-member component is consistent with that proposed by Hanan et al. (2004).
|
The E-2 end-member component is obvious in the most radiogenic Pb isotope and least radiogenic Hf isotope compositions of the 7–8 Ma lavas analyzed in this study. The 7–8 Ma lavas have 87Sr/86Sr and
Nd values comparable with those of lavas erupted before 12 Ma (Figs 7 and 11). More radiogenic Pb isotope compositions are observed in the postglacial Icelandic lavas, which are mostly alkali baslats erupted either in the southwestern rift (e.g. Hekla and Torfajökull) or in the off-axis regions (e.g. Snæfellsnes Peninsula and Vestmannnaeyjar) in Iceland (e.g. Stecher et al., 1999; Thirlwall et al., 2004; Kokfelt et al., 2006). The E-2 end-member component is thus most clearly identified in the postglacial alkaline lavas and must have 206Pb/204Pb >19·3, 207Pb/204Pb >15·56, and 208Pb/204Pb >38·9. This interpretation is supported by the PCA. The direction of the first principal eigenvector (v1) defined by the Tertiary lavas (Fig. 9) is almost identical to that defined by both Tertiary and postglacial Icelandic lavas (Fig. 10). The composition of the E-2 end-member should plot at the higher extension along the v1 vector (Fig. 10a). The radiogenic Pb isotope ratios indicate long-term higher U/Pb and Th/Pb in the E-2 end-member than in the other end-member components. Such higher ratios can be produced by fractionation as a result of extraction of Pb from the original source material. For example, preferential loss of Pb could occur during dehydration of subducted oceanic crust in subduction zones (e.g. Ayers, 1998). The lower Ba/Nb of the E-2 end-member can also be explained by recycling of dehydrated oceanic crust as a result of selective loss of Ba relative to Nb during dehydration (e.g. Stalder et al., 1998). On the other hand, higher 87Sr/86Sr (>0·7034) for the proposed E-2 end-member component may imply that such a process did not significantly change the Rb/Sr ratio (Fig. 11). The oceanic crust also has lower Lu/Hf and Sm/Nd than the residual mantle and dehydration processes seem to have little effect on these ratios (e.g. Stracke et al., 2003a). Thus, the oceanic crust can also be a possible candidate for the Nd–Hf isotope signature of the E-2 end-member (low 143Nd/144Nd–176Hf/177Hf). The isotopic characteristics of the E-2 end-member component, described above, are similar to those of mantle components previously called ‘FOZO’ or ‘C’ (Hanan & Graham, 1996; Hanan et al., 2004; Stracke et al., 2005), and are postulated to be due to the recycling of oceanic crust (Hanan & Graham, 1996; Stracke et al., 2005).
The D end-member component is most prominent in the <6·5 Ma magmatism relative to other periods. The <6·5 Ma lavas are depleted in incompatible elements, with lower (La/Sm)n and Ba/Nb than the older lavas (Figs 7 and 11). The 87Sr/86Sr and Nd of the <6·5 Ma lavas are the least and most radiogenic, respectively; that is, these lavas are geochemically the most depleted among the Tertiary lavas from eastern Iceland (Figs 4 and 11). The D end-member component is lower in 207Pb/204Pb than the other two end-member components (Fig. 6). This end-member component also contributes to the incompatible element depleted lavas in the Neovolcanic Zone and the Reykjanes and Kolbeinsey Ridges (Figs 4–6
). The Tertiary lavas <6·5 Myr old are similar to the Reykjanes Ridge basalts; both are characterized by a negative principal component score for the Pb isotope second eigenvector (v2) (Fig. 10), implying that the D end-member component is similar to that contributing to the ocean-ridge basalts; that is, depleted mid-ocean ridge basalt (MORB) mantle (DMM). Although this end-member component is likely to be peridotitic material depleted by ancient melt extraction, there are two hypotheses for the origin of this source: (1) ambient MORB source mantle (e.g. Schilling, 1973; Mertz et al., 1991; Hanan & Schilling, 1997; Mertz & Haase, 1997; Stracke et al., 2003b) or (2) ancient recycled oceanic lithosphere, which has been inferred to be an intrinsic component of the Iceland mantle plume (e.g. Fitton et al., 1997, 2003; Chauvel & Hémond, 2000; Skovgaard et al., 2000; Kokfelt et al., 2006). It should be noted that evaluation of the origin of the D end-member component is not the aim of this study. Our dataset does not provide new constraints on this end-member component, because there are no Tertiary Icelandic lavas with depleted geochemical and isotopic compositions similar to the postglacial primitive tholeiitic basalts. We, therefore, simply note that the D end-member is likely to be peridotitic material that was depleted in incompatible elements by ancient melt extraction.
Origin of temporal geochemical variations in the eastern Iceland Tertiary lavas
Estimation of relative contributions from the three end-member components
We have calculated the elemental Pb fraction from the three end-member components using two-dimensional isotope space representations (206Pb/204Pb vs 207Pb/204Pb and 206Pb/204Pb vs 208Pb/204Pb), following the method described by Schilling et al. (2003). The Pb isotope compositions of the three end-member components for the ternary mixing model are shown in Fig. 6 and in Table 4. Before calculation we checked the applicability of the Pb isotope compositions of the three end-members using PCA and confirmed that these mostly plot on the plane defined by the Pb isotope compositions of Icelandic lavas in the 3-D Pb isotope correlation space based on the fact that the first and second principal components account for 99% of the total variance. Figure 12 shows the temporal variations in elemental Pb fractions from each of the three end-member components.
|
E-1,
|
The contribution of elemental Pb from the E-1 end-member component gradually decreases from
50% at 13 Ma to 20% at around 8–7 Ma, and then increases to 40% immediately before 6·5 Ma, followed by a decrease to 20–30% at c. 6·5 Ma. The influence of the E-2 end-member component increases from 13 to 8–7 Ma (from 20 to 50%) and subsequently decreased to 20–40% at c. 6·5 Ma. The contributions of both the E-1 and E-2 end-member components show no systematic secular changes during the period younger than 6·5 Ma. The contribution of the D end-member is more clearly shown by the Pb elemental fraction calculated based on the 206Pb/204Pb–207Pb/204Pb correlation relative to those from 206Pb/204Pb–208Pb/204Pb. The contribution from the D end-member component increased to 50% at c. 10 Ma and to 50–60% during the period younger than 6·5 Ma. The temporal trends obtained in this study are essentially similar to those calculated by Hanan & Schilling (1997), confirming our estimations of mixing proportions.
Evaluation of ‘temporal’ variations in Icelandic magmatism
Although the temporal changes in the contributions from the three end-member components can account for geochemical variations in the Tertiary lavas analyzed as part of this study, it is important to realize that the data are confined to the lava piles of eastern Iceland and therefore inevitably give a biased view of the temporal variation in Icelandic magmatism as a whole. Additionally, the postglacial Icelandic lavas show greater variations in geochemical and isotopic compositions than the Tertiary lavas and there are systematic geochemical differences within each volcanic zone (i.e. northern, eastern and western volcanic zones) (e.g. Stracke et al., 2003b; Thirlwall et al., 2004; Kokfelt et al., 2006). These differences presumably reflect lateral heterogeneity in the mantle and may also be due to variations in plume–ridge configuration. The geochemical diversity in the Tertiary lavas can presumably be related to lateral geochemical variations in the mantle source that have existed consistently from the Tertiary to the present; that is, the bulk composition of the Icelandic mantle source has been in a steady state for a few tens of million years. For example, Barker et al. (2006) attributed the sequential Nd and Pb isotope variations in the East Greenland lava formations (56·1–55·0 Ma: Storey et al., 2007a, 2007b) to lateral heterogeneity in the proto-Iceland mantle plume tapped by partial melting processes. Below we discuss this issue to examine the ‘temporal’ variations in the compositions of the Tertiary Icelandic lavas.
In Iceland, Tertiary lavas occur on both the eastern and western sides of the current rift axis. Previous studies have revealed clear secular variations in the rare earth element (REE) and Sr–Pb isotope compositions of the Tertiary lavas from both sides (O’Nions & Pankhurst, 1973; Schilling et al., 1982; Hanan & Schilling, 1997; Hanan et al., 2000). The majority of samples analysed in these studies were collected from profiles sampled for paleomagnetic studies (e.g. Dagley et al., 1967; McDougall et al., 1977, 1984). Although breaks of over 50 km exist in some of these composite sections, the Sr and Pb isotope data for Tertiary lavas from western Iceland (O’Nions & Pankhurst, 1973; Hanan & Schilling, 1997) show temporal trends similar or parallel to those of the lavas from eastern Iceland (Fig. 7). These observations may indicate that regular temporal variations in the geochemical characteristics of the magmatism existed over the whole of Iceland, suggesting a systematic secular change in the source of all Icelandic magmatism, rather than variations in the tapping processes of a laterally heterogeneous source alone. The similarities between the temporal variations in (La/Sm)n of lavas from the Tertiary Iceland successions and from the Faeroe Islands (50–60 Ma) (Schilling & Noe-Nygaard, 1974) also seemingly demonstrate that there is a regular secular variation in the (proto-)Icelandic magmatism. In summary, the geochemical variations in the Tertiary lavas probably reflect a true secular variation in Icelandic magmatism. However, it is still difficult to estimate the effect of lateral heterogeneity in the mantle on secular variation in Icelandic magmatism. To advance our knowledge of such secular variation, detailed geochemical studies of the Tertiary lavas from the broad region must be carried out.
Secular changes in mantle composition or post-emplacement processes?
Processes other than secular changes in the mantle composition could lead to systematic variations in magma composition with time. For example, rift relocation and subsequent crustal accretion processes could disturb the preservation of old lavas, casting doubt on the interpretation of the compositional changes observed in the Tertiary lavas as a ‘temporal’ effect [see Hardarson & Fitton (1997) and Hardarson et al. (1997)]. Hardarson and coworkers attributed the much smaller range of variation in Nb/Zr of the Tertiary lavas relative to the postglacial lavas to biased sampling, resulting from the subsidence and burial of volumetrically smaller, more geochemically heterogeneous flows. The volumetrically dominant flows with homogeneous compositions erupted and flowed away from the rift axis, and these lavas remain in the upper portion of the crustal section exposed by glacial erosion. Wood et al. (1976) interpreted the REE variations in the Tertiary lavas as having been caused by secondary alteration. However, it seems unlikely that either model can be used exclusively to explain the geochemical variations of the Tertiary lavas for the following reasons: (1) Hf isotope compositions, which are thought to be unaffected by secondary alteration or weak metamorphism (e.g. Wood et al., 1976), show systematic temporal variations and correlations with other trace element and isotope ratios (Figs 7 and 11); (2) it seems difficult to explain the correlation between geochemical fluctuations and geological and geophysical observations (discussed in the following section), by subsidence and burial of lavas associated with crustal accretion processes alone (Fig. 13).
|
Rift relocation versus a pulsing mantle plume
A final issue concerns temporal changes in plume–ridge configuration as a cause of secular changes in the surface magmatism. Over the last 20 Myr, the rift axes have jumped eastwards in a series of steps so as to remain above the center of the Iceland mantle plume (e.g. Sæmundsson, 1974
; Hardarson et al., 1997; Garcia et al., 2003). Hardarson et al. (1997) suggested that temporal variations in magma productivity may be caused by rift relocation on the basis of an assumption that the mantle plume has been in a steady state in terms of composition and upwelling mass flux. They assumed that rift relocation plays a role in altering the pattern of mantle flow from the plume into the rift as a result of changes in distance between the center of the mantle plume and the spreading axis. As the spreading axis moves away from the mantle plume, magma productivity should decrease gradually. Hardarson et al. (1977) attributed the prominent V-shaped ridges in the sea floor south of Iceland to periodic relocations of the rift axis. However, recent numerical experiments have suggested that relocation of the spreading axis would be unlikely to disrupt the upwelling flow of the mantle plume (Ito, 2001; Jones et al., 2002). This is because the base of the rheological plate probably corresponds to the dry solidus (i.e. dehydration boundary), which probably lies at a constant depth of around 100 km and is independent of lithospheric age (Ito, 2001; Jones et al., 2002). Ito and Jones et al. proposed that the major cause of the V-shaped ridges is temporal changes in magma productivity caused by fluctuations in temperature (i.e. a pulsing mantle plume). In the following section, we evaluate the relationships between magma productivity and temporal changes in the composition of the magma source, to assess the relative contributions of the source composition to magma productivity.
Correlation between geochemical characteristics and productivity of the Tertiary Icelandic magmatism
Palaeomagnetic studies of the Tertiary lavas from both eastern and western Iceland indicate that there were several distinct periods of high magma productivity, as summarized in Fig. 13. The sea floor south of Iceland, as far away as a few hundred kilometers from Iceland, also records temporal changes in magma productivity during the Tertiary. For example, bathymetric features and gravity anomaly data show fluctuations that could be interpreted in terms of secular changes in magma productivity (e.g. Vogt, 1971; Johansen et al., 1984; Wright & Miller, 1996; Smallwood & White, 1998; Jones et al., 2002). A short-wavelength gravity anomaly profile obtained from the Irminger Basin on the western flank of the Reykjanes Ridge, located over 300 km south of Iceland (Jones et al., 2002), is shown in Fig. 13d, with an applied time shift of +0·5 Ma. This shift results in the gravity peaks corresponding to those in the eruption rate of Icelandic lavas and to peaks in a topographic transect across the Reykjanes Ridge and Faeroe–Iceland–Greenland aseismic ridge (Wright & Miller, 1996). The positive gravity anomalies correlate with the positive depth anomalies, indicating thicker crust produced by increases in the rate of magmatism. Jones et al. (2002) determined a propagation speed of anomalies within the Iceland plume of 200–250 km/Myr using the angle between the V-shaped ridges and the spreading axis, which corresponds to a time shift of +1·2–2·0 Myr. In contrast, Wright & Miller (1996) calculated propagation speeds, based on the lineation of escarpments, ranging from >1000 km/Myr for the youngest escarpment to 160 km/Myr for the oldest escarpment. Using the propagation speed estimated by Wright & Miller (1996), the possible range in acceptable time shifts is calculated to be from +0·3 to 2·5 Myr. A shift of +0·5 Myr falls within this range and so is an acceptable adjustment to align the gravity profile to the record of subaerial Icelandic magmatism. Because the validity of using of an elastic correction is not yet clear (Poore et al., 2006), more detailed numerical modeling is beyond the scope of this study. We emphasize, however, that, as shown in Fig. 13, there is a good correlation between the corrected gravity profile and the subaerial lava accumulation rate in Iceland.
To evaluate the relationship between magma productivity and temporal changes in the composition of the magma source, the elemental Pb fractions from the E-1 and E-2 end-member components relative to that of the D end-member component are compared with magma productivity (Fig. 13). During the period around 13 Ma, lavas with larger contributions from the E-1 end-member component erupted in larger volumes and produced relatively thicker sections of lavas. Subsequently, the magma productivity decreased toward 10 Ma as the contribution from the E-1 end-member component declined and the contribution from the E-2 end-member correspondingly increased. At around 10 Ma, the contribution from the D end-member increased (Fig. 12), accompanied by a decrease in magma productivity shown by the negative gravity anomaly. During the subsequent period from 10 to 8–7 Ma, the contribution from the E-2 end-member component increased dramatically. The increase in the E-2 contribution was associated with a significant increase in magma productivity towards 8–7 Ma, demonstrating a link between the source composition and magma productivity. After this period, the contribution from the E-2 end-member component decreased gradually, and the relative proportion of the E-1 and D end-member components increased again, associated with a gentle decrease in magma productivity toward 6·5 Ma. Magma productivity decreased after 6·5–6 Ma, correlated with the eruption of lavas having larger contributions from the D end-member component relative to both the E-1 and E-2 end-member components. The decrease in magma productivity from 6 to 4 Ma is indicated by both the accumulation rate of the subaerial lavas and the gravity anomaly data (Fig. 13). The increasing intensity of the gravity anomaly for crust younger than 4 Ma, however, is not correlated with the subaerial lava accumulation rate in Iceland. The reason for this discrepancy still remains unconstrained and the relationships between the gravity anomaly data and the estimation of eruption rate for younger lavas (<2 Ma) need to be investigated.
The most likely explanation for the correlation between source composition and productivity for the Tertiary lavas is that the geochemically enriched end-member components (i.e. E-1 and E-2 end-members) are more easily melted than the relatively depleted D end-member component. Thus, melting of source materials rich in the E-1 or E-2 end-members should result in higher melt productivity at a given temperature in the melting region than melting of source material dominated by the D end-member component (e.g. Hirshmann & Stolper, 1996; Yaxley, 2000). This interpretation is consistent with the E-1 and E-2 end-member components having been derived from recycled crustal materials (with low melting temperatures) as deduced from their geochemical and isotopic characteristics. The major and minor element compositions of the lavas also seem to be consistent with this hypothesis: the older lavas (13–11· 5 and 9–7 Ma) have lower SiO2 and CaO, and higher Al2O3 and Na2O contents than the younger lavas (5–2 Ma) (Fig. 2). Hirshmann et al. (2003) showed that low SiO2–CaO and high Al2O3–Na2O melts can be generated experimentally by melting of silica-deficient pyroxenite, a possible candidate for recycled material in the melting region, at moderately high pressure conditions (2 GPa), consistent with the setting of Iceland as a juxtaposed hotspot and mid-ocean ridge system. Higher Ni at a given MgO content in the 13–11· 5 and 9–7 Ma lavas may also indicate the involvement of recycled crustal lithologies in the source, because these materials (e.g. pyroxenite) are expected to have lower abundances of olivine, which might buffer the Ni content in the melts relative to mantle peridotite (Sobolev et al., 2005).
Implications for the evolution of Icelandic magmatism
Based on the above discussion, temporal variations in composition and the productivity of the Tertiary Icelandic magmatism can be explained by periodic changes in source composition. This may imply periodic transport of recycled materials from deep mantle regions to the shallow melting regime of the upper mantle. Temporal fluctuations in gravity anomaly data for the Irminger Basin appear to be cyclical on time scales of 5 Myr and can be traced to 35 Ma (Jones et al., 2002). Indirect effects of such fluctuations in the magmatism can also be detected in other ways. Magmatic underplating related to the activity of the mantle plume rapidly drove regional surface uplift and denudation, resulting in the periodic development of fan deposits along the continental margins surrounding the North Atlantic during the Palaeogene (White & Lovell, 1997). Temporal variations in the flux of Northern Component Water (NCW) were affected by the dynamic support that resulted from secular changes in the mantle plume activity during the Neogene: times of high mantle plume activity caused NCW production to cease (Wright & Miller, 1996; Poore et al., 2006). Additionally, seamount basalts from the Atlantic coast of Scotland also record periodic magmatism on time scales of 5–10 Myr (O’Connor et al., 2000). These studies seemingly demonstrate multiple episodes of Icelandic plume magmatism from c. 60 Ma to the present. The similarities between the Tertiary magmatism in the Faeroe Islands (60–50 Ma) and Iceland, in terms of secular variations in production rates and the (La/Sm)n ratios of the lavas, also support a model of volumetrically and geochemically periodic magmatism (Noe-Nygaard & Rasmussen, 1968; Schilling & Noe-Nygaard, 1974; Schilling et al., 1982). The blob model [i.e. a pulsing mantle plume, originally proposed by Schilling & Noe-Nygaard (1974)] seems to be an attractive model to account for these observations and for the geochemical data obtained in this study. To explain the observed secular variations in Icelandic magmatism, here we provide a conceptual model for the evolution of the Iceland mantle plume (Fig. 14), based on temporal variations in the geochemistry and productivity of the Tertiary lavas (Figs 12 and 13).
|
At >13 Ma, a blob dominated by the E-1 end-member component migrated upwards and encountered the melting region, resulting in enhanced magma productivity. A smaller amount of material with the E-2 end-member signature surrounded the tail of this blob. Therefore, from 13 Ma to >10 Ma, the magmatism changed temporally from being E-1 influenced to being E-2 influenced (Figs 12 and 13). Later, at around 10 Ma, this blob had been partly consumed by melting, and the residue became incorporated into the lithosphere. Therefore, the proportion of the D end-member-derived flux increased at c. 10 Ma, and magma productivity correspondingly decreased. During the subsequent period, the next blob ascended into the melting region, resulting in enhanced magma productivity to 8–7 Ma. This blob was mainly composed of the E-2 end-member component, possibly surrounded by the E-1 end-member component. The final stage of melting of this blob at c. 6·5 Ma would thus account for the slightly higher contribution from the E-1 end-member component. Later, this blob also became exhausted, and the contribution from the D end-member component again increased, associated with a decline in magma productivity.
Alternating pulses of blobs relatively rich in the E-1 and E-2 end-member components may have contributed to the temporal fluctuations in productivity and composition of the Tertiary Icelandic magmatism. An important aspect of our study is that it demonstrates the possibility that discrete mantle blobs containing recycled crustal materials played a major role in controlling the periodic increases in the rate of Icelandic magmatism. Recent numerical modeling (Lin & van Keken, 2005, 2006a, 2006b) demonstrates that the entrainment of dense recycled materials in the source region of a mantle plume can lead to multiple pulses of plume activity, a conclusion seemingly consistent with our model based largely on geochemical data.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTING AND SAMPLESANALYTICAL METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
The data presented in this study provide an 11 Myr record of temporal geochemical variations in Icelandic magmatism from 13 to 2 Ma, and yield constraints on the characteristics of the mantle end-member components involved in this magmatism. Temporal variations in the lava geochemistry can be explained by changes in the relative contributions from three mantle end-member components (E-1, E-2 and D), each with distinct geochemical characteristics. Larger contributions from the E-1 end-member component dominate the 13–12 Ma lavas, producing
7/4, 8/4 and Hf values higher than in lavas from the other periods. The characteristics of another enriched end-member component, E-2, are observed in the 8–7 Ma lavas, and these lavas have the most radiogenic Pb isotope compositions among the eastern Iceland Tertiary lavas. The origin of these two end-member components can be related to recycling of crustal materials. The lavas erupted at c. 10 and 6·5–2 Ma are geochemically less enriched, relative to those from the other periods, and are attributed to melts with a greater contribution from the D end-member component, interpreted to be peridotitic material. The temporal geochemical variations in the Tertiary lavas are well correlated with magma productivity: higher magma productivity is associated with the eruption of geochemically more enriched lavas, whereas lower magma productivity is coincident with the emplacement of less enriched lavas. Correspondence between productivity and the compositions of the Tertiary Icelandic lavas could be due to the periodic entrainment of recycled crustal lithologies into the Iceland plume at its source. |
ACKNOWLEDGEMENTS |
|---|
We are grateful to Y.-H. Lu and all members of the Pheasant Memorial Laboratory for their technical support, constructive discussion and encouragement. We would like to thank T. Yokoyama, K. Kuritani and E. Rose for discussion and help in the sampling, and J. Ö. Fridsteinsson and Á. Höskuldsson for their guidance during the field work in Iceland. We thank I. H. Campbell, K. Grönvold, G. E. Bebout, S.M. Jones, J. Maclennan, B. N. Nath, B. Paul, R. Tanaka and A. Ishikawa for discussion and improving the manuscript. Thomas Kokfelt, David Peate and one anonymous reviewer are thanked for their helpful reviews, and Colin Devey for his editorial handling. This study was supported by the program ‘Centers of Excellence for 21st century in Japan’ (E.N.), and grants-in-aid for scientific research from MEXT to E.N., A.M., and K.K.
*Corresponding author. Telephone: +81-858-43-3745. Fax: +81-858-43-3745. E-mail: eizonak{at}misasa.okayama-u.ac.jp
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONGEOLOGICAL SETTING AND SAMPLESANALYTICAL METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Anders E, Grevesse N. Abundances of the elements: Meteoritic and solar. Geochimica et Cosmochimica Acta (1989) 53:197–214.[CrossRef][Web of Science]
Ayers J. Trace element modeling of aqueous fluid–peridotite interaction in the mantle wedge of subduction zones. Contributions to Mineralogy and Petrology (1998) 132:390–404.[CrossRef][Web of Science]
Baker J, Peate D, Waight T, Meyzen C. Pb isotopic analysis of standards and samples using a 207Pb–204Pb double spike and thallium to correct for mass bias with a double-focusing MC-ICP-MS. Chemical Geology (2004) 211:275–303.[CrossRef][Web of Science]
Baker J, Peate DW, Waight TE, Thirlwall MF. Reply to the: Comment on ‘Pb isotope analysis of standards and samples using a 207Pb–204Pb double spike and thallium to correct for mass bias with a double focusing MC-ICP-MS’ by Baker et al. Chemical Geology (2005) 217:175–179.[CrossRef][Web of Science]
Barker A, Baker JA, Peate DW. Interaction of the rifting East Greenland margin with a zoned ancestral Iceland plume. Geology (2006) 34:481–484. doi:10.1130/G22366.1.
Blichert-Toft J, Frey FA, Albarède F. Hf isotope evidence for pelagic sediments in the source of Hawaiian basalts. Science (1999) 285:879–882.
Blichert-Toft J, Agranier A, Andres M, Kingsley R, Schilling J-G, Albarède F. Geochemical segmentation of the Mid-Atlantic Ridge north of Iceland and ridge–hot spot interaction in the North Atlantic. Geochemistry, Geophysics, Geosystems (2005) 6:Q01E19. doi:10.1029/2004GC000788.[CrossRef]
Breddam K. Kistfell: Primitive melt from the Iceland mantle plume. Journal of Petrology (2002) 43:345–373.
Carmichael ISE. The petrology of Thingmuli, a Tertiary volcano in eastern Iceland. Journal of Petrology (1964) 5:435–460.
Chauvel C, Hémond C. Melting of a complete section of recycled oceanic crust: Trace element and Pb isotopic evidence from Iceland. Geochemistry, Geophysics, Geosystems (2000) 1. 1999GC000002.
Dagley P, Wilson RL, Ade-Hall JM, Walker GPL, Haggerty SE, Sigurgeirsson T, Watkins ND, Smith PJ, Edwards J, Grasty RL. Geomagnetic polarity zones for Icelandic lavas. Nature (1967) 216:25–29.[CrossRef]
Debaille V, Blichert-Toft J, Agranier A, Doucelance R, Schiano P, Albarède F. Geochemical component relationships in MORB from the Mid-Atlantic Ridge, 22–35°N. Earth and Planetary Science Letters (2006) 241:844–862.[CrossRef][Web of Science]
Eisele J, Sharma M, Galer SJG, Blichert-Toft J, Devey CW, Hofmann AW. The role of sediment recycling in EM-1 inferred from Os, Pb, Hf, Nd, Sr isotope and trace element systematics of the Pitcairn hotspot. Earth and Planetary Science Letters (2002) 196:197–212.[CrossRef][Web of Science]
Elliott TR, Hawkesworth CJ, Grönvold K. Dynamic melting of the Iceland plume. Nature (1991) 351:201–206.[CrossRef]
Fitton JG, Saunders AD, Norry MJ, Hardarson BS, Taylor RN. Thermal and chemical structure of the Iceland plume. Earth and Planetary Science Letters (1997) 153:197–208.[CrossRef][Web of Science]
Fitton JG, Saunders AD, Kempton PD, Hardarson BS. Does depleted mantle form an intrinsic part of the Iceland plume? Geochemistry, Geophysics, Geosystems (2003) 4. doi:10.1029/2002GC000424.
Garcia S, Arnaud NO, Angelier J, Bergerat F, Homberg C. Rift jump process in Northern Iceland since 10 Ma from 40Ar/39Ar geochronology. Earth and Planetary Science Letters (2003) 214:529–544.[CrossRef][Web of Science]
Hanan BB, Graham DW. Lead and helium isotope evidence from oceanic basalts for a common deep source of mantle plumes. Science (1996) 272:991–995.[Abstract]
Hanan BB, Schilling J-G. The dynamic evolution of the Iceland mantle plume: the lead isotope perspective. Earth and Planetary Science Letters (1997) 151:43–60.[CrossRef][Web of Science]
Hanan BB, Blichert-Toft J, Kingsley R, Schilling J-G. Depleted Iceland mantle plume geochemical signature: Artifact of multicomponent mixing? Geochemistry, Geophysics, Geosystems (2000) 1. 1999GC000009.
Hanan BB, Blichert-Toft J, Pyle DG, Christie DM. Contrasting origins of the upper mantle revealed by hafnium and lead isotopes from the Southeast Indian Ridge. Nature (2004) 432:91–94.[CrossRef]
Hardarson BS, Fitton JG. Mechanisms of crustal accretion in Iceland. Geology (1997) 25:1043–1046.
Hardarson BS, Fitton JG, Ellam RM, Pringle MS. Rift relocation—a geochemical and geochronological investigation of a palaeo-rift in northwest Iceland. Earth and Planetary Science Letters (1997) 153:181–196.[CrossRef][Web of Science]
Hards VL, Kempton PD, Thompson RN. The heterogeneous Iceland plume: new insights from the alkaline basalts of the Snaefell volcanic centre. Journal of the Geological Society, London (1995) 152:1003–1009.
Hards VL, Kempton PD, Thompson RN, Greenwood PB. The magmatic evolution of the Snæfell volcanic centre; an example of volcanism during incipient rifting in Iceland. Journal of Volcanology and Geothermal Research (2000) 99:97–121.[CrossRef][Web of Science]
Hart SR. A large-scale isotope anomaly in the Southern Hemisphere mantle. Nature (1984) 309:753–757.[CrossRef]
Hemond C, Arndt NT, Lichtenstein U, Hofmann AW, Oskarsson N, Steinthorsson S. The heterogeneous Iceland plume: Nd–Sr–O isotopes and trace element constraints. Journal of Geophysical Research (1993) 98:15833–15850.[CrossRef]
Hirshmann MM, Stolper EM. A possible role for garnet pyroxenite in the origin of the ‘garnet signature’ in MORB. Contributions to Mineralogy and Petrology (1996) 124:185–208.[CrossRef][Web of Science]
Hirshmann MM, Kogiso T, Baker MB, Stolper EM. Alkalic magmas generated by partial melting of garnet pyroxenite. Geology (2003) 31:481–483.
Irvine TN, Baragar WRA. A guide to the chemical classification of the common volcanic rocks. Canadian Journal of Earth Sciences (1971) 8:523–548.
Ito G. Reykjanes ‘V’-shaped ridges originating from a pulsing and dehydrating mantle plume. Nature (2001) 411:681–684.[CrossRef]
Jancin M, Young KD, Voight B, Aronson JL, Sæmundsson K. Stratigraphy and K/Ar ages across the west flank of the northeast Iceland axial rift zone, in relation to the 7 Ma volcano-tectonic reorganization of Iceland. Journal of Geophysical Research (1985) 90:9961–9985.
Johansen B, Vogt PR, Eldholm O. Reykjanes Ridge: further analysis of crustal subsidence and time transgressive basement topography. Earth and Planetary Science Letters (1984) 68:249–258.[CrossRef][Web of Science]
Jones SM, White N, Maclennan J. V-shaped ridges around Iceland: Implications for spatial and temporal patterns of mantle convection. Geochemistry, Geophysics, Geosystems (2002) 3. doi:10.1029/2002GC000361.
Kempton PD, Fitton JG, Saunders AD, Nowell GM, Taylor RN, Hardarson BS, Pearson G. The Iceland plume in space and time: a Sr–Nd–Pb–Hf study of the North Atlantic rifted margin. Earth and Planetary Science Letters (2000) 177:255–271.[CrossRef][Web of Science]
Kokfelt TF, Hoernle K, Hauff F, Fiebig J, Werner R, Garbe-Schönberg D. Combined trace element and Pb–Nd–Sr–O isotope evidence for recycled oceanic crust (upper and lower) in the Iceland mantle plume. Journal of Petrology (2006) 47:1705–1749.
Kuritani T, Nakamura E. Highly precise and accurate isotopic analysis of small amounts of Pb using 205Pb–204Pb and 207Pb–204Pb, two double spikes. Journal of Analytical and Atomic Spectrometry (2003) 18:1464–1470.[CrossRef]
Lin S-C, van Keken PE. Multiple volcanic episodes of flood basalts caused by thermochemical mantle plumes. Nature (2005) 436:250–252.[CrossRef][Medline]
Lin S-C, van Keken PE. Dynamics of thermochemical plumes: 1. Plume formation and entrainment of a dense layer. Geochemistry, Geophysics, Geosystems (2006a) 7:Q02006. doi:10.1029/2005GC001071.[CrossRef]
Lin S-C, van Keken PE. Dynamics of thermochemical plumes: 2. Complexity of plume structures and its implications for mapping mantle plumes. Geochemistry, Geophysics, Geosystems (2006b) 7:Q03003. doi:10.1029/2005GC001072.[CrossRef]
Lu Y, Makishima A, Nakamura E. Coprecipitation of Ti, Mo, Sn and Sb with fluorides and application to determination of B, Ti, Zr, Nb, Mo, Sn, Sb, Hf and Ta by ICP-MS. Chemical Geology (2007a) 236:13–26.[CrossRef][Web of Science]
Lu Y, Makishima A, Nakamura E. Purification of Hf in silicate materials using extraction chromatographic resin, and its application to precise determination of 176Hf/177Hf by MC-ICP-MS with 179Hf spike. Journal of Analytical and Atomic Spectrometry (2007b) 22:69–76. doi:10.1039/b610197f.[CrossRef]
Makishima A, Nakamura E. Suppression of matrix effects in ICP-MS by high power operation of ICP: Application to precise determination of Rb, Sr, Y, Cs, Ba, REE, Pb, Th and U at ng g–1 levels in milligram silicate samples. Geostandards Newsletter (1997) 21:307–319.[Web of Science]
Makishima A, Nakamura E. Determination of major, minor and trace elements in silicate samples by ICP-QMS and ICP-SFMS applying isotope dilution-internal standardisation (ID-IS) and multi-stage internal standardisation. Geostandards and Geoanalytical Research (2006) 30:245–271.[CrossRef][Web of Science]
Makishima A, Nakamura E, Nakano T. Determination of boron in silicate samples by direct aspiration of sample HF solutions into ICPMS. Analytical Chemistry (1997) 69:3754–3759.
Makishima A, Nakamura E, Nakano T. Determination of zirconium, niobium, hafnium and tantalum at ng g–1 levels in geological materials by direct nebulisation of sample HF solution into FI-ICP-MS. Geostandards Newsletter (1999) 23:7–20.[Web of Science]
McDonough WF, Sun S-s. The composition of the Earth. Chemical Geology (1995) 120:223–253.[CrossRef][Web of Science]
McDougall I, Watkins ND, Kristjansson L. Geochronology and paleomagnetism of a Miocene–Pliocene lava sequence at Bessastadaa, eastern Iceland. American Journal of Science (1976a) 276:1078–1095.
McDougall I, Watkins ND, Walker GPL, Kristjansson L. Potassium–argon and paleomagnetic analysis of Icelandic lava flows: Limits on the age of anomaly 5. Journal of Geophysical Research (1976b) 81:1505–1512.[CrossRef]
McDougall I, Sæmundsson K, Johannesson H, Watkins ND, Kristjansson L. Extension of the geomagnetic polarity time scale to 6·5 m.y.: K–Ar dating, geological and paleomagnetic study of a 3,500-m lava succession in western Iceland. Geological Society of America Bulletin (1977) 88:1–15.
McDougall I, Kristjansson L, Sæmundsson K. Magnetostratigraphy and geochronology of northwest Iceland. Journal of Geophysical Research (1984) 89:7029–7060.
Mertz DF, Haase KM. The radiogenic isotope composition of the high-latitude North Atlantic mantle. Geology (1997) 25:411–414.
Mertz DF, Devey CW, Todt W, Stoffers P, Hofmann AW. Sr–Nd–Pb isotope evidence against plume–asthenosphere mixing north of Iceland. Earth and Planetary Science Letters (1991) 107:243–255.[CrossRef][Web of Science]
Morgan WJ. Convection plumes in the lower mantle. Nature (1971) 230:42–43.[CrossRef]
Moriguti T, Makishima A, Nakamura E. Determination of lithium contents in silicates by isotope dilution ICP-MS and its evaluation by isotope dilution thermal ionisation mass spectrometry. Geostandards and Geoanalytical Research (2004) 28:371–382.[CrossRef][Web of Science]
Mussett AE, Ross JG, Gibson IL. 40Ar/39Ar dates of eastern Iceland lavas. Geophysical Journal of the Royal Astronomical Society (1980) 60:37–52.[Web of Science]
Nakamura E, Makishima A, Moriguti T, Kobayashi K, Sakaguchi C, Yokoyama T, Tanaka R, Kuritani T, Takei H. Comprehensive geochemical analyses of small amounts (<100 mg) of extraterrestrial samples for the analytical competition related to the sample return mission MUSES-C. Institute of Space and Astronautical Science Report SP (2003) 16:49–101.
Noe-Nygaard A, Rasmussen J. Petrology of a 3,000 metre sequence of basaltic lavas in the Faeroe Islands. Lithos (1968) 1:286–304.[CrossRef]
O’ Connor JM, Stoffers P, Wijbrans JR, Shannon PM, Morrissey T. Evidence from episodic seamount volcanism for pulsing of the Iceland mantle plume in the past 70 Myr. Nature (2000) 408:954–958.[CrossRef]
O’ Nions RK, Pankhurst RJ. Secular variation in the Sr-isotope composition of Icelandic volcanic rock. Earth and Planetary Science Letters (1973) 21:13–21.[Medline]
Óskarsson N, Steinthorsson S, Sigvaldason GE. Iceland geochemical anomaly: origin, volcanotectonics, chemical fractionation and isotope evolution of the crust. Journal of Geophysical Research (1985) 90:10011–10025.[CrossRef]
Pálmason G, Sæmundsson K. Iceland in relation to the Mid-Atlantic Ridge. Annual Review of Earth and Planetary Sciences (1974) 2:25–50.[CrossRef][Web of Science]
Patchett PJ, White WM, Feldmann H, Kielinczuk S, Hofmann AW. Hafnium/rare earth element fractionation in the sedimentary system and crustal recycling into the Earth's mantle. Earth and Planetary Science Letters (1984) 69:365–378.[CrossRef][Web of Science]
Plank T, Langmuir CH. The chemical composition of subducting sediment and its consequences for the crust and mantle. Chemical Geology (1998) 145:325–394.[CrossRef][Web of Science]
Poore HR, Samworth R, White NJ, Jones SM, McCave IN. Neogene overflow of Northern Component Water at the Greenland–Scotland Ridge. Geochemistry, Geophysics, Geosystems (2006) 7:Q06010. doi:10.1029/2005GC001085.[CrossRef]
Prestvik T, Goldberg S, Karlsson H, Grönvold K. Anomalous strontium and lead isotope signatures in the off-rift Öræfajökull central volcano in south-east Iceland. Evidence for enriched endmember(s) of the Iceland mantle plume? Earth and Planetary Science Letters (2001) 190:211–220.[CrossRef][Web of Science]
Ross JG, Mussett AE. 40Ar/39Ar dates for spreading rates in eastern Iceland. Nature (1976) 259:36–38.[CrossRef]
Rudnick RL, Fountain DM. Nature and composition of the continental crust: a lower crustal perspective. Reviews of Geophysics (1995) 33:267–309.[CrossRef][Web of Science]
Rudnick RL, Goldstein SL. The Pb isotopic compositions of lower crustal xenoliths and the evolution of lower crustal Pb. Earth and Planetary Science Letters (1990) 98:192–207.[CrossRef][Web of Science]
Saunders AD, Fitton JG, Kerr AC, Norry MJ, Kent RW. Large Igneous Provinces: Continental, Oceanic, and Planetary Flood Volcanism. American Geophysical Union, Geophysical Monograph—Mahoney JJ, Coffin MF, eds. (1997) 100:45–93.
Sæmundsson K. Evolution of the axial rifting zone in northern Iceland and the Tjörnes fracture zone. Geological Society of America Bulletin (1974) 85:495–504.
Schilling J-G. Iceland mantle plume: geochemical study of Reykjanes Ridge. Nature (1973) 242:565–571.[CrossRef]
Schilling J-G, Noe-Nygaard A. Faeroe–Iceland plume: rare-earth evidence. Earth and Planetary Science Letters (1974) 24:1–14.[CrossRef][Web of Science]
Schilling J-G, Meyer PS, Kingsley RH. Evolution of the Iceland hotspot. Nature (1982) 296:313–320.[CrossRef]
Schilling J-G, Kingsley R, Fontignie D, Poreda R, Xue S. Dispersion of the Jan Mayen and Iceland mantle plumes in the Arctic: A He–Pb–Nd–Sr isotope tracer study of basalts from the Kolbeinsey, Mohns, and Knipovich Ridges. Journal of Geophysical Research (1999) 104:10543–10569.[CrossRef]
Schilling J-G, Fontignie D, Blichert-Toft J, Kingsley R, Tomza U. Pb–Hf–Nd–Sr isotope variations along the Galápagos Spreading Center (101°–83°W): Constraints on the dispersal of the Galápagos mantle plume. Geochemistry, Geophysics, Geosystems (2003) 4. doi:10.1029/2002GC000495.
Schmitz MD, Vervoort JD, Bowring SA, Patchett PJ. Decoupling of the Lu–Hf and Sm–Nd isotope systems during the evolution of granulitic lower crust beneath southern Africa. Geology (2004) 32:405–408.
Skovgaard AC, Storey M, Baker J, Blusztajn J, Hart S. Osmium–oxygen isotopic evidence for a recycled and strongly depleted component in the Iceland mantle plume. Earth and Planetary Science Letters (2000) 194:259–275.[CrossRef]
Smallwood JR, White RS. Crustal accretion at the Reykjanes Ridge, 61°–62°N. Journal of Geophysical Research (1998) 103:5185–5201.[CrossRef]
Sobolev AV, Hofmann AW, Sobolev SV, Nikogosian IK. An olivine-free mantle source of Hawaiian shield basalts. Nature (2005) 434:590–597.[CrossRef][Medline]
Stalder R, Foley SF, Brey GP, Horn I. Mineral–aqueous fluid partitioning of trace elements at 900–1200°C and 3·0–5·7 GPa: New experimental data for garnet, clinopyroxene, and rutile, and implications for mantle metasomatism. Geochimica et Cosmochimica Acta (1998) 62:1781–1801.[CrossRef][Web of Science]
Stecher O, Carlson RW, Gunnarsson B. Torfajökull: a radiogenic end-member of the Iceland Pb-isotope array. Earth and Planetary Science Letters (1999) 165:117–127.[CrossRef][Web of Science]
Storey M, Duncan RA, Swisher CC III. Paleocene–Eocene Thermal Maximum and the opening of the Northeast Atlantic. Science (2007a) 316:587–589.
Storey M, Duncan RA, Tegner C. Timing and duration of volcanism in the North Atlantic Igneous Province: Implications for geodynamics and links to the Iceland hotspot. Chemical Geology (2007b) 241:264–281.[CrossRef][Web of Science]
Stracke A, Bizimis M, Salters VJM. Recycling oceanic crust: Quantitative constraints. Geochemistry, Geophysics, Geosystems (2003a) 4. doi:10.1029/2001GC000223.
Stracke A, Zindler A, Salters VJM, McKenzie D, Blichert-Toft J, Albarède F, Grönvold K. Theistareykir revisited. Geochemistry, Geophysics, Geosystems (2003b) 4. doi:10.1029/2001GC000201.
Stracke A, Hofmann AW, Hart SR. FOZO, HIMU, and the rest of the mantle zoo. Geochemistry, Geophysics, Geosystems (2005) 6:Q05007. doi:10.1029/2004GC000824.[CrossRef]
Sun S-S, Jahn B-m. Lead and strontium isotopes in post-glacial basalts from Iceland. Nature (1975) 255:527–530.[CrossRef]
Sun S-S, Tatsumoto M, Schilling J-G. Mantle plume mixing along the Reykjanes axis: Lead isotope evidence. Science (1975) 190:143–147.[Web of Science]
Takei H. Development of precise analytical techniques for major and trace element concentrations in rock samples and their applications to the Hishikari Gold Mine, southern Kyushu, Japan. In: PhD thesis. (2002) Okayama University. 164.
Taylor RN, Thirlwall MF, Murton BJ, Hilton DR, Gee MAM. Isotopic constraints on the influence of the Iceland plume. Earth and Planetary Science Letters (1997) 148:E1–E8.[CrossRef][Web of Science]
Thirlwall MF, Gee MAM, Taylor RN, Murton BJ. Mantle components in Iceland and adjacent ridges investigated using double-spike Pb isotope ratios. Geochimica et Cosmochimica Acta (2004) 68:361–386.[CrossRef][Web of Science]
Vervoort JD, Patchett PJ, Blichert-Toft J, Albarède F. Relationships between Lu–Hf and Sm–Nd isotopic systems in the global sedimentary system. Earth and Planetary Science Letters (1999) 168:79–99.[CrossRef][Web of Science]
Vogt PR. Asthenosphere motion recorded by the ocean floor south of Iceland. Earth and Planetary Science Letters (1971) 13:153–160.[CrossRef][Web of Science]
Walker GPL. Geology of the Reydarfjördur area, eastern Iceland. Quarterly Journal of Geological Society of London (1959) 114:367–393.
Walker GPL. Geological investigations in eastern Iceland. Bulletin of Volcanology (1964) 27:351–363.[CrossRef]
Watkins ND, Walker GPL. Magnetostratigraphy of eastern Iceland. American Journal of Science (1977) 277:513–584.
Welke H, Moorbath S, Cumming GL. Lead isotope studies on igneous rocks from Iceland. Earth and Planetary Science Letters (1968) 4:221–231.[CrossRef][Web of Science]
White N, Lovell B. Measuring the pulse of a plume with sedimentary record. Nature (1997) 387:888–891.[CrossRef]
Wood DA. Spatial and temporal variation in the trace element geochemistry of the eastern Iceland flood basalt succession. Journal of Geophysical Research (1976) 81:4353–4360.
Wood DA. Major and trace element variations in the Tertiary lavas of eastern Iceland and their significance with respect to the Iceland geochemical anomaly. Journal of Petrology (1978) 19:393–436.
Wood DA, Gibson IL, Thompson RN. Elemental mobility during zeolite facies metamorphism of the Tertiary basalts of eastern Iceland. Contributions to Mineralogy and Petrology (1976) 55:241–254.[CrossRef][Web of Science]
Wright JD, Miller KG. Control of North Atlantic Deep Water circulation by the Greenland–Scotland Ridge. Paleoceanography (1996) 11:157–170.[CrossRef][Web of Science]
Yaxley GM. Experimental study of the phase and melting relations of homogeneous basalt + peridotite mixtures and implications for the petrogenesis of flood basalts. Contributions to Mineralogy and Petrology (2000) 139:326–338.[CrossRef][Web of Science]
Yoder HS, Tilley CE. Origin of basalt magmas: an experimental study of natural and synthetic rock systems. Journal of Petrology (1962) 3:342–532.
Yokoyama T, Nakamura E. Precise determination of ferrous iron in silicate rocks. Geochimica et Cosmochimica Acta (2002) 66:1085–1093.[CrossRef][Web of Science]
Yokoyama T, Makishima A, Nakamura E. Evaluation of the coprecipitation of incompatible trace elements with fluoride during silicate rock dissolution by acid digestion. Chemical Geology (1999) 157:175–187.[CrossRef][Web of Science]
Yoshikawa M, Nakamura E. Precise isotope determination of trace amounts of Sr in magnesium-rich samples. Journal of Mineralogist, Petrologist and Economic Geologist (1993) 88:548–561.
Zindler A, Hart S. Chemical geodynamics. Annual Review of Earth and Planetary Sciences (1986) 14:493–571.[CrossRef][Web of Science]
Zindler A, Hart SR, Frey FA, Jakobsson SP. Nd and Sr isotope ratios and rare earth element abundances in Reykjanes Peninsula basalts: Evidence for mantle heterogeneity beneath Iceland. Earth and Planetary Science Letters (1979) 45:249–262.[CrossRef][Web of Science]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  J. HANSEN, D. A. JERRAM, K. McCAFFREY, and S. R. PASSEY The onset of the North Atlantic Igneous Province in a rifting perspective Geological Magazine, May 1, 2009; 146(3): 309 - 325. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||