Introduction

East and West Gondwanaland collided during the Pan-African orogeny (950–550 Ma), forming a major Precambrian orogenic belt (Kröner 1984; Stern 1994). Part of this belt is the Arabian–Nubian Shield, which comprises accreted island arcs and continental fragment terranes from the rifting of Rodinia and the opening of the Mozambique Ocean (> 870 Ma). The amalgamation of these geological terranes occurred along ophiolite suture zones that separate the East and West Gondwana fragments (~ 715–620 Ma) (Stern 1994; Abdelsalam and Stern 1996; Johnson et al. 2011; Abd-Allah et al. 2014; Matsah et al. 2018; Baggazi et al. 2019b; AlKashghari et al. 2020). Generally, post-collisional extension and relaxation (~ 620–540 Ma) reactivated terrane boundaries and are characterized by igneous intrusions of and plutons of linear to curved subvertical dike swarms of various compositions, and different aspects of graben sedimentation (Al-Shanti and Mitchell 1976; Greenwood et al. 1976; Stoeser and Camp 1985; Genna et al. 2002; Collins and Pisarevsky 2005; Johnson et al. 2011; Abd-Allah et al. 2014; Matsah et al. 2018; Baggazi et al. 2019a; El-Fakharani et al. 2019; AlKashghari et al. 2020; El-Fakharani et al. 2020). Igneous intrusions often follow northwest-southeast trending transform faults referred to as the Najd Faults System (NFS). This fault system also accommodated the formation of post-amalgamation basins in the ANS (Genna et al. 2002; Johnson and Woldehaimanot 2003; Johnson et al. 2011; Baggazi et al. 2019b; El-Fakharani et al. 2021) (Fig. 1).

Fig. 1
figure 1

Tectonic Map of the Arabian Shield (after Nehlig et al. 2002; Johnson and Woldehaimanot 2003). Small rectangle shows the location of the study area

Modern remote sensing techniques offer effective information regarding lithologic discriminations, mineral explorations (Zoheir and Lehmann 2011; Pour et al. 2017; Emam et al. 2018; Matsah et al. 2018; AlKashghari et al. 2020), and lineament extraction which is usually carried out, based on digital satellite image data, either manually (Suzen and Toprak 1998) or automatically (Koike et al. 1998; Casas et al. 2000; Mostafa and Bishta 2005; Bishta 2018) to facilitate structural mapping and mineral explorations. However, automatic alinements extraction could lead to selecting non-structural lineaments such as roads, and cliff edges. Therefore, to avoid mixing up lineaments, dikes, and non-structural features, the lineaments and dikes extractions were done manually in this study.

Geochemical analysis is a powerful technique to distinguish dike compositional types and define their tectonic environments, where incompatible element signatures help to determine magma sources and corresponding tectonic settings (Pearce et al. 1984; Rollinson 1993; Hargrove et al. 2006; Abd-Allah et al. 2014; Asran et al. 2017; Pearce and Cann 1971; Pearce 1996).

Volcanic dikes in the western Arabian Shield exhibit different sizes, compositions, and cross-cutting relationships (Matsah et al. 2018). In the study area, dike swarms with different orientations and corresponding field relative ages are presented. Dike swarms of the study area widely vary in thickness and compositions, where they cut the surrounding older metasedimentary, metavolcanic, and granitic rock units. These dike swarms are prominent but do not extend into other younger rocks; their thickness varies from a few centimeters up to several meters across (up not 5 m thick), and extend along strike up to 1 km. These dike swarms also vary in composition and probably formed in different tectonic environments. This study applies remote sensing processes including manual lineaments and dikes extraction from SPOT 7 satellite images, field observations, and geochemical analysis to identify the geochemical compositions of dike swarms, relationship between structural lineaments and dike swarms, and evaluating the relative age, magma source, and tectonic settings of different types of dike swarms in the study area.

Geological setting and lithology

The study area (Hada Alsham area) is located within the Jiddah terrane and to the north of the NE-oriented Fatima Shear Zone (Fig. 1). The Jiddah terrane contains Cryogenian tectonic ophiolitic rocks, metavolcanic and metasedimentary rocks, volcano-sedimentary rocks, and plutonic intrusions, in addition to Ediacaran sedimentary and subordinate volcanic rocks as well as dikes (Johnson et al. 2011; Johnson and Kattan 2012; Abd-Allah et al. 2014; Matsah et al. 2018; Baggazi et al. 2019a). The area under consideration mainly contains metasedimentary, metavolcanic, and igneous complexes that are intruded by different sets of mafic and felsic dikes that are usually less than 5 m thick (Fig. 2a).

Fig. 2
figure 2

a Geological map of the study area showing major rock units, b manually extracted structural lineaments (yellow) and dikes (light blue) from a SPOT-7 satellite image, with associated rose diagrams showing their frequency distribution

According to Johnson (2006), the Precambrian rocks in the Hada Alsham area are represented by the Samran group, Hishash granite, Kamil suite, and Shayma Nasir group, while Cenozoic rocks are represented by the Haddat AshSham Formation, Usfan Formation, and basaltic lava (Harrat) flows (Taj 2011). The rocks in the study area (Fig. 2a) encompass Precambrian and Cenozoic units, and more details are provided in Table 1 (Moore and Al-Rehaili 1989; Taj 2011). Slightly foliated Precambrian mafic to felsic dike swarms with different orientations cut the older Precambrian rock units in the study area (Moore and Al-Rehaili 1989). To the south of the study area, similar deformed and mainly NE to ENE and NW to WNW trending mafic and felsic dike swarms intrude the older Precambrian pre-Fatima basement but not the Fatima Formation, indicating that these dikes are older than the deposition of these Fatima sedimentary rocks (Matsahet al. 2018). The pre-Fatima basement, dike swarms, and the Fatima Formation were affected by NW–SE compression, resulting in the folding and thrusting of the Fatima Formation (Abd-Allah et al. 2014; Matsah et al. 2018; Baggazi et al. 2019a).

Table 1 Summary of main characteristics of all rock units and formations presented in the Hada Alsham area, based on previous studies (Moore and Al-Rehaili 1989)

The Hada Alsham dikes transect different Precambrian complexes, such as the Shiwan Complex (Rb–Sr age of 769 ± 39 Ma by Fleck 1981). Moreover, these dikes do not crosscut the Tertiary Haddat AshSham Formation. To the south of the study area, the dikes in the Wadi Fatima area only transact the rocks of pre-Fatima basement complexes such as the Qattanah Complex (763 ± 159 Ma by Fleck 1981, and Rumayda Complex (773 ± 16 Ma by Duyvermanet al. 1982). Additionally, the Rb–Sr ages for the lava and igneous bodies within the Fatima Formation are 576 ± 28 Ma and 592 ± 23 Ma, respectively (Moore and Al-Rehaili 1989). Based on the previous age dating and field relationships, the mafic and felsic dike swarms in the Hada Alsham area and Wadi Fatima area (to the south) were emplaced between < 763 Ma and > 576 Ma.

Methodology and results

Different methods and techniques are adopted in this study, including (1) lineament and dike orientation analysis based on satellite images, (2) field observations, measurements, and relationships, and (3) geochemical analysis techniques. These techniques and analytical methods are described below together with their corresponding results.

Lineament and dike extraction

Digital SPOT-7 data were processed in ArcMap to manually extract the structural lineaments likely fractures, faults, and dikes. The extracted structural lineaments in the study area exhibit trends that cluster in the NNE–ENE direction, NNW–NW direction, and E–W direction. Additionally, a statistical analysis of extracted dike swarms reveals that these dikes attain almost similar NE–ENE, NNW–NW, and E–W trends (Fig. 2b). Since the thickness of the dikes are relatively small (< 5 m thick), and the structural lineaments and dike trends are consistent with each other, and many of the dikes have thickness (up to 5 m thick), both are likely formed during the same tectonic events following the regional stress patterns at the time of emplacement rather than any pre-existing structures.

Field observations

Different rock types and cross-cutting relationships were observed in the field. These rocks are classified into three main groups: (1) host or country rocks, including ultramafic rocks, metasedimentary rocks, and granitic rocks; (2) hosted dike swarms of different compositions and orientations, including mafic dikes mostly trending NNW- to NW (mafic NNW dikes), mafic dikes mostly trending NE- to ENE (mafic NE dikes), felsic dikes mostly trending NE, and mafic dikes mostly trending E–W directions; and (3) tertiary rocks, including the Haddat AshSham Formation and basalt flows (Harrat). The rocks in the study area are described below in chronological order.

The ultramafic rocks are the oldest rock unite in the study area, which are very scarce and found only as elongated spots in the lower SW corner of the mapped area (Fig. 2a). The ultramafic units are highly deformed, metamorphosed and contain dark-green serpentinized pyroxenite and are crosscut by Precambrian intrusions related to the Hishash Complex (Moore and Al-Rehaili 1989).

Metasedimentary rocks are also not widely distributed in the study area; they mostly described from the most northeastern corner and, to a lesser extent, in the northwestern part of the study area (Fig. 2a). The metasedimentary rocks belong to the Madrakah Formation (Moore and Al-Rehaili 1989), which is encountered in the NW and NE corners of the mapped area (Fig. 2a) and is elongated in a NE–SW direction. They are highly foliated, sheared, and altered, outcropping as dark brown to yellowish-brown exposures. The metasedimentary rocks are cut by Precambrian mafic dikes of different orientations but mainly parallel to the foliation direction, and both the host rocks and dikes are cut by different fracture sets (Fig. 3). The rocks of the Madrakah Formation forms a part of the Samran Group and are unconformably overlain by the Fatima Group. The Madrakah group rocks consist of andesitic to basaltic lavas, andesitic volcaniclastic rocks, and pyroclastic rocks combined with meta-sandstone and marble. Regional greenschist facies metamorphism affected the Madrakah Formation that is intruded by a tonalite pluton with an Rb–Sr age of 759 ± 39 Ma (Fleck 1981).

Fig. 3
figure 3

Field photos showing highly foliated metasedimentary rocks dissected by a mafic dikes and b different sets of joints with about 15 cm of sinistral displacement

Tonalite-diorite-quartz diorite plutons are the third oldest rock units in the study area and occur as isolated spots along the southern border of the mapped area (Fig. 2a). These rock units are intruded by the Shiwan Complex, which is described later. The plutons are rather melanocratic and contain approximately 35% hornblende and biotite (Moore and Al-Rehaili 1989). Tonalite deformation could have been related to the NE-SW Fatima shear zone event (Moore and Al-Rehaili 1989), which underwent dextral transpression prior to the formation of the Fatima basin (Abd-Allah et al. 2014; Baggazi et al. 2019a).

Other igneous intrusions (diorite, granodiorite, and monzogranite) in the study area are very common in the northwestern part of the study area (Fig. 2a). These igneous rocks belong to the Shiwan Complex, which is part of a Kamil Suite and consists of hornblende tonalite, hornblende-biotite tonalite, quartz diorite, diorite, granodiorite, and monzonite (Johnson 2006). The Shiwan Complex is transected mainly by mafic NW and NE dikes (Fig. 4a). The Shiwan complex is overlain by Fatima Group rocks, which contain lavas with K–Ar ages of 576 ± 28 Ma (Brown et al. 1978) and 592 ± 23 Ma (Duyverman et al. 1982). In the southern and northeastern parts of the study area, other igneous rocks belong to the Hishash Complex (Fig. 2a), which intrude ultramafic rocks, metasedimentary rocks (the Madrakah Formation), tonalite, diorite and quartz diorite, as well as the Shiwan Complex (Moore and Al-Rehaili 1989). The pinkish to gray Hishash Complex is cut by various dikes and fractures of different orientations (Fig. 4b–d), and also transected by NW trending left-lateral faults (Fig. 4e, f). In hand specimens, the rocks of Hishash Complex mainly comprise quartz, orthoclase, plagioclase, with minor muscovite, biotite, and opaque minerals.

Fig. 4
figure 4

Fractured granitoid rocks and their hosted dikes, a mafic NW- and NNE dikes cutting granitoid host rocks of Shiwan Complex, b, c sheared mafic NNE-, to NE dikes cutting fractured granitic host rocks of the Hishash Complex, d xenoliths of clustered and/or elongated amphibolite within granite boulder of the Hishash Complex, e, f NW left lateral faults cutting the Hishash complex granitic rocks

Dike swarms

Generally, dikes are formed when magma fills fractures, which are normally the result of tensional, shear, or compressional regimes (e.g., Ragland et al. 1983; Ernst et al. 1995; Hamilton et al. 2001; Philpotts and Ague 2022). The tectonic setting during the formation of dikes plays an important role in the type of magma that rise and fill fractures. In the study area, dike swarms with different orientations, thicknesses (up to 5 m), and compositions cut the metasedimentary, metavolcanic, and granitic rock units. These dike swarms are prominent but do not extend into other younger rocks, such as the Haddat AshSham, Alshoumasi, and Fatima Formations. Relative cross-cutting mafic dike relationship can be determined from satellite images. The mafic NNE- and ENE-trending dikes are younger than the mafic NNW- to NW-trending dikes (Fig. 5a–c).

Fig. 5
figure 5

Spot-7 satellite image of the rocks of Hada Alsham area, a, b mafic NNE dikes cutting the mafic NW dikes, c mafic ENE dikes cutting the mafic N–S to NNW dikes

Field and crosscutting relationships (Fig. 6a–d) in addition to satellite image analysis (Fig. 5a–c) revealed that these dike swarms could be chronologically arranged as mafic N-S to NNW trending dikes, mafic NW trending dikes, mafic NE to ENE trending dikes, and rhyolitic NE dikes.

Fig. 6
figure 6

Dikes field relationship, a, b, and d mafic NE dike cuts a mafic NNW dike hosted by granitic rocks of the Hishash Complex, c a felsic NE dike cuts a mafic NNW dike hosted by granitic rocks of the Shiwan Complex

Mafic NW–SE trending dikes The orientations of mafic NW trending dikes generally trend between N10° and 60° W, with certain dikes even trending in an N–S direction (Fig. 5). Based on field crosscutting relations with other dikes, these represent the oldest dikes in the study area (Fig. 6). These dikes vary in length from a few meters to more than 1 km and from a thickness of a few centimeters up to 5 m in width. These dikes are highly fractures, fine- to medium-grained, and range in composition from basalt and basaltic andesites to andesites with a light gray to grayish-green color. These mafic NW trending dikes are highly fractured and jointed by at least three sets of vertical joints that systematically trend NNE-SSW, NE-SW and WNW-ESE. At certain locations, the mafic NW-trending dikes are displaced by right-laterally movement along NE-SW trending faults (Fig. 7a, b).

Fig. 7
figure 7

Different dike directions cut by faults, a right lateral fault (white dashed line) cuts thin mafic NNW dikes, and left lateral faults (yellow dashed line) cut mafic NE dikes, b right lateral fault cuts mafic NW dike, c mafic E–W dike along left lateral fault that cuts mafic NE dike, d left lateral fault that cuts felsic NE dike

Mafic NE–SW trending dikes These mafic NE trending dikes generally trend between N30° and 70° E and are younger than the mafic NW-trending dikes (Fig. 6a–c). These dikes are only slightly deformed and mainly basaltic in composition. They attain thicknesses ranging from 0.5 m to a few tens of meters (Fig. 7). At certain locations, mafic NE-trending dikes are generally transected by sinistrally displaced along E–W trending faults (Fig. 7a, c) that have been utilized by thin mafic E–W trending dikes (Fig. 7c), indicating that these thin mafic E–W trending dikes are the youngest.

Felsic NE–SW trending dikes Rhyolitic NE trending dikes are scarce, light yellowish red to dark red in color and exhibit fine-grained aphyric to porphyritic textures. The dikes range in thickness from a few centimeters to a few meters and extend up to a few hundreds of meters. In addition, these dikes are highly fractured and displaced by E–W trending sinistral faults (Fig. 7d). Cross-cutting relationship with the mafic NW dikes indicates that these rhyolitic NE dikes are younger (Fig. 6d).

Tertiary rock units

Tertiary rocks are common and represented by the Haddat AshSham Formation and basaltic flows (Harrat) (Fig. 2a). The Haddat AshSham Formation rests unconformably on Precambrian units and consists of conglomerates, sandstone, thin beds of shale, and shaly siltstones (Moore and Al-Rehaili 1989). These clastic beds are very conspicuous where they reveal variations in both color and thickness. Additionally, they are typically faulted and slightly to moderately dipping to the east. The basaltic flows occur as cap rocks horizontally cover any older rock units and comprise thick plateaus ranging in thickness from a few meters up to a few tens of meters. In hand specimens, these basaltic rocks exhibit a gray to grayish black color and a fine-grained texture with an occasional vesicles (Moore and Al-Rehaili 1989).

Geochemical analysis

Analytical techniques

Twenty-three samples were analyzed for major and trace elements using inductively coupled plasma-atomic emission spectrometry (ICP-AES) and inductively coupled plasma-mass spectrometry (ICP-MS), respectively, at the ACME Analytical Laboratories Ltd., Canada. The prepared powders of selected sample (0.100 g) are added to lithium metaborate/lithium tetraborate flux, mixed well and fused in a furnace at 1025 °C. The resulting melt is then cooled and dissolved in an acid mixture containing nitric, hydrochloric, and hydrofluoric acids. This solution is then analyzed by ICP-AES and ICP-MS for major, trace and REE, and the results are corrected for spectral inter-element interferences. Analytical precision, as calculated from replicate analyses, is better than 0.5% for major elements and varies from 2 to 5% for trace element. Duplicate crushed and pulverized (pulps duplicate) fractions are also analyzed to define samples homogeneity and analytical precision. A blank and certified standard reference materials (CRMs) STD SO-15 are used to monitor the accuracy of the analysis. Major, trace, and REE contents of the analyzed dike samples are listed in Table 2.

Table 2 Major and trace elements of dike swarms samples collected from the Hada Alsham area

Bulk major, trace, and rare-earth elements compositions

The process of hydrothermal fluid and rock interactions is complex and could involve changes in the mineral composition, texture, and chemical composition of the host rocks (Pirajno and Smithies 1992). Hydrothermal alteration of rocks can remove and introduce large amounts of more mobile elements such as Na2O, K2O, Rb, Sr, MgO and CaO, while rare earth elements (REEs) and high-field-strength elements (HFSEs) remain more immobile (Polat and Hofmann 2003). Therefore, the least altered samples were selected for geochemical analysis. Since alteration typically hydrates and or carbonates a rock, alteration minerals, such as calcite, actinolite, and chlorite, will break down at 1050 °C and the resulting loss of ignition (LOI) can assist in the investigation of the elemental distribution extent due to the presence of any excess H2O and/or CO2 (Ali et al. 2009). The LOI in the studied volcanic samples ranges from 0.3 to 3.5 wt.% (mean = 1.7%), indicating that most of the samples are only slightly altered. The samples obtained from the study area reveal large variations in Cr2O3 (< 13.7–157.3 ppm) and NiO (< 20–85 ppm) contents, as well as Mg# (= 100 × MgO/(MgO + FeOt) (9.4–59.1) (Table 2).

A classification diagram by Winchester and Floyd (1977) (Fig. 8a) indicates that the samples mainly include basalts, basaltic andesites, and rhyolites. The same rock types are indicated by the Nb/Y vs. Zr/Ti diagram revised by Pearce (1996), originally developed by Winchester and Floyd (1977). Thus, the mafic NW, NE, and E–W dike samples plot in the basalt, basaltic andesite and andesite fields, respectively, while the felsic NE dike samples plot in the rhyolite field (Fig. 8b).

Fig. 8
figure 8

Major elements and compatible trace elements for dike swarms from Hada Alsham area, Saudi Arabia. a Zr/TiO2 versus SiO2 diagram (Winchester and Floyd 1977) showing that the studied dikes are of two major types, mafic and rhyolitic; b Nb/Yb versus Zr/Ti diagram (Winchester and Floyd 1977) showing that the NNW, NE–ENE, and NE mafic dikes are basaltic and basalt andesite, and andesite; c SiO2 versus K2O relationship (Le Maitre 1989); and d AFM diagram (A = Na2O + K2O; F = FeO*; M = MgO). Calc-alkaline-tholeiitic boundary (solid curve) as proposed by Irvine and Baragar (1971), showing that the studied samples are calc-alkaline

The studied dikes indicate broad ranges of SiO2 (48.7–76.3 wt.%; Fig. 8c), TiO2 (0.04–1.42 wt.%), Al2O3 (12.22–18.95 wt.%), K2O (0.78–4.07 wt.%), and MgO (0.07–6.40 wt.%). The Le Maitre et al. (1989) diagram of K2O against SiO2 shows that the studied mafic NW-, NE- and E–W-trending dikes all belong to the same low- to medium-K suites, whereas the felsic NE-trending dikes extend into the high calc-alkaline series (Fig. 8c). In the AFM diagram (c.f., Irvine and Baragar, 1971; Fig. 8d), all dike swarms plot in the calc-alkaline field.

Trace element data further reveal the following geochemical characteristics: (1) La/Nb is generally > 1.1 (4.7 to 7.0), suggesting a typical tectonic setting of an intraoceanic island arc (Straub 1995); (2) spider diagrams (Fig. 9e–h) show large-ion lithophile elements (LILE) enrichment (Rb, K, Sr), and depletion of Ta, Nb, and Ti, which also typical for subduction zones; and (3) the REE patterns (Fig. 9a–d) are slightly fractionated, i.e., mafic NW dikes: [(La/Yb)N = 2.1–5.8], mafic NE dikes: [(La/Yb)N = 1.6–3.9], mafic E–W dikes: [(La/Yb)N = 2.5–3.4], and felsic NE dikes: [(La/Yb)N = 3.6–4.5]. The felsic NE dike samples have negative Eu anomalies (Fig. 9d). The heavy rare earth element (HREE) flat patterns indicate that elemental partitioning during melting or fractionation was not controlled by garnet.

Fig. 9
figure 9

Chondrite-normalized rare earth element (REE) patterns and primitive mantle-normalized trace element spider diagrams for the dike swarms. (All elements are normalized to the values of chondrite and primitive mantle reported by Sun and McDonough (1989)

Discussion and conclusions

Tectonic setting implication

Important trace and minor elements (e.g., Zr, Yb, Cr, Ti, Y, Th and V) and corresponding ratios can help infer the tectonic setting of the studied samples. The tectonic setting of the dike samples was inferred based on an analysis of different discrimination diagrams (Fig. 10). The following provides a discussion of the geochemical results based on each of these discrimination diagrams:

Fig. 10
figure 10

Trace element discrimination diagrams for mafic and intermediate lavas from different tectonic settings showing data for dike swarms samples: a (F) Cr–Ce/Sr (Pearce 1996); b Cr–Y (Pearce et al. 1981); c V–Ti/1000 (Shervais 1982); d Th/Yb–Ta/Yb (Pearce et al. 1981); e Th–Hf/3–Nb/116 (Wood, 1980); and f Th–Zr/117-Nb/16 (Wood 1980). WPB within-plate basalt, MORB mid-ocean ridge basalt, E-MORB enriched mid-ocean ridge basalt, VAB volcanic-arc basalt, BAB back-arc basalt, IAT island-arc tholeiitic, OIB ocean island basalt, Thol tholeiitic, Alk alkali, Tran transitional, C-A bas calc-alkali basalt

Visualization of our data in a Cr-(Ce/Sr) Cartesian diagram (Fig. 10a; Pearce 1982) demonstrates that all samples plot in the volcanic arc basalt (VAB) field. According to Ali et al. (2009), this diagram may be used for less alteration rocks. Relatively low Ce/Sr and low Cr contents are typical for VAB. In the Cr–Y tectonic discrimination diagram (Fig. 10b; Pearce et al. 1981), all samples plot in the VAB field. The Ti–V discrimination diagram (Fig. 10c; Shervais 1982) is highly recommended to distinguish different tectonic settings, such as mid-ocean ridge basalt (MORB), within-plate basalt (WPB) and tholeiitic arc basalt (TAB). This diagram indicates that most samples plot in the field of mid-ocean ridge basalt (MORB) and back-arc basalt (BAB), except for one sample that plots in the island arc tholeiite (IAT) field. Because Ti and V exhibit different behaviors in basaltic melts, this diagram is very sensitive and easily distinguishes tectonic settings as a consequence of partitioning changes with the oxygen fugacity. Consequently, MORB and WPB are less sensitive to oxidation than arc magmas (Rollinson, 1993). Th/Yb–Ta/Yb is one of the most effective discrimination diagrams (Fig. 10d; Pearce et al. 1981) because, in the absence of garnet in the residuum (Fig. 9), the Th/Yb ratios do not change much during melting and fractionation and are typically higher in subduction-related magmas. This occurs because This added to the mantle melt via flux melting during the subduction process, whereas Ta is not. Therefore, any change in Th versus Ta is clearly indicated in this discrimination diagram. All studied samples plot above the mantle array, suggesting that the corresponding arc magmas were generated in a convergent margin setting and/or became mixed with continental crust materials. The last ternary Th–Hf/3-Nb/16 and Th–Zr/117-Nb/16 diagrams (Fig. 10e, f; Wood 1980) reveal that all dike magmas were formed in an arc tectonic setting.

The tectonic setting of the felsic dikes (rhyolites) can be determined with the same discrimination diagrams as those considered for plutonic granite rocks (Stern et al. 1991). All felsic NE trending dikes plot in the volcanic arc/syn-collisional fields (VAG + syn-COLG) in the Y + Nb vs. Rb, Yb + Ta vs. Rb, Y vs. Nb, and Yb vs. Ta diagrams (Fig. 11a–d) (Pearce et al. 1984). The mafic and intermediate dikes—added for comparison—also plot in the same volcanic arc/syn-collisional fields. In addition, the petrogenesis of the felsic NE-trending dikes may represent extensive fractionation of the mafic magma, which intruded prior to these as mafic NE dikes, because their incompatible elements signatures are so similar (Fig. 9).

Fig. 11
figure 11

Trace elements tectonic discrimination diagrams. a Y + Nb vs. Rb; b Yb + Nb vs. Rb, c Y vs. Nb, and d Yb vs. Ta diagrams (after Pearce et al. 1984), showing the studied samples have been formed in a volcanic-arc/syn-collisional tectonic setting (VAG + syn-COLG). VAG volcanic arc granite, syn-COLG syn-collision granite, ORG ocean ridge granite, WPG within plate granitem post-COLG post-collisional granite

In conclusion, the cross-cutting field relationships and geochemical analysis results of the dike swarms confirm that the relative age order (from oldest to youngest) of the dikes is mafic NW dikes, mafic NE dikes, felsic NE dikes, and then mafic E–W dikes. Based on their similar incompatible element signatures, felsic NE dikes likely represent extreme differentiates from mafic NE dikes. Various geochemical discrimination diagrams and multi-elements diagrams consistently confirm that all dike swarms of the studied area in the west-central Arabian Shield formed within a volcanic arc/syn-collisional setting. Based on field relationships and geochemical discrimination diagrams, the dike swarms were likely emplaced during post-collisional extension and relaxation (~ 620–575 Ma) following upon the collision between the East and West Gondwana fragments (~ 715–620 Ma). Moreover, the dikes are younger than the Shiwan Complex (~ 769 Ma) and Madrakah Formation (> 759 Ma), and older than the Fatima Group (~ 576 Ma). Since dikes are tensional features, the dike swarms in the study area were likely emplaced between 620 and 575 Ma.