SEARCH

SEARCH BY CITATION

Keywords:

  • Oman ophiolite;
  • extrusives;
  • geochemistry;
  • ICP-MS;
  • paleomagnetism

Abstract

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. The Oman Ophiolite
  5. 3. Results
  6. 4. Discussion
  7. 5. Conclusion
  8. Acknowledgments
  9. References
  10. Supporting Information

[1] The multielemental composition of 31 lavas sampled in the extrusive section of the Oman ophiolite was determined by inductively coupled plasma-source mass spectrometry (ICP-MS). This study allowed us to define clear geochemical criteria to characterize the different lavas types in Oman. Most of the Oman ophiolite extrusive sequence is composed of lavas of composition similar to present-day MORB with the exception of slight Nb-Ta negative anomalies (V1 magmatism). V1 lavas display REE patterns and Zr/Hf ratios in the range of MORB (Zr/Hf = 33.15–38.7). In contrast, the overlaying V2 lavas, which outcrop only in the northern part of the ophiolite, are REE depleted relative to V1 and display low Zr/Hf ratios (23.6–30.5). V2 lavas may be further classified into two sub-groups. V2 type I lavas display a continuous decrease from HREE to LREE. The upper V2 type II lavas are more depleted in HREE and MREE but they are enriched in LREE. They are also distinguished by their low Nb/Ta ratios (10.53–11.65) relative to other Oman basalts (V1: 12.5–14; V2 Type I: 12.35–14.4). As for N-MORB, melting of an upwelling mantle beneath an oceanic spreading centre formed V1 lavas. V2 resulted from fluid-enhanced melting of previously depleted mantle residual after V1 extraction. This process probably occurred at different melt/rock ratios thus resulting in the two V2 sub-groups, and was enhanced in the northern part of the ophiolite. V3 lavas overlie the V1–V2 sequence in the Salahi area. They display incompatible element rich patterns (LREE > MREE > HREE) and a high Nb/Ta ratio typical of within-plate basalts. In the different studied areas, the transition from one lava type to the other seems to correlate with the block rotations revealed by paleomagnetic data.

1. Introduction

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. The Oman Ophiolite
  5. 3. Results
  6. 4. Discussion
  7. 5. Conclusion
  8. Acknowledgments
  9. References
  10. Supporting Information

[2] The Oman ophiolite is a sliver of the Neo-Tethys oceanic lithosphere obducted onto the Arabian plate at the end of the Cretaceous. It comprises several large structural massifs, each exposing a more or less continuous ophiolitic sequence, from the mantle section to the upper oceanic crust [see Nicolas et al., 2000, and references therein].

[3] A paleomagnetic Euler pole analysis [Perrin et al., 1994], conducted on the volcanic sequence in the Salahi area (Hilti Massif) in the central part of the ophiolite (Figure 1), has shown that the movement of this area could be modeled by a single, large rotation (about 145–150°) around an Euler pole located close to the nappe (less than 200 km away). The question was then whether these results were representative of the Hilti Massif, or of the entire ophiolite nappe. In 1999, a field campaign was carried out to complete the paleomagnetic data set in terms of geographical coverage of the ophiolite. Sampling was focused on the upper extrusive sequence, as lavas are usually better paleomagnetic recorders than intrusive or sedimentary rocks.

image

Figure 1. (a) The Oman ophiolite (red area) is set in the northeastern part of the Arabian plate, along the coast of the Sultanate of Oman and of the United Arab Emirates on the Gulf of Oman. (b) It is composed of 12 large structural massifs (names indicated in black upper case) as shown on the map of the Oman ophiolite structures [after Nicolas et al., 2000]. Sampling locations are indicated by green dots.

Download figure to PowerPoint

[4] A major and trace element study was carried out together with paleomagnetic analysis. The objectives of the geochemical study were (1) to characterize the geochemical variability of the upper extrusive sequence of the ophiolite and, (2) to evaluate possible correlations between geochemical changes and paleomagnetic rotations, in order to better understand the geodynamic environment during which block rotation occurred.

2. The Oman Ophiolite

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. The Oman Ophiolite
  5. 3. Results
  6. 4. Discussion
  7. 5. Conclusion
  8. Acknowledgments
  9. References
  10. Supporting Information

2.1. Geological Setting

[5] The Oman ophiolite is the best-exposed piece of oceanic lithosphere at the surface of the Earth and also, one of the best-documented ophiolite, after the systematic work of several research groups (see for instance Boudier and Nicolas [1988], Coleman and Hopson [1981], Glennie et al. [1974], Lippard et al. [1986], Peters et al. [1991], and, more recently, Boudier and Juteau [2000]). It represents part of the Neo-Tethys ocean obducted onto the Arabian plate at the end of the Cretaceous (∼78 Myr [Hacker et al., 1996]).

[6] The ophiolite is made up of several large (up to 4500 km2) massifs aligned along the Oman coast (Figure 1b). Each massif exposes a more or less complete ophiolitic sequence, comprising a thick mantle section, gabbros and sheeted dykes topped by a 0.5–2 km thick extrusive sequence and interbedded oceanic sediments. Three magmatic sequences within the lavas constituting the extrusive section can be distinguished using petrographic and geochemical studies.

[7] The lower sequence, V1 [Beurrier, 1987; Ernewein et al., 1988] also called Geotimes by Pearce and co-workers [Alabaster et al., 1982; Lippard et al., 1986; Pearce et al., 1981] represents most of the exposed extrusives (>60% [Nicolas et al., 2000]). It consists mainly of poorly vesicular brownish pillow basalts with scarce massive lava flows. Its composition is close to mid-ocean-ridge basalts -MORB- [Alabaster et al., 1982; Einaudi et al., 2000; Ernewein et al., 1988]. The eruption of V1 lavas is associated with the accretion of the Oman paleo-ridge, dated at 94–95 Ma [Hacker et al., 1996].

[8] The middle sequence outcrops mainly in the northern part of the ophiolite and represents most of the remaining lavas (>35%). It is separated from the V1 sequence by a more or less continuous layer of pelagic sediments. Three units, Lasail, Alley and clinopyroxene-phyric, were recognized by Pearce and co-workers [Alabaster et al., 1982; Lippard et al., 1986; Pearce et al., 1981] and later merged together under the same name, V2, by Beurrier [1987] and Ernewein et al. [1988]. V2 lavas are of tholeiitic affinity and are distinguished from V1 by low-Ti and low incompatible trace element contents [Beurrier et al., 1989; Ernewein et al., 1988; Lippard et al., 1986]. V2 lavas were emplaced shortly after the V1 lavas (<2 Ma [Hacker et al., 1996]). They are thought to represent either the first stages of island arc volcanism in an immature arc environment [Beurrier et al., 1989; Pearce et al., 1981], either the product of the hydrous melting of the over-ridden lithosphere during intraoceanic thrusting at the Oman paleo-ridge [Boudier et al., 1988; Ernewein et al., 1988].

[9] The uppermost sequence, Salahi [Alabaster et al., 1982], later defined as V3 [Beurrier, 1987; Ernewein et al., 1988], is volumetrically the least significant. It has been observed only in the Salahi area in the Hilti massif, where it is separated from the lower sequences by a layer of pelagic sediments. It comprises alkaline to transitional within-plate basalts and they are thought to result from intraplate seamount volcanism, produced after the beginning of oceanic thrusting but preceding the end of the obduction of the ophiolite onto the Arabian continental margin [Ernewein et al., 1988; Lippard et al., 1986].

[10] Hydrothermal circulations, contemporaneous with the various phases of volcanic activity, induced secondary recrystallization in the lava and interbedded sediments [Alabaster and Pearce, 1985; Pflumio, 1991].

2.2. Sampling

[11] 31 samples were selected amongst the 1999 paleomagnetic sampling: 26 cores were collected by Perrin et al. [2000] in the northern part of the ophiolite and 4 by Weiler [2000] in the Sumail massif (Figure 1, Table 1). A V3 sample from the Salahi area (Hilti massif) earlier sampled by Perrin et al. [1994] was also analyzed.

Table 1. Location of Sampled Areas
Site NameMassifLava TypeLocation
LatitudeLongitude
AswadAswadV224°49′–24°50′56°19′
Western FizhFizhV224°33′56°28′
Eastern FizhFizhV224°35′56°23′
BarghahHiltiV124°13′56°31′
SalahiHiltiV324°11′56°33′
SaramiSaramiV223°52′56°49′
AbdahSumailV123°06′58°13′

[12] Distinction between V1 and V2 units was made according to the 1:50 000 geological maps published by the Bureau de Recherches Geologiques et Minieres [1993] and on the basis of the field criteria (sample color, flow jointing, and erosion resistance) defined by Alabaster et al. [1982]. However, the V1 and V2 volcanic units are often hardly distinguishable in the field and to find irrefutable discrimination criteria is one of the challenges of the study of the Oman ophiolite extrusive section. Even petrologic criteria, such as texture and clinopyroxene composition, are inefficient because both V1 and V2 basalts are hyalopilitic and contain identical low Ti and Na endiopside. Furthermore their groundmasses cannot be used because alteration has obliterated their original texture: both are altered to chlorite, quartz, hematite, Fe hydroxide, epidote, prehnite and zeolite. The best approach to discriminate the two series is geochemical analysis [Alabaster et al., 1982; Ernewein et al., 1988; Lippard et al., 1986; Pearce et al., 1981].

3. Results

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. The Oman Ophiolite
  5. 3. Results
  6. 4. Discussion
  7. 5. Conclusion
  8. Acknowledgments
  9. References
  10. Supporting Information

[13] Results are given in Table 2. Major element whole rock data were obtained by ICP-AES at the CRPG (Nancy, France). Whole rock trace element analyses were performed on a PQ2 ICP-MS (ISTEEM, Montpellier, France): Rare earth elements, Cs, Rb, Ba, Th, U, Zr, Hf and Y were determined following the method described by Ionov et al. [1992], and Nb and Ta by surrogate calibration following the procedure described in Godard et al. [2000].

Table 2a. Whole Rock Major and Trace Elements Analysis for the 31 Studied Lavasa
Sample name99 M 01599 M 03899 M 06099 M 008-199 M 008-299 M 02099 M 02799 M 04599 M 06599 M 07499 M 08699 M 09499 M 100-199 M 100-299 M 113-199 M 113-299 M 11699 M 12499 M 13599 M 14199 M 14699 M 15499 M 16099 M 16999 M 17989M152a99 M 18299 M 19099 M 19499 M 20299W90D99W107D99W120C99 W 156
  • a

    The lava type (V1, V2 or V3) according to the map and to field criteria (Map) and to the results of the geochemical study (this study) is indicated. Flow numbers are also indicated for samples analysed by Perrin et al. [2000]. Major element compositions were obtained by inductively coupled plasma-atomic emission spectrometry at the CRPG (Nancy, France). Trace elements (REE, Cs, Rb, Ba, Th, U, Pb, Nb, Ta, Zr, Hf and Y) were analysed by ICP-MS -PQ2+ Turbo, VG Elemental- (ISTEEM-University of Montpellier 2, France). The analytical procedure is given in Godard et al. [2000] for Nb and Ta and in Ionov et al. [1992] for other trace elements. Detection limit values are given in Godard et al. [2000].

SiteAswadAswadAswadAswadAswadAswadAswadAswadAswadAswadW FizhW FizhE FizhE FizhE FizhE FizhE FizhE FizhBarghahBarghahBarghahBarghahBarghahBarghahBarghahSalahiSaramiSaramiSaramiSaramiAbdahAbdahAbdahAbdah
Flow n°25711346892344556712345671234
Type
MapV2V2V2V2V2V2V2V2V2V2V2V2V2V2V2V2V2V2V1V1V1V1V1V1V1V3V2V2V2V2V1V1V1V1
This studyV1V1V1V2 tIV2 tIV2 tIV2 tIV2 tIIV2 tIIV2 tIIV2 tIIV2 tIIV2 tIV2 tIV2 tIV2 tIV2 tIV2 tIV1V1V1V1V1V1V1V3V2 tIV2 tIV2 tIV2 tIV1V1V1V1
% wt.
SiO248.6950.8848.955.3655.2554.7559.9856.8954.9255.9151.5451.8156.5956.552.2852.3557.1560.0459.561.9158.9159.0755.5461.1759.2847.2357.2356.7357.1555.8656.4956.954.5956
Al2O315.3715.4414.8214.8814.8415.1912.9814.915.0914.9515.0713.9915.3715.415.9915.9514.6713.9213.7213.2914.5214.4615.613.6214.4715.1413.5713.6713.3714.3215.4214.9514.9514.78
Fe2O312.9212.0612.0110.9710.988.669.7211.8311.2311.019.418.910.9610.979.779.7310.510.768.677.518.269.569.0488.610.7311.79.479.859.9111.0611.0211.3811.59
MnO0.20.150.210.430.430.250.210.160.150.160.140.120.160.160.640.630.180.150.120.10.090.130.140.10.120.160.190.140.10.170.190.180.30.33
MgO4.193.886.494.14.095.613.753.644.784.737.459.33.493.516.386.363.192.642.271.861.822.592.792.031.86.763.063.852.933.672.763.44.543.9
CaO9.154.838.484.334.345.83.625.76.465.767.37.664.944.974.534.564.742.573.063.33.432.163.683.273.48.65.767.667.126.921.932.042.082.16
Na2O3.655.994.584.764.735.114.423.673.333.814.173.64.834.854.934.925.866.25.725.336.896.476.576.673.645.14.694.623.767.316.496.15.92
K2O0.020.010.170.330.330.050.120.490.530.510.720.40.250.260.070.080.020.050.060.160.020.050.120.070.061.10.380.130.791.170.05  0.02
TiO21.831.731.410.730.740.470.660.60.470.510.380.320.730.740.560.560.790.781.151.21.231.161.31.211.31.350.870.740.80.831.621.712.051.61
P2O50.220.210.150.080.10.040.090.050.050.040.080.060.050.050.060.060.080.070.230.220.230.210.20.160.240.220.080.050.070.080.310.30.260.32
LOI3.624.653.073.954.13.894.271.932.842.444.213.782.462.434.614.632.662.635.34.924.44.014.43.63.865.072.332.73.033.162.652.733.53.19
Total99.8699.83100.2999.9299.9399.8299.8299.8699.8599.83100.4799.9499.8399.8499.8299.8399.8499.8199.899.899.899.899.8199.899.8100100.2799.8399.8399.8599.7999.7299.7599.82
Mg number39.339.251.942.842.756.443.638.146.046.261.367.638.939.056.656.737.832.934.433.130.635.138.233.729.555.834.344.837.342.633.338.244.440.2
ppm
Rb0.150.171.851.982.000.731.214.984.805.0211.175.472.312.310.310.330.150.120.631.390.160.420.910.570.6810.047.421.6817.5737.271.171.070.989.00
Sr162.7266.33104.7083.2284.0138.4469.88124.89118.09122.71342.31618.74140.90142.46137.62138.0781.1749.0761.5068.5040.2644.2999.7354.0053.691093.9057.7464.43122.03207.1755.8533.9455.52210.96
Yn.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.n.a.50.2148.6840.4239.25
Zr113.47110.2775.9739.5039.4321.3636.8230.6722.5524.7217.8816.0331.5730.9126.0926.2429.4725.80164.20151.58155.98160.96127.39105.33150.8583.0434.6019.0435.0453.30217.13172.18171.08160.99
Nb2.642.721.900.630.640.290.591.090.850.940.790.730.600.590.220.220.730.693.373.103.253.332.552.023.0811.720.680.680.640.986.596.245.274.30
Cs0.0090.0120.0320.0110.0160.0120.0230.1190.2170.1750.8770.9410.1250.0940.0170.0150.0090.0050.0220.0130.0050.0090.0120.0110.0080.0960.1570.1050.1760.4990.1110.1150.0680.355
Ba3.637.1512.7231.8632.465.3716.3227.5834.1929.2636.2230.6827.3627.6017.9718.083.023.938.6810.245.567.0514.268.708.59398.4422.5520.3233.6181.7310.238.7211.8360.16
La4.304.573.151.341.320.641.971.611.171.451.121.011.071.060.860.901.541.426.467.947.096.295.234.616.8012.001.361.381.302.0510.319.198.236.58
Ce13.5213.859.993.893.892.004.334.343.163.842.772.453.183.152.792.894.584.3320.6023.4320.6419.6815.4513.3521.0525.883.984.023.855.7829.1326.5422.5219.63
Pr2.212.221.620.660.660.370.750.630.450.560.390.340.530.520.520.530.750.703.263.693.183.132.572.133.353.200.670.680.671.004.404.133.363.02
Nd12.2412.009.333.853.892.244.303.392.403.012.051.823.093.053.183.264.454.2417.5219.1816.5816.8013.8211.4218.0414.494.033.913.925.7323.0421.8317.4415.98
Sm4.033.803.081.461.480.911.521.190.881.070.750.661.191.181.231.261.661.595.305.694.885.384.353.535.323.431.531.451.452.166.986.695.234.95
Eu1.581.421.220.610.620.340.750.460.360.440.300.260.520.540.500.490.720.631.631.821.511.831.591.261.901.290.640.600.590.892.422.311.921.68
Gd5.765.254.442.402.411.482.441.901.461.721.251.141.991.981.931.982.592.547.127.566.797.296.124.937.404.052.532.302.373.389.128.927.066.64
Tb1.020.940.810.440.450.280.440.380.290.330.250.230.380.380.370.370.490.481.231.311.161.281.040.851.250.630.460.420.430.631.581.541.241.15
Dy7.226.655.763.253.302.063.262.862.222.491.981.782.882.892.662.683.653.528.909.108.499.027.476.038.784.063.473.083.274.7111.0210.568.718.13
Ho1.511.421.210.710.720.450.700.640.500.560.460.410.640.650.560.580.780.761.911.891.841.911.571.261.820.810.760.660.711.032.312.211.831.72
Er4.344.133.552.092.121.312.091.991.591.751.481.321.932.001.701.742.302.295.725.435.465.664.633.645.342.222.291.952.093.066.506.135.205.04
Tm0.640.610.510.310.320.200.320.310.240.270.230.210.300.300.250.250.340.340.850.790.810.840.680.540.790.310.340.290.320.460.950.890.760.74
Yb4.124.053.372.042.091.322.122.131.671.861.641.472.052.061.651.672.292.235.575.095.295.494.373.565.111.942.331.982.133.056.145.704.914.76
Lu0.650.630.540.330.330.210.340.360.290.310.280.250.340.340.260.270.340.360.900.820.850.900.720.580.830.310.380.310.350.500.980.900.800.78
Hf3.173.102.311.301.320.731.211.020.780.860.600.551.091.090.920.931.221.024.784.384.544.753.833.184.402.311.270.811.221.855.914.814.714.16
Ta0.190.190.150.040.040.020.040.090.070.080.080.070.050.050.020.020.050.050.260.240.250.260.200.160.240.740.050.050.050.080.500.470.410.29
Pb4.511.630.7110.1210.312.011.850.691.550.842.721.341.451.383.113.191.341.001.632.391.901.661.811.741.181.390.500.731.101.620.350.510.660.57
Th0.240.290.180.120.120.060.120.290.210.240.250.220.130.130.070.070.130.120.460.420.430.450.440.370.421.940.120.100.120.180.670.600.560.44
U0.160.330.090.130.130.060.130.230.180.200.230.230.130.130.070.070.140.120.480.190.210.190.200.240.150.390.100.090.090.160.250.260.230.28
Table 2b. Values Obtained for BIR and BEN Standards During This Study are Compared to Those Given by Govindaraju [1994] (GSTL in Table 2b)a
 BIR1BEN
this studyGSTNLthis studyGSTNL
  • a

    Analytical reproducibility was also assessed by analysing 3 samples twice (99 M 008-1 & 2, 99 M 100-1 & 2, and 99 M 113-1 & 2).

ppm
Rb0.2048.8947
Sr101.431081360.801370
Y14.731630
Zr14.2115.5265.69260
Nb0.580.6109.24105
Cs0.0080.750.8
Ba6.4871040.61025
La0.620.6283.6082
Ce1.971.95159.87152
Pr0.380.3817.5017.5
Nd2.452.569.7267
Sm1.091.112.0812.2
Eu0.530.543.703.6
Gd1.871.859.929.7
Tb0.360.361.271.3
Dy2.692.56.476.4
Ho0.590.571.071.1
Er1.741.72.582.5
Tm0.260.260.320.34
Yb1.671.651.821.8
Lu0.270.260.270.24
Hf0.590.65.795.6
Ta0.040.045.785.7
Pb334.084
Th0.0320.0311.3110.4
U0.0110.012.582.4

3.1. Major Elements

[14] Alteration, revealed by thin section observation, leads to substantial modifications of the primary concentrations of CaO, Na2O and K2O, and to a lesser extent, may also modify MgO and SiO2 [e.g., Ernewein et al., 1988]. Nevertheless, as shown in Figure 2, Oman lavas show variations in MgO and SiO2 correlated with TiO2, an element thought to be less affected by alteration [e.g., Winchester and Floyd, 1977]. These variations are consistent enough to be interpreted in terms of various degrees of differentiation, and reveal two differentiation trends, one for V1 lavas and the second for V2 lavas, characterized by lower TiO2 contents [Ernewein et al., 1988]. However, because of their various degrees of differentiation, both the V1 and V2 lavas display a wide range of TiO2 content and the V2 and V1 fields may overlap.

image

Figure 2. (a) #Mg (100 × Mg/(Mg + Fe2+) with total Fe as Fe2+) and (b) SiO2 content (anhydrous wt.%) versus TiO2 content (anhydrous wt.%) of the studied lavas. These values are compared to a compilation of Oman lava compositions [Alabaster et al., 1982; Beurrier et al., 1989; Einaudi et al., 2000; Ernewein et al., 1988; Lippard et al., 1986; Rochette et al., 1990].

Download figure to PowerPoint

[15] The Abdah (Sumail Massif) and Barghah (Hilti Massif) V1 samples display high TiO2 concentration (>1.1 wt.%). Because of their lower SiO2 content, the Abdad samples are thought to be less differentiated than the Barghah samples. Most lava sampled in V2 sites display low TiO2 content (<0.9 wt.%), a characteristic considered to be typical of V2 lavas. These V2 lavas appear on average less differentiated than the V1 samples in this study (lower SiO2, higher #Mg (100 × Mg/(Mg + Fe2+)). The less differentiated V2 samples are from the western part of the Fizh massif and the most differentiated from the Sarami massif. The other V2 sites, from the eastern part of Fizh and in Aswad, show more variable compositions. This may be an effect of alteration as noted before. However, this process cannot explain the high TiO2 content (1.4–1.8 wt.%), typical of V1, in three of the Aswad samples supposed to be part of V2.

[16] Because of alteration, major elements do not allow V3 lavas to be distinguished. The studied samples appear slightly less differentiated than those analyzed in previous studies.

3.2. Trace Elements

[17] Four groups of lavas can be distinguished using trace element data (Figures 3 and 4). The first group comprises lavas sampled in V1 sites and the three high Ti Aswad samples. Their chondrite normalized Rare Earth Element (REE) patterns and contents are similar to or slightly more light REE (LREE) enriched compared to N-MORB (Figure 5). On Primitive Mantle (PM) normalized diagrams, most studied samples display moderate to strong Pb and Sr negative anomalies (also often encountered in MORB, e.g., Hofmann [1997]) and a slight Nb-Ta negative anomaly relative to neighboring elements. Note that depletions of Nb and Ta relative to La and Th are relatively unlikely to be modified by hydrothermal alteration, whereas Pb and Sr concentrations can be modified by alteration processes [e.g., Winchester and Floyd, 1977]. The studied lava compositions plot in the same field as V1 lavas analyzed in previous studies (Figure 5). They are referred to as V1 lavas following the classification proposed by Ernewein et al. [1988]. Both Abdah and Barghah lavas are trace element enriched relative to Aswad V1 samples, yet these lavas display similar TiO2 contents. Such relatively low TiO2 contents probably reflect late precipitation of Fe-Ti oxides and hence a higher degree of differentiation.

image

Figure 3. (a) V1, (b) V2 type I, (c) V2 type II and (d) V3 Rare Earth Element content normalized to Chondrite [Sun and McDonough, 1989]. Symbols are indicated in insert. The blue shaded area on Figures 3b, 3c, and 3d indicates the values obtained for V1 lavas in this study.

Download figure to PowerPoint

image

Figure 4. (a) V1, (b) V2 type I, (c) V2 type II and (d) V3 trace element contents normalized to Primitive Mantle [Sun and McDonough, 1989]. Symbols are the same as in Figure 3.

Download figure to PowerPoint

image

Figure 5. (a) Yb content (ppm), (b) Zr content (ppm) and (c) chondrite normalized La/Yb ratio (La/YbN) [Sun and McDonough, 1989] versus TiO2 content (anhydrous wt.%) of the Oman lavas. Symbols are the same as on Figure 2 and otherwise indicated in insert.

Download figure to PowerPoint

[18] The second and third groups of lavas were sampled in V2 sites and are referred to, respectively, as V2 type I lavas and V2 type II lavas. They are both REE depleted relative to V1 lavas (Figures 3 and 5). The V2 type I group comprises samples coming from Sarami, the eastern part of Fizh and 3 samples from Aswad. It is characterized by a steady decrease of REE content from heavy REE (HREE) to LREE on chondrite normalized diagrams (Figure 3), strong Nb-Ta negative anomalies and Pb and Sr positive anomalies relative to neighboring elements on PM normalized patterns. The samples coming from the western part of Fizh and the 3 remaining samples from Aswad form the V2 type II group. These lavas display “spoon-shaped patterns” on chondrite and PM normalized diagrams, with a strong enrichment in Th, Pb and Sr, a slight Nb and Zr negative anomaly relative to neighboring elements and low Nb/Ta ratios (Figures 36). Both groups plot in the same field as the previously analyzed V2 lavas (Figure 5). In the Aswad massif, where both V2 types are observed, V2 type II lavas were emplaced on top of V2 type I lavas.

image

Figure 6. (a) Zr/Hf versus La/YbN, (b) Nb/Ta versus Zr/Hf and (c) (La/Sm)N versus (Sm/Yb)N of the Oman lavas. Symbols are the same as on Figure 5.

Download figure to PowerPoint

[19] The V3 sample from Salahi (Hilti Massif) represents the fourth type of composition. It is characterized by a steady increase of trace element content of the most incompatible elements on chondrite and PM normalized diagrams (Figures 3 and 4).

[20] All analyzed samples display strong positive U anomalies, as well as a wide range of Cs, Rb, and Ba contents. These features mainly reflect the different secondary processes that affected the Oman lavas [e.g., Alabaster and Pearce, 1985; Pflumio, 1991; Regba et al., 1991].

4. Discussion

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. The Oman Ophiolite
  5. 3. Results
  6. 4. Discussion
  7. 5. Conclusion
  8. Acknowledgments
  9. References
  10. Supporting Information

4.1. Relationships Between Geochemistry and Paleomagnetic Orientations

[21] This study provides the first whole rock trace element data set on the Oman extrusive section since the geochemical studies published in the 1980s [Alabaster et al., 1982; Ernewein et al., 1988; Pearce et al., 1981]. These works have shown that the Oman ophiolite extrusive sequence is formed by three consecutive magmatic sequences, distinguished by their geochemistry. This study provides improved chemical criteria, little affected by alteration and differentiation, to discriminate the different magmatic episodes: (1) V1 lavas display LREE/HREE and Zr/Hf ratios in the range of N-MORBs (Figures 5 and 6a). (2) V2 lavas are REE depleted relative to MORB, and display lower Zr/Hf ratios (Figure 6a). On average, they display lower TiO2, REE and HFSE content than V1 lavas, although the two fields might overlap. V2 can be further divided into two sub-groups using ICP-MS data (Figures 6b and 6c): (3) Type I lavas display a continuous decrease from HREE to LREE and have Nb/Ta ratio similar to V1 (V1: 12.5–14; V2 Type I: 12.35–14.4). (4) Type II lavas were emplaced on top of type I lavas. They are more depleted in HREE and MREE but they are enriched in LREE. They are also distinguished by low Nb/Ta ratios (10.53–11.65) relative to other Oman lavas. (5) V3 lavas display incompatible element rich patterns (LREE > MREE > HREE) and a high Nb/Ta ratio typical of within-plate basalts. (It should be noted however that only one V3 sample was analyzed during this study.)

[22] In the Aswad massif, both V2 types are observed in areas mapped as “seamount volcanic sequences” by Alabaster et al. [1982] (i.e., where both the lower Lasail and upper Alley units where observed). On inter element diagrams, V2 type I lavas plot in the Lasail field and V2 type II lavas in the Alley field (Figure 5). However, we couldn't establish a definite correspondence, based on lava geochemistry, between V2 type I and V2 type II lavas and the Lasail and the Alley types defined by Pearce and co-workers [Alabaster et al., 1982; Pearce et al., 1981], because (1) only new analytical techniques allow precise determination of the geochemical criteria that we use to distinguish V2 type I from V2 type II and (2) complementary analyses on clinopyroxene to determine their primary composition were precluded by alteration or absence of phenocrysts. It should be noted also that, in contrast to Alabaster et al. [1982] observations, the lower V2 type I sub-unit (Lasail?) is not limited to specific areas. Instead, it has been found throughout the northern massifs of the Oman ophiolite. In contrast, the upper V2 type II sub-unit (Alley?) has a very limited outcrop extent. Because it was difficult to ascertain unambiguously the correspondence between the lava types and sub-types revealed by our ICP-MS study and those defined by Pearce and co-workers [Alabaster et al., 1982; Pearce et al., 1981], we chose to use the classification proposed by Ernewein et al. [1988] to describe the extrusive sequence of the Oman ophiolite.

[23] Paleomagnetic studies were carried out on the samples later analysed by ICP-MS (this study). Paleomagnetic results are summarized in Figure 8 [after Perrin et al., 2000; Perrin et al., 1994; Weiler, 2000]. For a given massif, the three main magma types (V1, V2 and V3) correspond to different paleomagnetic directions [Perrin et al., 1994]: V1 and V2 directions are statistically different but very close to each other, in agreement with the short time interval between their relative emplacement, while V3 directions are further away but in good agreement with African Late Cretaceous results when corrected for the post-Miocene opening of the Red Sea. At the scale of individual massifs within the Oman ophiolite, the V1 lavas record three significantly different paleomagnetic directions for the Aswad, Hilti-Sarami and Sumail massifs, probably indicative of tectonic rotations between these three massifs, at the time or after emplacement of the V1 lavas [Perrin et al., 2000]. Paleomagnetic data were obtained only for the V2 type I lavas. These lavas display homogeneous paleomagnetic directions, which suggest that, by the time of emplacement of the lower V2 series, the northern domain was behaving as a single tectonic unit.

4.2. Variations of Oman Extrusive Composition With Spatial Distribution and Time

[24] Both V1 and V2 lavas display HFSE/REE and HFSE/LILE fractionation, typically interpreted as the result of the contamination of their mantle source(s) by slab derived fluid-rich melts [e.g., Pearce et al., 1981; Peate and Pearce, 1998]. At each studied site, the “arc signature” increases from V1 to V2 lavas; this evolution is characterized by a decrease in the lavas HFSE/REE and HFSE/LILE ratios and an increase in LILE/REE ratios (Figure 7, Figure 8). The V2 sequence is characterized also, upward the volcanic sequence, by a decrease of, first, Zr/Hf (lower V2 Type I), then Nb/Ta ratios (upper V2 Type II). Such chemical variations are commonly attributed to arc magmatism [e.g., Peate and Pearce, 1998]. A continuous evolution in highly incompatible element fractionation is observed from the south to the north of the Oman ophiolite. The most highly incompatible element enriched are V2 type II lavas, which were only found in highly tectonized areas in the northernmost parts of the ophiolite (Fizh, Aswad). These variations can be interpreted as resulting from a stronger influence of fluid-rich melts in the northern part of the ophiolite. It should be noted that in spite of their differences in trace element concentrations, V1 and V2 lavas may display similar highly incompatible element ratios (e.g., Hilti V1 lavas and Sarami V2 lavas display similar Nb/Th ratios, Figure 7); this suggests small-scale heterogeneities in the mantle below the Oman paleo-ridge. They might reveal the progressive infiltration of the V1–V2 mantle source(s) by slab derived melts, and its effects on trace element enrichments.

image

Figure 7. N-MORB [Sun and McDonough, 1989] normalized (a) Ta/La, (b) Th/La, (c) Ce/Pb and (d) Zr/Sm ratios versus N-MORB normalized Nb/Th ratio for the studied lavas. Symbols are the same as on Figure 5 for the Oman lavas. They are compared to Ocean Island Basalts (OIB [Kalfoun et al., 1999]), basalts from open oceanic settings in fast spreading areas (East Pacific Rise: Pito Deep and Juan Fernandez microplate area (M. Godard and R. Hékinian, unpublished data, 1998), East Pacific seamounts [Niu and Batiza, 1997]) and from back arc basins (Lau Basin [Ewart et al., 1994], North Fiji Basin [Fleutelot, 1997]) as well as to more differentiated lavas from the Lau-Tonga arc and backarc systems [Ewart et al., 1994] and from the Izu-Bonin forearc [Murton et al., 1992].

Download figure to PowerPoint

image

Figure 8. Summary of the results of the paleomagnetic studies of Oman ophiolite lavas in areas sampled for both paleomagnetic and geochemical studies. Geochemical results are shown in Table 2. The V1 lavas from Sarami Massif were not analyzed; the results of the study of the V1 and V2 lavas from the Salahi Area (Hilti Massif) will be shown in a forthcoming paper. Paleomagnetic results were published in Perrin et al. [2000] for most samples coming from the northern part of the ophiolite, and in Weiler [2000] for samples coming from the Abdah area in the southern Sumail massif (Figure 1, Table 1). Paleomagnetic results for the Salahi area (Hilti massif) were published earlier by Perrin et al. [1994].

Download figure to PowerPoint

[25] Because the “arc signature” apparently is present from the early stages of ridge accretion in Oman lavas (V1 magmatism), Pearce and co-workers [Alabaster et al., 1982; Lippard et al., 1986; Pearce et al., 1981] proposed a model in which the entire Oman ophiolite was formed in a subduction zone environment. This model is also supported by the wide range of degree of differentiation that characterizes Oman lavas, similar to that observed in subduction-related magmatism [e.g., Davidson, 1996]. In this model, the continuous increase of the “arc signature” from V1 to V2 lavas is interpreted as the transition from accretion to localized arc-type volcanism in a marginal basin close to the subduction trench [e.g., Pearce et al., 1981]. However, the structure of the ophiolite (see review in Nicolas et al. [2000]), the short time interval between V1 and V2 emplacement, and the similar “arc signatures” found in both V1 and V2 lavas are difficult to reconcile with this model. In addition, it should be noted that recent marine geological studies have shown also the lack of clear correlation of trace element signatures with location relative to subduction zones [e.g., Bach et al., 1996].

[26] In the Oman ophiolite, changes in paleomagnetic orientations and magma types are correlated along strike, suggesting a connected evolution. Boudier et al. [1997], combining paleomagnetic data and structural observations, suggested that the Oman ophiolite recorded block rotation within a “micro-plate”, and that V2 lavas were formed during associated intraoceanic thrusting. The presence of fluids in the mantle below the future Oman ophiolite may then have favored locally higher degrees of partial melting or re-melting of peridotites [e.g., Wyllie, 1981] and therefore the formation of the depleted V2 lavas. However, there are no analogues in present-day fast spreading centers for V1-type lavas overlying V2 type lavas. More recently, Moores et al. [2000] suggested that the mantle source(s) of Tethyan ophiolites (including the Oman ophiolite) had been contaminated by an ancient subduction. In this model, the “arc signature” characterizing most Oman lavas would not be representative of the regional tectonic environment during the Oman ophiolite formation. However, this model does not explain the observed evolution of composition of the Oman extrusives from V1 to V2 magmatism.

[27] A better understanding of the processes controlling the lava arc signature in present-day oceanic environments should provide the means to improve our knowledge of the geodynamic environment in which the Oman ophiolite was formed.

5. Conclusion

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. The Oman Ophiolite
  5. 3. Results
  6. 4. Discussion
  7. 5. Conclusion
  8. Acknowledgments
  9. References
  10. Supporting Information

[28] Four groups of lavas in the Oman ophiolite can be unambiguously characterized using new ICP-MS data. The proposed classification and interpretation of the different magma sequences is in agreement with previously published geochemical studies of Oman lavas. At the base of the extrusive sequence, V1 lavas were formed via a process similar to the genesis of N-MORB, by melting of an upwelling mantle beneath an oceanic spreading centre. The overlying V2 lavas probably resulted from fluid-enhanced melting of previously depleted mantle residual after V1 extraction. This process probably occurred at different melt/rock ratios, thus resulting in the two V2 sub-groups and in the various highly incompatible element signatures from the north to the south of the ophiolite. The V2 type I lavas set at the base of the V2 sequence and the overlying V2 type II lavas could correspond to, respectively, the lower Lasail unit and the upper Alley unit defined by Pearce and co-workers [Alabaster et al., 1982; Pearce et al., 1981]. The third magmatic event, V3, resulted from melting of an incompatible element enriched mantle source.

[29] The three main magma types (V1, V2 and V3) correspond to different paleomagnetic directions. At the scale of the ophiolite, the V1 lavas record three significantly different paleomagnetic directions for the Aswad, Hilti-Sarami and Sumail massifs, indicating tectonic rotations between these three massifs before emplacement of V2 lavas [Perrin et al., 2000]. Incompatible elements indicate that the mantle source(s) of Oman V1–V2 lavas was progressively infiltrated by slab-derived melts, yet the variability of the “arc signature” within V1 and V2 lavas along the Oman ophiolite implies that this process did not control the V1–V2 transition. Block rotations recorded during or after V1 emplacement and the emplacement of V2 less than 2 Ma after V1 could be interpreted as being associated with a same complex accretion process, similar to that observed in “microplates” along present-day fast spreading centers.

Acknowledgments

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. The Oman Ophiolite
  5. 3. Results
  6. 4. Discussion
  7. 5. Conclusion
  8. Acknowledgments
  9. References
  10. Supporting Information

[30] We thank S. Pourtales and O. Bruguier for technical assistance on ICP-MS (ISTEEM, Montpellier, France) for the analyses of Oman, Pito deep and Juan-Fernandez lavas and F. Kalfoun for sharing data on French Polynesia OIB in advance of publication. This work has benefited from discussions with F. Boudier, C. Thoraval, and D. Bosch. We thank Hans Schouten and Julian Pearce for helpful reviews, Peter Kelemen and William White for editorial handling. This work was supported by the Centre National de la Recherche Scientifique (France).

References

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. The Oman Ophiolite
  5. 3. Results
  6. 4. Discussion
  7. 5. Conclusion
  8. Acknowledgments
  9. References
  10. Supporting Information
  • Alabaster, T., and J. A. Pearce, The interrelationship between magmatic and ore-forming hydrothermal processes in the Oman ophiolite, Econ. Geol., 80, 116, 1985.
  • Alabaster, T., J. A. Pearce, and J. Malpas, The volcanic stratigraphy and petrogenesis of the Oman ophiolite complex, Contrib. Mineral. Petrol., 81, 168183, 1982.
  • Bach, W., J. Erzinger, L. Dosso, C. Bollinger, H. Bougault, J. Etoubleau, and J. Sauerwein, Unusually large Nb-Ta depletions in North Chile ridge basalts at 36 50′ to 38 56′S: Major element, trace element, and isotopic data, Earth Planet. Sci. Lett., 142, 223240, 1996.
  • Beurrier, M., Geologie de la nappe ophiolitique de Samail dans les parties orientales et centrales des montagnes d'Oman, Univ. e Pierre et Marie Curie, Paris VI, Paris, France, 1987.
  • Beurrier, M., M. Ohnenstetter, B. Cabanis, J.-L. Lescuyer, M. Tegyer, and J. Le Métour, Géochimie des filons doléritiques et des roches volcaniques ophiolitiques de la nappe de Semail: Contraintes sur leur origine géotechnonique au Crétacé supérieur, Bull. Soc. Géol. Fr., 8, 205219, 1989.
  • Boudier, F., and T. Juteau, The ophiolite of Oman and United Arab Emirates, Mar. Geophys., 21, 2000.
  • Boudier, F., and A. Nicolas, The ophiolites of Oman, Tectonophysics, 390 pp., 1988.
  • Boudier, F., G. Ceuleneer, and A. Nicolas, Shear zones, thrusts and related magmatism in the Oman Ophiolite; initiation of thrusting on an oceanic ridge, Tectonophysics, 151, 275296, 1988.
  • Boudier, F., A. Nicolas, B. Ildefonse, and D. Jousselin, EPR microplates, a model for the Oman ophiolite, Terra Nova, 9, 7982, 1997.
  • Bureau de Recherches Geologiques et Minieres (BRGM), Geological maps, Ministry of Petroleum and Minerals, Muscat, Sultanate of Oman, 1993.
  • Coleman, R. G., and C. A. Hopson (Eds.), Oman ophiolite, J. Geophys. Res., 86, 283 pp., 1981.
  • Davidson, J. P., Deciphering mantle and crust signatures in subduction zone magmatism, in Subduction: Top to Bottom, Geophys. Monogr. Ser., vol. 96, edited by G. E. Bebout et al., pp. 251262, AGU, Washington, D.C., 1996.
  • Einaudi, F., P. Pézard, J.-J. Cochemé, C. Coulon, C. Laverne, and M. Godard, Petrography, geochemistry and physical properties of a continuous extrusive section from the Hilti massif, Oman ophiolite, Mar. Geophys. Res., 21, 387407, 2000.
  • Ernewein, M., C. Pflumio, and H. Whitechurch, The death of an accretion zone as evidenced by the magmatic history of the Sumail ophiolite (Oman), Tectonophysics, 151, 247274, 1988.
  • Ewart, A., W. B. Bryan, B. W. Chappell, and R. L. Rudnick, Regional geochemistry of the Lau-Tonga arc and backarc systems, Proc. Ocean Drill. Program Sci. Results, 135, 385425, 1994.
  • Fleutelot, C., Propagation de segments de dorsales océaniques et modélisation pétro-géochimique de l'évolution des réservoirs magmatiques (Dorsale Juan de Fuca et Bassin Nord-Fidjien), Ph.D. thesis, Universite de Bretagne Ouest, Brest, 1997.
  • Glennie, K. W., M. G. A. Boeuf, M. W. Hugues Clark, M. Moody-Stuart, W. F. H. Pilaar, and B. M. Reinhardt, Geology of the Oman Mountains, 423 pp., Neder. Mijn. Geol. Genoot., Delft, Netherlands, 1974.
  • Godard, M., D. Jousselin, and J.-L. Bodinier, Relationships between geochemistry and structure beneath a palaeo-spreading centre: A study of the mantle section in the Oman Ophiolite, Earth Planet. Sci. Lett., 180, 133148, 2000.
  • Govindaraju, K., 1994 compilation of working values and sample description for 383 geostandards, Geostand. Newsl., 18, 1158, 1994.
  • Hacker, B. R., J. L. Mosenfelder, and E. Gnos, Rapid emplacement of the Oman Ophiolite; thermal and geochronologic constraints, Tectonics, 15, 12301247, 1996.
  • Hofmann, A. W., Mantle geochemistry: The message from oceanic volcanism, Nature, 385, 1997.
  • Ionov, D. A., L. Savoyant, and C. Dupuy, Application of the ICP-MS technique to trace element analysis of peridotites and their minerals, Geostand. Newsl., 16, 311315, 1992.
  • Kalfoun, F., et al., Nb-Ta geochemistry: New constraints from ICP-MS analysis, paper presented at 10th EUG Meeting, Terra Nova, Strasbourg, France, 1999.
  • Lippard, S. J., A. W. Shelton, and I. G. Gass, The Ophiolite of Northern Oman, Geol. Soc. Mem., 178 pp., Backwell Sci., Malden, Mass., 1986.
  • Moores, E., L. H. Kellogg, and Y. Dilek, Tethyan ophiolites, mantle convection, and tectonic “historical contingency”: A resolution of the “ophiolite conundrum,”, in Ophiolites and Oceanic Crust; New Insights From Field Studies and the Ocean Drilling Program, edited by Y. Dilek et al., Spec. Pap. Geol. Soc. Am., 349, 312, 2000.
  • Murton, B. J., D. W. Peate, R. J. Arculus, J. A. Pearce, and S. Van der Laan, Trace element geochemistry of volcanic rocks from site 786: The Izu-Bonin forearc, Proc. Ocean Drill. Program Sci. Results, 125, 211235, 1992.
  • Nicolas, A., F. Boudier, B. Ildefonse, and E. Ball, Accretion of Oman ophiolite and United Emirates ophiolite. Discussion of a new structural map, Mar. Geophys. Res., 21, 147179, 2000.
  • Niu, Y., and R. Batiza, Trace element evidence from seamounts for recycled oceanic crust in the eastern Pacific mantle, Earth Planet. Sci. Lett., 148, 471483, 1997.
  • Pearce, J. A., T. Alabaster, A. W. Shelton, and M. P. Searle, The Oman ophiolite as a cretaceous arc-basin complex: Evidence and implications, Philos. Trans. R. Soc. London Ser. A, A300, 299317, 1981.
  • Peate, D. W., and J. A. Pearce, Causes of spatial compositional variations in Mariana Arc lavas: Trace element evidence, The Island Arc, 7, 479495, 1998.
  • Perrin, M., M. Prévot, and F. Bruere, Rotation of the Oman ophiolite and initial location of the ridge in the hotspot reference frame, Tectonophysics, 229, 3142, 1994.
  • Perrin, M., G. Plenier, J.-M. Dautria, E. Cocuaud, and M. Prevot, Rotation of the Semail ophiolite (Oman): Additional paleomagnetic data from the volcanic sequence, Mar. Geophys. Res., 21, 181194, 2000.
  • Peters, T., A. Nicolas, and R. G. Coleman, Ophiolite Genesis and Evolution of the Oceanic Lithosphere, 903 pp., Kluwer Acad., Norwell, Mass., 1991.
  • Pflumio, C., Evidences for polyphased oceanic alteration of the extrusive sequence of the Semail ophiolite from the Salahi block (Northern Oman), in Ophiolite Genesis and Evolution of the Oceanic Lithosphere, edited by T. Peters et al., pp. 313352, Kluwer Acad., Norwell, Mass., 1991.
  • Regba, M., P. Agrinier, C. Pflumio, and M. Loubet, A geochemical study of a fossil oceanic hydrothermal discharge zone in the Oman ophiolite (Zuha sulphide prospect): Evidence for apolyphased hydrothermal history, in Ophiolite Genesis and Evolution of the Oceanic Lithosphere, edited by T. Peters et al., pp. 352384, Kluwer Acad., Norwell, Mass., 1991.
  • Rochette, P., L. Jenatton, C. Dupuy, F. Boudier, and I. Reuber, Diabase dyke emplacement in the Oman ophiolite: A magnetic fabric study with reference to geochemistry, in Ophiolite Genesis and Evolution of the Oceanic Lithosphere, edited by T. Peters et al., pp. 5582, Kluwer Acad., Norwell, Mass., 1990.
  • Sun, S. S., and W. F. McDonough, Chemical and isotopic systematics of oceanic basalts: Implications for mantle composition and processes, in Magmatism in the Ocean Basins, edited by A. D. Saunders, and M. J. Norry, pp. 313345, Geol. Soc. London, London, 1989.
  • Weiler, P. D., Differential rotations in the Oman ophiolite: Paleomagnetic evidence from the southern massifs, Mar. Geophys. Res., 21, 195210, 2000.
  • Winchester, J. A., and P. A. Floyd, Geochemical discrimination of different magma series and their differentiation products using immobile elements, Chem. Geol., 20, 325343, 1977.
  • Wyllie, P. J., Plate tectonics and magma genesis, Geol. Rundsch., 70, 128153, 1981.

Supporting Information

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. The Oman Ophiolite
  5. 3. Results
  6. 4. Discussion
  7. 5. Conclusion
  8. Acknowledgments
  9. References
  10. Supporting Information
FilenameFormatSizeDescription
ggge337-sup-0001-t01.txtplain text document0KTab-delimited Table 1.
ggge337-sup-0002-t02a.txtplain text document9KTab-delimited Table 2a.
ggge337-sup-0003-t02b.txtplain text document1KTab-delimited Table 2b.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.