Introduction

The igneous rocks are generated mostly at divergent and convergent plate margins, but there are exceptional sites of volcanism where they are generated within plates by hotspots; areas in the earth’s mantle where columns of hot magma rise up to melt through the crust, resulting in igneous activity (e.g., Wilson 1963). Along with magmatic activity of convergent margin in the Chagai-Ras Koh magmatic arc (e.g., Siddiqui et al. 2015) and the divergent setting in Bela–Muslim Bagh–Waziristan ophiolites belt (Kakar et al. 2014a, b). the geology of western and north-western Pakistan is also characterized by the intra-plate volcanic activity (e.g., Khan et al. 1999). These rocks are found intruded in the Indian continent passive margin sediments: Jurassic Alozai Group (Kerr et al. 2010). Cretaceous Parh Group (Mahoney et al. 2002). and the subduction-accretion complexes found beneath the ophiolites (Gnos et al. 1998; Kakar et al. 2014a).

The Muslim Bagh Ophiolite is one of the largest ophioites in Pakistan and is underlain by an accretionary-wedge complex known as Bagh Complex. This complex comprises Mesozoic sedimentary, igneous, and metamorphic assemblages and is exposed around the Muslim Bagh Ophiolite (Fig. 1). The geology, biostratigraphy, and tectonics of this complex have already been reported by many workers (e.g., Otsuki et al. 1989; Kojima et al. 1994; Kimura et al. 1993; Naka et al. 1996 and Kakar et al. 2014a). The volcanic rocks of the Bagh Complex are identified as mid-oceanic ridge basalts (MORBs) and oceanic island basalts (OIBs) (Sawada et al. 1992). The petrogenesis of both types of the basalts is well documented in Kakar et al. (2014a). The petrogenetic model of the OIB-type basalts with reference to the regional tectonic reconstruction was still waited. This paper presents a brief account of geological, petrographic, and petrogenetic aspects of Hamrani volcanic rocks and relates their emplacement with mantle plume activity on the Ceno-Tethys Ocean floor prior to the passage of Indian plate over it.

Fig. 1
figure 1

Geological map of the Muslim Bagh area north-western Pakistan showing the locations of the Hamrani volcanic rocks (modified after Siddiqui et al. 2011; Kakar et al. 2014a)

Geology and petrography

The Bagh Complex is divided into five tectonic/biostratigraphic units: (1) sedimentary rock unit undivided (Permo-Triassic; Anwar et al. 1993), (2) sedimentary rock unit (Jurassic-Cretaceous; Jones 1961), (3) basalt-chert unit (Early-Late Cretaceous; Kojima et al. 1994), (4) hyaloclastite-mudstone unit (Late Cretaceous; Sawada et al. 1995). and (5) melange unit (Late Cretaceous; Mengal et al. 1994). The volcanic rocks (called here the Hamrani volcanics) are situated at about 12 km south of Muslim Bagh town (Fig. 1) and occur within the hyaloclastite mudstone unit in a narrow strip of Bagh Complex toward north-western margin of Jang Tor Gar Massif of Muslim Bagh Ophiolite (Fig. 1). These volcanic rocks comprise two lenticular bodies of porphyritic and amygdaloidal basalt and hyaloclastite and are found within the Hyaloclastite mudstone unit of the Bagh Complex. The lower basaltic flow is pillowed (30 to 1 m in diameter). The size of the pillow lava body is 30 × 300 m. The upper body of basaltic flow is massive and 15 × 150 m in size. The lower contact of the basaltic hyaloclastite is faulted and intercalated with siliceous mudstone and limestone derived from the sedimentary rock unit of Bagh Complex, whereas its upper contact is covered by subrecent to recent alluvial gravel (Fig. 2).

Fig. 2
figure 2

Generalized stratigraphic sequence of the Hyaloclastite-mudstone unit of Bagh Complex near Hamrani village

Most of the basalt samples collected from the study area is amygdaloidal. The main textures exhibited by these basalts include porphyritic, cumulophyric, and intersertal. The main minerals identified in these volcanic rocks include titaniferous augite, hornblende, phlogopite, plagioclase (An37–88) and devitrified volcanic glass, nosean, and olivine, with aegirine augite in some samples. Apatite, ilmenite, magnetite, and hematite occur as accessories, and chlorite, calcite, stilbite, antigorite, and clay minerals are found as secondary minerals.

Geochemistry

Analytical methods

Ten samples from Hamrani volcanic rocks were analyzed for major and trace elements (Table 1) in the Geoscience Laboratory, Geological Survey of Pakistan, Islamabad, Pakistan. For major elements, the sample powder (<200 mesh) was thoroughly mixed with lithium tetra-borate (flux) with a 1:5 sample flux ratio and the glass beads thus obtained were analyzed on X-ray fluorescence spectrophotometer, RIGAKU 3370E (XRF). For trace elements, powdered pellets of all the samples were prepared by putting 5–7 g powered sample (<200 mesh) in an aluminum cup and compressed between two tungsten carbide plates (within circular briquettes) at about 20 t per square inch pressure, in a hydraulic press. The pellets thus obtained were analyzed on XRF, using corresponding Geological Survey of Japan (GSJ) standard samples with every batch of ten samples. A check of precision of the instrument was made using JA-3 standard sample (Govindaraju 1989).

Table 1 Whole rocks major oxides and trace elements data of the Hamrani volcanic rocks

Geochemical characteristics

The samples from Hamrani are classified on the alkali versus SiO2 diagram (Fig. 3a) of Le Bas et al. (1986). It can be seen that all the rocks are alkaline and plot in the combined field of basanites and tephrites. In some instances, nonalkaline basalts show high contents of alkalies owing to submarine alteration (e.g., Staudigel 2003). In order to ascertain whether the alkaline character of these rocks is primary or due to alteration, elements considered immobile during alteration were used The Zr/Ti versus Nb/Y diagram, commonly used for the purpose (Fig. 3b), confirms the primary alkaline character of the studied Hamrani volcanic rocks.

Fig. 3
figure 3

a Alkali versus SiO2 classification plot (after Le Bas et al. 1986). b Zr/Ti versus Nb/Y classification plot (after Pearce 1996). c primordial mantle normalized spider diagram; d oceanic islands normalized spider diagram. Average N-MORB values in c and d are after Sun and McDonough (1989)

It is obvious that the Hamrani volcanic rocks SiO2 contents were affected by post-magmatic alteration processes. This is also reflected in nonsystematic variations in alkalies and CaO. The Na2O content in some samples reaches up to 6.43 wt.%, possibly due to albitization. The CaO content is also greatly variable as some of the samples show up to 16.64 wt.% CaO. The presence of calcite in the vesicles suggests that these enrichments can be due to partial replacement of some minerals by calcite. Similarly, the original contents of SiO2, Al2O3, and other major elements may have been changed due to alteration processes and secondary infillings of certain minerals like chlorite, chalcedony, and zeolites in the vesicles. Compared to N-MORB, the Hamrani volcanic rocks show a narrower range of TiO2 (2.36–3.43) and a wider range of MgO (5.09–7.14). The rocks have low abundances (13.22–16.15 wt.%) of Al2O3. They have highly variable concentrations of P2O5 (0.68–1.29 wt.%) and have low (48–55) Mg # (100 × Mg/(Mg + Fe2+). The major elements of the basalts show enrichment in K2O, Na2O, CaO, TiO2, MnO, and P2O5 and depletion in MgO, Fe2O3, Al 2O3, and SiO2 relative to N-MORB (Tables 1 and 2), which are consistent with the reported values of alkaline rocks (Weaver et al. 1987; Baker 1987).

Table 2 Comparison of trace elements and their ratios of Hamrani volcanic rocks with volcanic rocks from Bibai volcanic rocks, N- and E-MORBs, Oceanic islands, Hawaii, Reunion hotspot, and Mount Kenya

The Hamrani volcanic rocks are enriched in the whole range of large ion lithophile (LIL) elements (Rb, Sr, and Ba) and high field strength (HFS) elements (Nb, Zr, Ti, and Y) relative to average N-MORB. However, these amounts are consistent with reported values of basaltic alkaline rocks (Tables 1 and 2). The multi-elements spider diagrams are generally used to study the behavior of incompatible trace elements in rocks to constrain their source region with reference to N-MORB, primordial mantle or any other tectonically important composition.

All the samples from the Hamrani volcanic rocks were plotted on primordial mantle-normalized spider diagram (Fig. 3c). Incompatible trace element patterns exhibit variable enrichments of the whole range of trace elements (including LIL elements and HFS elements) relative to both N-MORB and primordial mantle. These patterns also show higher enrichments of LIL elements relative to HFS elements, which are consistent with an enriched mantle source (Pearce 2014). These patterns exhibit marked positive anomalies on Nb, which confirms their origins from an enriched mantle source (e.g., Hofmann 1997).

On OIB-normalized spider plot (Fig. 3d), the Hamrani volcanic rocks exhibit patterns almost parallel to those of OIB, suggesting a source similar to OIB. In the same diagram, all the trace elements are enriched, when compared with N-MORB. In compatible elements, these rocks are generally greatly variable and low in Cr (57–355 ppm), Ni (47–408 ppm), and Co (47–50 ppm) (Table 1).

Discussion

Nature of parent magma

SiO2 versus alkali diagram (Fig. 3a) and Zr/Ti versus Nb/Y plot (Fig. 3b) show that the Hamrani volcanic rocks are alkaline. SiO2 versus Na2O + K2O plot (Fig. 4a) further suggest that most of the rocks are mildly to strongly alkaline in nature. The primordial mantle, OIB-normalized spider diagrams (Fig. 3c, d) strongly suggest the derivation of the parent magma from the partial melting of an enriched mantle source. The primordial mantle normalized pattern also shows depletion in Y as compared to other HFS elements. Y is generally partitioned in the garnet, and the depletion of this element in the patterns indicate the presence of residual garnet in the parent magma source (Clague and Frey 1982; Wilson 1989). As mentioned earlier, the OIB-normalized pattern of these volcanic are parallel to OIB, indicating a mantle plume-related mantle source.

Fig. 4
figure 4

Tectonic discrimination diagrams of the Hamrani volcanic rocks: a Na2O + K2O versus SiO2 (after Schawarzer and Roger 1974). b Zr/Y versus Nb/Y (after Fitton et al. 1997); c Zr/Y versus Ti/Y (after Pearce and Gale 1977). d Ti versus V (after Shervais 1982). Filled triangles are the data from Hamrani volcanic rocks published in Kakar et al. (2014a)

The Z versus Zr/Y plot (Pearce and Norry 1979) provides useful information about the nature of source, degree of partial melting, and fractionations. Plot in this diagram suggest fractionated nature and about 15 % partial melting of an enriched mantle source for the parent magma. In addition, their Zr/Y (10.11–12.19), Zr/Nb (3.63–3.99), Ti/Zr (26.98–57.95), Y/Nb (0.30–0.39), K/Rb (361–553), K/Ba (07.42–46.25), and Ti/V (75.60–101.44) ratios (Table 1) are consistent with an enriched mantle source (Pearce and Norry 1979; Shervais 1982; Pearce 1983; Wilson 1989; Winter 2001). The aforementioned studies suggest that parent magma of these volcanic rocks was fractionated from a garnet-lherzolite mantle source.

The criteria generally used in support of basaltic rocks being primary melt from a mantle peridotite source or a product of fractionated liquids are (a) presence of mantle peridotite (lherzolite) xenoliths, (b) high magnesium number (Mg # = 100 × Mg / (Mg + Fe2+)) and low FeOt/MgO ratios, and (c) high contents of compatible elements (Ni, Cr, and Co). The basaltic magma coming from up to 30 % partially melted mantle peridotite source must have Mg # in a range of 68–75 (Green 1976; Frey et al. 1978; Hanson and Langmuir 1978). Gill (1981) has suggested an Mg # ≥ 67, whereas Tatsumi and Eggins (1995) have documented Mg # > 70 for primary basaltic magmas. In addition, basalts with Ni (250–300 ppm) and Cr (500–600 ppm) contents are considered to be derived from a primary mantle source (Perfit et al. 1980; Wilkinson and Le Maitre 1987). The Co contents in primary basaltic magma should range from 27 to 80 ppm (Frey et al. 1978)

No mantle lherzolite xenoliths have so far been reported from any of the Muslim Bagh volcanic rocks. The Mg # (48–55), Ni (47–408 ppm), Cr (57–355 ppm), and Co (47–50 ppm) contents in the Hamrani volcanic rocks are well below the values just mentioned for mantle-derived melts (Table 1). It is, therefore, concluded that the Hamrani volcanic rocks are a product of mantle-derived magma that fractionated en route to eruption, mainly due to olivine crystallization.

Tectonic setting

A number of plots and tectonomagmatic discrimination diagrams based on major, minor, or trace elements are designed to study the parent magma and tectonic setting of volcanic rocks. The diagrams based on major elements or LIL elements should be used with caution, as these elements are more mobile during post-magmatic alteration or metamorphic processes as compared to HFS elements (e.g., Weaver et al. 1987; Hastie et al. 2007).

The plots of samples from Hamrani volcanic rocks on various discrimination diagrams [Zr/Y versus Nb/Y (Fig. 4b), Nb/Y versus Ti/Y (Fig. 4c), and Ti versus V (Fig. 4d)], involving elements which are considered quite immobile during post-magmatic alteration or metamorphic processes, suggest that these volcanic rocks have an intra-plate origin and are erupted in an OIB setting. This is supported by their spider patterns (Fig. 3c, d) which exhibit enrichment of the whole range of trace elements with marked positive Nb anomalies, confirming their OIB signatures (Saunders and Tarney 1991; Hofmann 1997; Pearce 2014).

Nature of the source of parent magma

The Zr versus Zr/Y plot of Pearce and Norry 1979 (Fig. 5) provides useful information about the nature of source, degree of partial melting, and fractionation. On this diagram, the Hamrani volcanic rocks plot closely to the Reunion hotspot alkali basalts (the ~0–2 Ma; Fisk et al. 1988) suggesting a similar degree of partial melting (about 15 %) from an enriched mantle source for the parent magma, and a similar degree of fractionation for both the volcanic groups.

Fig. 5
figure 5

Zr/Y versus Zr plot of the Hamrani volcanic rocks (after Pearce and Norry 1979). The Reunion hotspot data is from Fisk et al. (1988)

The marked positive Nb anomalies in the spider diagram (Fig. 3c, d) are explained by the addition of this element in the magma source from the mantle plume (e.g., Hofmann 1997). The positive spikes of certain LIL elements are generally considered to have formed by incorporation of these elements in the source from the upper part of the crust (e.g., Pearce 1982). The spider patterns clearly suggest enriched magma sources for these volcanic rocks.

In Table 2, average trace element chemistry of the Hamrani volcanic rocks are compared with average N-MORB, E-MORB OIB, Reunion hotspot, Bibai volcanic rocks (Siddiqui et al. 2010). Hawaiian and continental rift basalts from the Mount Kenya. The Hamrani volcanic rocks show close similarity with Bibai, Reunion, Hawaii, and Mount Kenya. The source diagnostic ratios (including Zr/Y, Rb/Sr, and Zr/Nb; cf. Floyd 1991) of Hamrani, Bibai, Reunion, Hawaii, and Mount Kenya basalts are similar (Table 2), but K/Y, Sr/Y, and Ba/Y ratios in Hamrani, Reunion, and Hawaii volcanics have lower values, which suggest that the parent magma of the Hamrani volcanic rock was not affected by continental crustal contamination en route to eruption.

The petrogenetic considerations in the foregoing pages strongly suggests that the Hamrani volcanic rocks represent the intra-plate mantle plume activity possibly of the Reunion hotspot and were erupted during the passage of Ceno-Tethys Ocean floor prior to the passage of Indian Plate over it. Sinha and Mishra (1992) have suggested a similar origin for the Late-Cretaceous intra-plate volcanics found in the ophiolite melange in Ladakh, NW Himalaya. Reunion hotspot related origin for ~71 Ma intra-plates Bibai volcanic rocks in Pakistan is already documented (cf., Khan et al. 1999. Mahoney et al. 2002; Siddiqui et al. 2010; Kerr et al. 2010).

The Reunion Island and the Chagos-Laccadive ridge have been postulated to be formed in a hotspot-related setting (Whitemarsh 1974; Albarede and Tamagnan 1988). Backmann et al. (1989) suggested a hotspot origin for both the Chagos-Laccadive ridge and Deccan basalts. Duncan and Pyle (1988) and Fisk et al. (1988) have documented that the Deccan Trap and the Chagos-Laccadive ridge represent manifestation of Reunion hotspot formed by the passage of Indian continent and Indian Ocean floor during ~68–66 Ma, respectively. The Late Cretaceous to Pliocene (~120–05 Ma) journey of Ceno-Tethys Ocean floor, Indian continent, and Indian Ocean floor is illustrated in Fig. 6a–f. Figure 6a exhibits rifting of India from Gondwana, suturing of Afghan block with Eurasia, intra-oceanic convergence in Ceno-Tethys, and convergence of Ceno-Tethys below the Afghan block. In this figure, two important sutures are also shown including the Late-Triassic Herat-N-Pamir-Kun Lun (between Turan block and Farah block) and Early Cretaceous Panjao-C-Pamir-Banggang (between Farah block and Afghan block). The Turan and Farah blocks were separated from northern margin of Gondwana during Late Devonian and Early Permian and sutured with Eurasia in Early Permian and Late Triassic, respectively (Boulin 1988, 1990; Sengor et al. 1988; Stocklin 1989; Brookfield 1993; Metcalfe 1995; Zaman et al. 2013; Siddiqui et al. 2010, 2012; Rehman et al. 2011). Figure 6b shows eruption of Hamrani volcanic rocks when the Ceno-Tethys Ocean floor passed over the Reunion hotspot in Late Cretaceous (~81 Ma). Figure 6c indicates the eruption ~71 Ma of Bibai volcanic rocks when north-western continental margin of Indian plate passed over this hotspot. Figure 6d represents the Palaeocene (~66 Ma) eruption of Deccan basalt within the Indian continent when it passes over the Reunion hotspot. Figure 6e depicts the obduction of Muslim Bagh Ophiolite and associated Bagh Complex having slivers of Ceno-Tethys Ocean floor accompanying with ~81 Ma Hamrani volcanic rocks on to the north-western margin of the Indian plate during ~55 Ma, formation of Chagos Laccadive in the Indian Ocean floor and initial development of Katawaz basin between Afghan block and Indian plate (e.g., Qayyum et al. 1997; Kasi et al. 2012). Figure 6f shows collision of north-western margin of Indian plate and Afghan block in Pliocene (Treloar and Izatt 1993).

Fig. 6
figure 6

af Schematic cross sections showing the tectonic evolution of Ceno-Tethys Ocean and the Indian continent from Early Cretaceous to Pliocene (120–05 Ma) (based on the published work of Boulin 1988, 1990; Sengor et al. 1988; Stocklin 1989; Brookfield 1993; Treloar and Izatt 1993; Metcalfe 1995; Zaman et al. 2013; Siddiqui et al. 2010, 2012; Rehman et al. 2011). See the text for further detail

Conclusions

  1. 1.

    This petrogenetic study shows unambiguously that the Hamrani basalts belong to mildly to strongly alkaline intra-plate volcanic rock series.

  2. 2.

    The parent magma of these rock suites was generated by about 15 % partial melting of an enriched mantle source. Their low Mg # and low Cr, Ni, and Co contents suggest that the magma of these volcanic rocks underwent fractionation in an upper level magma chamber, en route to eruption.

  3. 3.

    3. Geochemical signatures of Hamrani volcanic rocks indicate that their parent magma are not affected by crustal contamination and suggest that these volcanic rocks may have been erupted on the ocean floor of the Ceno-Tethys when it passed over mantle plume. This may represent the earliest stage of impingement of the Reunion hotspot with the lithosphere’s base just prior to the migration of Indian Plate across it