Introduction

After the Arabia–Eurasia collision during the Oligocene and before the regional uplift of the Arabia–Eurasia suture zone during the late Miocene, a marine seaway connected the Mediterranean Sea and the Indian Ocean in the early–middle Miocene period (Hempton 1987; Allen and Armstrong 2008; Zadeh et al. 2017; Koshnaw et al. 2019, 2021; Mohammadkhani et al. 2022). During the early and middle Miocene, a relatively warm climate, known as the Middle Miocene Climatic Optimum (MMCO), was characterized by a high sea level (Zachos et al. 2001; Miller et al. 2020). The preserved dispatched shallow marine lower-middle Miocene carbonate rocks in the Arabia–Eurasia suture zone in the Kurdistan region of Iraq represent a record of this post-collisional marine incursion. The marine limestone beds of the Govanda Formation were deposited on the nonmarine clastic beds of the Suwais Group and are stratigraphically followed by nonmarine clastic beds of the Merga Group, suggesting a period of marine water incursion. These three rock units collectively, known as the Red Beds Series (RBS), were deposited in an intermontane basin setting in the Arabia–Eurasia suture zone after the collision (van Bellen et al. 1959; Jassim et al. 2006; Koshnaw et al. 2021).

The rare earth element and Y (REE + Y) signature of carbonates has been widely exploited in recent decades to reconstruct the history of deep-marine paleoenvironment (Piper 1974; Elderfield et al. 1990; Armstrong-Altrin et al. 2003; Madhavaraju et al. 2010; Nagarajan et al. 2011; Özyurt et al. 2020). REEs + Y have systematic changes in the chemical properties of the marine system due to the progressive filling of the outer shell. This causes different fractionations of REEs + Y in natural systems. Except for europium (Eu+2, Eu+3) and cerium (Ce+3, Ce+4), all REEs have a trivalent character which is redox-sensitive elements, and they show distinct geochemical behavior compared to other Lanthanides (Caetano-Filho et al. 2018; Özyurt et al. 2020). Through these geochemical characterizations, carbonate sedimentologists can reconstruct ancient environmental conditions. The carbonate succession of the Govanda Formation is well defined in terms of its stratigraphical and sedimentological characteristics (Smail 2015; Abdula et al. 2018). Sediments are platform carbonates which are represented by microfacies analysis indicating a transition towards a quieter lagoonal/reefal environment upward in the sequence and the compaction is the main diagenetic process that occurs between the grains that are returned to the close of the formation to the thrust zone and plate boundary (Smail 2015). Although there were several studies that dealt with Govanda Formation in terms of sedimentology, stratigraphy, and paleoecology, there has not been an attempt to investigate Bulk-rock geochemistry and elemental variation that provide proxies for paleo-redox conditions controls on their geochemistry. Consequently, this study focuses on the limestone strata outcropping in the Soran area in the northwestern Zagros fold–thrust belt within the intermontane Red Bed basin (NE Iraq) and presents a petrographic analysis coupled with REE + Y characterization of the limestone succession. The main purpose of the current study is (1) to determine ancient seawater composition using marine carbonates of intermountain basin NW Zagros thrust zone and affected by detrital materials; (2) to define the redox conditions and discuss the pale-oceanographic proxies for discrimination of depositional tectonic settings and paleoenvironments of the limestones and (3) to discuss the role of potential diagenetic influence on REE signals of ancient carbonates. The geochemical results were substantiated by field and petrographic studies and an extensive literature review.

Geological setting

From southeast Turkey and across the Kurdistan area of Iraq to southwestern Iran, the NW–SE trending Zagros orogenic belt extends for around 2000 km. It was produced during the Cenozoic Arabia–Eurasia convergence. During the Late Cretaceous (Berriasian–Maastrichtian), ophiolitic terranes were obducted onto the Arabian plate, leading to the Paleogene flexural subsidence and marine clastic input into the developing proto‐Zagros foreland basin. Following the Arabia–Eurasia collision, the Zagros orogenic belt expanded forelandward via shortening throughout the Neogene, resulting in clastic material input into the Neogene Zagros foreland basin (Alavi, 2004; Jassim and Goff, 2006; Koshnaw et al. 2020).

The RBS basin, which was developed at the suture zone of Arabia–Eurasia after the Oligocene, was accumulated in a narrow depozone on the plate boundary in a strongly subsiding intermontane basin (Jassim and Goff,

2006; Koshnaw et al. 2021). According to van Bellen et al. (1959), the RBS is subdivided into the lower Suwais Red Beds, the middle Govanda Limestone, and the upper Merga Red Beds. The Govanda Formation features a series of patchy reef bodies along the NW Zagros fold–thrust belt.

According to Buday (1980), the Govanda Formation consists of a shallow marine reef-fore-reef environment and is influenced by sediment supply by nearby paleo highs. The lower contact of the Govanda Formation is unconformable with the Suwais Group. The upper contact is gradational and conformable, gradually transitioning into the red shales and sandstones of the Merga Group (van Bellen et al. 1959; Buday 1980). Along the Arabia–Eurasia suture zone, the RBS rock units are thrusted onto by allochthonous rock units, Naopurdan and Walash. Naopurdan consists of fly-ash sediments and nummulitic limestone beds, whereas the Walash Group consists of igneous rocks (basalt and tuff), sandstone, shale, limestone, and radiolarite (Jassim and Goff 2006; Hassan 2012). The study area in the northeast of Erbil city in the Kurdistan region of Iraq includes three exposed sections with a variance thickness along-strike of ~ 144 m, 60 m, and 23 m, respectively. The first section is Sherwan Mazin section is in the northwestern extremity of the study area (lat. 36º 59′ 7″ N and long. 44º 12ʹ 18ʺ E). This stratigraphic secession and thickness changed toward the second (Hababiyan) section (lat. 36º 45′ 7.9″ N and long. 44º 28ʹ 52.5ʺ E). To the southeast, the Govanda Formation's thickness decreases further at the Smelan section (lat. 36º 40′ 2.3″ N and long. 44º 42ʹ 34.8ʺ E). All three sections are exposed in road-cut outcrops along the Arabia–Eurasia suture zone in the NW Zagros fault-thrust belt (Figs. 1 and 2).

Fig. 1
figure 1

a Regional tectonic map showing location of the study area the Arabia–Eurasia suture zone; b Generalized stratigraphic column depicting major rock units in the study area; c Geological map of the study area indicating locations of the studied sections in this study (Sh  Sherwan Mazen, Hb  Hababiyan, and Se  Semelan) and geographic distribution of the Govanda Formation. Figures modified after Sissakian and Foad (2014) and Koshnaw et al. (2021)

Full size image
Fig. 2
figure 2

Stratigraphic sections representing the Govanda Formation in the study area show notable thickness change from northwest to southeast

Full size image

Methods

For this study, 29 samples were collected from three sections to establish the petrographic and geochemical variations of the Govanda Formation. Thin sections have been prepared using the standard methods of thin section preparation and followed Dunham’s method for the classification of the studied carbonates. To eliminate any contamination, the selected samples were washed with distilled water. They were then air-dried and finely ground in a tungsten carbide mortar. For further refinement, the ground samples were sieved through a 200 ASTM mesh. X-ray fluorescence (XRF) was utilized with a spectrometer to conduct a chemical analysis of major oxides, including Al2O3, Fe2O3, TiO2, CaO, MgO, MnO, Na2O, K2O, and P2O5. One gram of sample was heated to 1000 °C in porcelain crucibles for two hours to measure the loss on ignition (LOI). Inductively coupled plasma-mass spectrometry (ICP-MS) was used to analyze the content of trace elements (Ba, Rb, Cs, Sr, Th, U, Y, Zr, Hf, Nb, Sc, V, Cr, Co, Ni, Cu, and Zn) and rare earth elements (La, Ce, Pr, Nd, Sm, Eu, Gd, Tb, Dy, Er, Ho, Tm, Yb and Lu) after four acid digestion (ME-MS81d and ME-4ACD81 codes). The abovementioned geochemical analyses were performed at the ALS International Laboratory in Spain. To ensure analytical accuracy and precision, which ranged from 0 to 9.3% and 2 to 8.8%, respectively, the internationally acknowledged standard materials OREAS102a and MRGeo08 were used as references. The Post-Archean Australian Shale (PASS) values given by Taylor and McLennan (1985) were used to prepare REE normalized diagrams.

Results

Petrography

The middle Miocene carbonates of the Govanda Formation are petrographically described according to Dunham’s carbonate classification (1962). The petrographic analysis reveals that the carbonate succession has five major microfacies in all studied stratigraphic sections. Benthic (Miliolida) foraminiferal Packstone microfacies (MF-1) is dominated by shallow water benthic foraminiferal packstone with rare wackestone layers. It has a mud-supported texture with some bioclasts in the micritic matrix and the occurrence of calcite cementation (Fig. 3a, b). Foraminiferal Wackestone microfacies (MF-2), Most of the grains are composed of skeletal grains like Miliolida, algae, and Alveolinids, and most of the non-skeletal grains are intraclasts in the micritized matrix (Fig. 3c, d). Foraminiferal Grainstone microfacies (MF-3) is mainly composed of fossils of miliolida, alveolinoids, and different kinds of bioclast as well as extra clasts, embedded within sparry calcite groundmass (Fig. 3e). Red algae boundstone (MF-4), The main components this microfacies are encrusted and bound with the other components, and micrite is red coralline algae sheet-like and lamellar skeletons; the predominant faunal association of these facies is coralline red algae, coral, and some foraminiferal-like milliolids (Fig. 3f, g). Lime Mudstone microfacies (MF-5) display a color range from dark gray to dark brown and contain detrital material like clay, silt, and sand. These facies contain a high content of micrite and a smaller number of grains, like some bioclasts with sutured seam stylolite, an irregular type with peaks of low amplitude, in metamorphosed lime mudstone is especially restricted at the upper part of Sherwan Mazin section (Fig. 3h, i). Foraminiferal Wackestone microfacies are primarily found in the middle parts of all studied sections.

Fig. 3
figure 3

Thin section views of representative microfacies found in the Govanda Formation; a, b Benthic (Miliolida) Packstone microfacies (MF-1), samples Sh8 &BS4, cp. c, d Wackestone microfacies (MF-2), samples Sh15, pp. & Sh9, cp. e Foraminiferal Grainstone microfacies (MF-3), sample BS3, pp. f, g Red algae Boundstone microfacies (MF-4), samples SG3, cp. & Hg2, pp. h, i Mudstone microfacies (MF-5), sample Sh13, pp., Bf Benthic foraminifera, Ba Bafflestone, St sutured stylolite, Alg laminated coralline algae, (cp crossed polarizers, pp plane polarized)

Full size image

Elemental geochemistry

Major and trace elements

Depending on the variation in the content of Al2O3, SiO2 vs. CaO oxides in the studied carbonate samples, which are the sources of different phases (clay and carbonate content), the samples are classified into massive limestones (MSL) and marly limestones (ML) (Fig. 4). The massive limestone is characterized by high purity, approaching the composition of standard calcite (56.03% CaO + 43.97% CO2; Haldar 2000); it consists of 52.40–55.10% CaO (53.87% av.), 40.80–43.50% LOI (42.41% av.), and 3.1% impurities including SiO2 (2.18%), Al2O3 (0.52%), Fe2O3t (0.40%), and traces of other oxides (Table 1). The marly limestone is less pure than the massive limestone but still can be considered semi-pure because it also contains relatively high amounts of CaO (42.90–52.90, 47.28% av.),

Fig. 4
figure 4

Bivariation of CaO% vs. Al2O3% and CaO% vs. SiO2% oxides separated the carbonate samples of the studied area into two types: massive limestone (MSL < 4% SiO2, < 1% Al2O3) and marly limestone (ML > 4% SiO2, > 1% Al2O3)

Full size image
Table 1 Major oxides contents (wt%) of the carbonates of the Govanda Formation, Northeastern Iraq
Full size table

LOI (34.60–42.20, 38.22% av.), and 14.5% impurities dominated by SiO2 (9.65%), Al2O3 (2.03%), Fe2O3t (1.08%),

and traces of other oxides (Table 1). The geochemical compositions reveal that the average content of the CaCO3 in ML is lower than in MSL (83.04% and 95.80%, respectively; Table 3). CaO has a negative correlation with SiO2, Al2O3, Fe2O3t, and P2O5 (r = − 0.750, − 0.76, − 0.53, and 0.012 in MSL; r = − 0.98, − 0.93, -0.89, and -0.64 in ML, respectively; n = 29; Fig. 5) indicating that the source of most major elements is not calcite but rather the impurities in these limestones such as clays, ferruginous material and detrital silica-bearing minerals such as quartz (Lokesh 2015)

Fig. 5
figure 5

Bivariate plots show the relationships of CaO% and ΣREE with some major oxides and trace elements and Eu/Eu* relations with K/Al and Na/Al ratios for the limestones of the Govanda Formation (blue circle = MSL, orange triangle = ML)

Full size image

Madhavaraju et al. 2016). In addition, CaO% content is significantly higher than the PAAS (Post-Archean Australian Shale)-normalization values of the major elements (Table 1).

Trace element contents (Table 2) were normalized using PAAS values (Taylor and McLennan 1985) and are plotted in Fig. 6. Within the large-ion lithophile elements (LILE) (Rb, Cs, Sr, Ba), Rb, Cs, and Ba are highly depleted compared to the PAAS in both MSL and ML. Sr content within these limestones is higher than the PAAS value (200 ppm; Taylor and McLennan 1985) (Fig. 6; Table 2). An average PAAS normalized of ferromagnesian (Transitional) trace elements (Co, Ni, Cr, and V) showing moderate depletion, except Cr, which shows enrichment in some samples. Within all the high-field strength elements (HFSE) of ML and MSL samples (Zr, Y, Nb, Hf, Th, and U), U shows enrichment in some samples, and others are moderate to highly depleted compared to PAAS (Fig. 6).

Table 2 Trace elements concentrations (ppm) for the carbonates from the Govanda Formation, Northeastern Iraq
Full size table
Fig. 6
figure 6

Post-Archean Australian Shale (PAAS) normalized Major and trace elements: A, C and D = Massive limestone; B, and E Marly limestone, F PAAS-normalized comparison of average MSL and ML values, ASC Arabian Sea Carbonate sediments (Nath et al. 1997), SMPC shallow marine Carbonate platform (Mazumdar, 2003)

Full size image

Rare earth elements

ΣREE content shows significant variation among the limestone types (average ML = 20.20 ppm and MSL = 7.02 ppm; Table 3). This indicates that the ML is enriched ~ 3 times in REEs relative to MSL, related to the impurities in these limestones. The low ΣREE in these limestones can also be linked to the conditions under which they have been deposited since the marine environments contain minor amounts of detrital constituents and heavy minerals (Piper 1974). REE concentrations in massive and marly limestones are normalized to PAAS values (Taylor and McLennan 1985) in Fig. 6. These limestones are exhibited: (1) seawater-like REE pattern with depletion of light rare earth elements (LREE) relative to heavy rare earth elements (HREE) (av. (La/Yb)n = 0.64, 0.71 and (Nd/Yb)n = 0.64, 0.76 in MSL and ML, respectively; Table 4); (2) moderate negative Ce* and positive La/La* anomalies (av. 0.79, 1.66, and 0.89, 1.08 in MSL and ML, respectively; Table 4); (3) distinctive positive Gd/Gd* and Eu/Eu* anomalies (av. = 1.34, 1.18 and 0.94, 1.14 in MSL and ML, respectively; Table 4); (4) moderately high Y/Ho ratio (av. Y/Ho = 41.5, 30.4 in MSL and ML, respectively; Table 4).

Table 3 Rare-earth element concentrations (in ppm) for the carbonates of the Govanda Formation, Northeastern Iraq
Full size table
Table 4 Rare-earth element ratios for the carbonates of the Govanda Formation, Northeastern Iraq
Full size table

Discussion

Source of the REE of the carbonates

The distribution pattern of REEs in limestones can be influenced by several factors, such as the addition of terrigenous materials from the continent (Piper 1974; McLennan 1989), scavenging process related to depth, salinity and oxygen levels (Elderfield 1988; Piepgras and Jacobsen 1992; Greaves et al. 1999), and diagenetic alterations which cause the REEs enrichments (Sholkovitz 1988). Therefore, clarifying the processes responsible for REE fractionation is essential since REEs are effective tracers of different geological and oceanographic processes (Piper 1974; Murray and Leinen 1993). The ΣREE content varies significantly between two types of limestone (MSL and ML) in all Govanda Formation sections, and they are less comparable with Arabian Sea carbonate sediments but relatively more or less comparable to shallow marine platform carbonate (78 ± 40, and 3.36 ± 2.55 ppm, respectively; Table 5) (Nath et al. 1997; Madhavaraju and Ramasamy 1999; Mazumdar et al. 2003). Significant positive correlation of ΣREE with SiO2, Al2O3, Fe2O3t, Zr, Th, and Y and negative correlation with CaO% (Fig. 5), indicates that the source of REEs are impurities, including clays, ferruginous cement, quartz and heavy minerals like zircon.

Table 5 Average values of elemental ratios of the studied area compared to the values of shallow marine platform carbonate and Arabian Sea Carbonate sediments
Full size table

The (La/Yb)n average value studied samples is similar to shallow marine platform carbonate ((La/Yb)n = 0.68 ± 0.47, Mazumdar et al. 2003). However, a few samples show values very close to that of the Arabian Sea carbonate sediments ((La/Yb)n = 0.85 ± 0.2, Nath et al. 1997; Table 5) and less than the value proposed by Condie (1991) [(La/Yb)n = 1] and Sholkovitz (1990) [(La/Yb)n = 1.3] for terrigenous particulate matter. In addition, the Dy/Yb ratio of the MSL: 1.17 and ML: 1.01 found in the studied samples are similar to the Dy/Yb ratio observed in the Shallow Marine Platform Carbonate, which is 1.10 ± 0.25 (Mazumdar et al. 2003), and the Arabian Sea carbonates ratio of 1.12 ± 0.11 (Nath et al. 1997; Table 5). This information suggests that these limestones retained their original seawater characters. Meanwhile, in normal seawater, the Er/Nd ratio is about 0.27 (De Baar et al. 1988) and limestones with high Er/Nd ratios effectively represent the seawater signature that marine carbonates retain. However, the addition of detrital material or the effect of the diagenesis process can result in the reduction of the Er/Nd value to below 0.1 (De Baar et al. 1988; German and Elderfield 1989; Bellanca et al. 1997; Nagarajan et al. 2011). The Er/Nd ratios of the carbonates of the Govanda Formation range from 0.08 to 0.23 in the MSL and from 0.09 to 0.14 in the ML. This supports the pristine character of most samples, but some limestone samples exhibit values closer to 0.10, comparable to Arabian Sea carbonates (Nath et al. 1997; Table 5), indicating the influence of detrital material.

The Y/Ho ratio is a useful proxy for identifying the presence or absence of siliciclastic components in carbonate samples (Bau 1991; Özyurt et al. 2020). Uncontaminated marine carbonates by terrigenous material exhibit Y/Ho ratios that vary widely between ~ 44 and 74 or approximate the chondritic value of ~ 28 (Liu et al. 2019; Özyurt M. et al. 2020). In the present study, variations in Y/Ho ratios were observed among different types of limestone Govanda Formation (Table 4). The Y/Ho ratio in chemical sediments decreases as Nd increases, approaching the Y/Ho ratios observed in schists, which also have high Nd content. This implies a significant alteration of the REY budget by detrital aluminosilicates (Viehmann et al. 2015). The Y/Ho vs. Nd diagram of the studied carbonate shows a low concentration of Nd suggesting that the samples lean more toward pure chemical precipitates with the presence of some detrital contaminants (Fig. 7a).

Fig. 7
figure 7

a = Y/Ho vs. Nd concentration for the Govanda carbonates (Mishra and Mohanty 2021; detritus and chondrite trends are adopted from Viehmann et al. 2015). b Correlation diagram of Th/U vs. Mn/Sr ratios (after Li et al. 2022). c, d Evaluation of REEs anomaly to discriminate the alteration of Govanda Formation Samples. (After Li et al. 2022)

Full size image

The carbonate rock units of the Govanda Formation were deposited in an intermontane basin, flanked by positive lands, and show little or no negative Eu anomaly (Table 4) this might have been derived from the weathering of diverse lithologies during deposition. Some elemental ratios such as K/Al and Na/Al, are helpful for determining the presence of detrital feldspar in bulk sediments (Madhavaraju and Lee 2009). However, in the present study, K/Al and Na/Al ratios do not show any correlation with Eu/Eu* (Fig. 5), this suggests that the observed positive Eu anomaly in these limestones may not be due to the inclusion of feldspars. As well, the hydrothermal alteration probably did not contribute to positive Eu anomalies since LREE is depleted relative to HREE. According to Bau (1991); Wang et al. (2018) and Mishra and Mohanty (2021), the positive europium anomalies in studied carbonates can be attributed to the reducing environment of deposition.

Evaluation of diagenetic alteration

Hydrothermal alteration and burial metamorphic can occur during the early and late stages of diagenesis, leading to the destruction of the original depositional information in carbonate rocks (Knauth and Kennedy, 2009; Li et al. 2022). The findings of our chemical analyses of major oxides in the present study revealed that calcite is the main mineral component of limestones of the Govanda Formation, which is consistent with the average content of CaO (51.6%). In addition, these limestones have a low Mg content, with an average of MgO values are: 1.16% in MSL and 1.39% in ML, indicating that little or a lack of dolomitization processes. Some geochemical parameters such as Mn/Sr ratios can be effectively used to evaluate the diagenetic alteration of carbonate rock (Kaufman et al. 1993; Wu, 2013). As the Mn content of precipitation is higher than that of seawater, the diagenetic alteration will be exhibited as an increase in the Mn/Sr ratio, therefore, the Mn/Sr ratio < 2 is considered to suggest that carbonate rock has not been affected by water interaction (Kaufman et al. 1993; Kaufman and Knoll 1995). The bivariate diagram between Mn/Sr and Th/U ratios can also reveal whether trace elements contain information relevant to the original deposition. Figure 7b illustrates a weak association between the Th/U ratio and Mn/Sr, suggesting that Th and U likely retain a record of the sedimentary information from ancient seawater. All limestone samples were characterized by Mn/Sr values of < 2 (Table 6) indicating that these limestones were not affected by diagenesis. REEs patterns in carbonate rocks are not affected by alteration or strong weathering (Webb and Kamber 2000). However, the behavior of certain elements, such as Ce and Eu, may vary due to differences in valence states under different redox conditions, leading to anomalies in their behavior and differentiation in rare earth element associations (Webb and Kamber 2000; Tribovillard et al. 2006; Li et al. 2022). According to Webb and Kamber (2000), in this study, we assessed the correlations between Ce/Ce*; Eu/Eu* and Pr/Yb; Pr/Sm as indicators of diagenetic alteration for all limestone samples of the Govanda Formation (PAAS-normalizing data). Except for the weak correlation between Ce/Ce* and Pr/Sm, the parameters (Ce/Ce*, Eu/Eu*, Pr/Yb, and Pr/Sm) showed a moderate linear relationship (with correlation coefficients of 0.60, 0.55 and 0.57) (Fig. 7c, d), indicating that the analyzed samples had not undergone diagenetic alterations.

Table 6 Major and trace elements ratios for the carbonates of the studied area
Full size table

Paleo-environmental analysis of the carbonates

The microfacies analysis revealed that the carbonates under investigation were formed in the shallow marine reef-fore-reef environment. This conclusion was drawn by combining the findings from petrography, and field studies and comparing them to standard facies models.

The enrichment of certain trace elements in marine sediments and their solubility in seawaters are influenced by oxygen levels, which determine their oxidation states (Hua et al. 2013; Tobia and Aqrawi 2016). For instance, uranium ions maintain a higher oxidation state (U+6) in seawater but when trapped in marine carbonates change to a lower oxidation state (U+4). On the other hand, thorium remains unaffected by the redox conditions of the water column and can persist indefinitely in its insoluble state of Th. (Hua et al. 2013). The use of authigenic U calculation can further support interpretations of the oxidative depositional condition using the following formula: authigenic U = Total U – (Th/3), the values of oxic conditions are indicated by < 2, whereas values exceeding 2 indicate dysoxic conditions (Wignall and Myers 1988). The results of the authigenic U calculations for the MSL and ML samples range from 0.58 to 4.39 (ave. = 2.66); and 0.88 to 5.22 (ave. = 2.84), respectively Table 6). These results suggest the limestones deposited in dyoxic conditions environment.

For evaluation of paleo-redox conditions, the trace element ratio indices, V/Cr, U/Th, Ni/Co, and V/(V + Ni), are used widely (e.g., Hatch and Leventhal 1992; Rimmer, 2004). According to Mishra and Mohanty (2021), the boundary of the Ni/Co ratio is at 3 for suboxic-anoxic and at 2 for dysoxic to oxic conditions. The V/(V + Ni) ratio is considered an anoxic condition of > 0.5, and suboxic to the oxic boundary at 0.3, while in oxic water the V/(V + Ni) ratio is < 0.5. We have employed the limited values of the U/Th ratio, which was suggested by Wignall and Richard (1996), to define the boundaries of the oxic conditions: oxic to suboxic is set at 0.1, the dysoxic limit ranges between 0.1 and 0.5, and the dysoxic to anoxic occurs at 0.5. According to the geochemical ratios, the studied carbonates were possibly deposited under suboxic–anoxic conditions within shallow water environments (Fig. 8).

Fig. 8
figure 8

Cross-plots of geochemical data of Govanda carbonates for assessing paleo-redox conditions: a Cr/V vs. Ni/Co; b V/(V + Ni) vs. Ni/Co; c U/Th vs. Ni/Co (Mishra and Mohanty 2021)

Full size image

The Ce* anomaly in carbonates is a more reliable proxy for determining ancient seawater redox state (Azmy et al. 2011; Liu et al. 2019; Özyurt et al. 2020). The depletion of Ce* anomalies represent modern oxygenated seawater (Bau 1991; Liu et al. 2019; Özyurt et al. 2020). However, it is important to mention that the partial leaching of clays containing high levels of REE + Y can mask any Ce anomalies present within the authigenic carbonate phase. Suboxic seawater mobilizes Ce-containing sediments, releasing Ce into the water column and creating a less negative to positive Ce anomaly in seawater (De Baar 1991). A significant negative Ce* anomaly can be categorized into three types: (a) smaller than < 0.5; (b) ~ 0.6–0.9; and (c) ~ 0.9–1.0, representing oxic, suboxic, and anoxic marine water, respectively (Chen et al. 2015; Özyurt et al., 2022). The negative Ce anomalies observed in this study (MSL: 0.63 to 0.96; ML: 0.79 to 0.95; Table 4) suggest that these limestones were deposited under a suboxic/anoxic environment, but the difference in Ce anomalies in these limestones indicates fluctuations of oxygen levels in the water which are close to those in Arabian sea sediments (Nath et al. 1997; Table 5).

Tectonic setting of Govanda Formation carbonates

Various studies have demonstrated that there were different tectonic settings for the deposition of Limestone successions, including continental margin basins (Wilson 1975), oceanic floors above the carbonate compensation depth (Liu and Schmitt 1990; Wang et al. 1986; Nath et al. 1997), local inland freshwater lakes (Alonso-Zarza 2003), and oceanic highs (Zhang et al. 2014). In general, limestones contain trace elements that originate from terrigenous materials and are scavenged from seawater (Murray et al. 1991; Siby et al. 2008). Numerous studies have demonstrated that the quantity and types of these terrigenous and trace elements in seawater are generally influenced by the plate tectonic environment of the basins (Murray et al. 1991; Holser 1997). Post-depositional alterations can modify the primary textures of limestones. Consequently, the relationship between the geochemistry and plate tectonics setting of limestones provides additional information for understanding ancient plate tectonic environments from the current variations in seawater chemistry (Webb and Kamber 2000). The Sr content of limestones found inland is lower than limestones in the open ocean, which exhibit the lowest Rb concentrations and the lowest Ba concentrations in limestones found along the continental margins (Zhang et al. 2017). The Rb–Sr–Ba ternary diagram to explain the studied samples reveals most of the samples fall within the marginal field (Fig. 9a). Furthermore, limestones from the continental margin can be distinguished from those from inland freshwater sources by their high Sr/Ba and Sr/Rb ratios in the Sr/Ba vs. Sr/Rb diagram. (Fig. 9b). Open-ocean limestones are mostly influenced by hydrothermal Fe–Mn-oxyhydroxide flux, while the inland and margin limestones are mostly influenced by terrigenous materials (Zhang et al. 2017). Al2O3 contents display a positive correlation with Fe2O3 for the marine limestones (Fig. 9c), but these two variables are invariant for the inland limestones. According to the variation trend of the Ce/Ce* ratios and the proximity of the basin to a continental plate margin, the limestones can be classified into two groups: inland + margins vs. open ocean (Zhang et al. 2017). Figure 9d reveals the limestones of the studied area represent the passive margin tectonic setting, which is less affected by terrigenous materials.

Fig. 9
figure 9

a Rb–Sr–Ba triangular diagram, b Sr/Ba vs. Sr/Rb, c Al2O3% vs. Fe2O3%, and d Ce/Ce* vs. Sm/Yb bivariate analyses represent the passive margin tectonic setting which is less affected by terrigenous materials (after Zhang et al. 2017)

Full size image

Conclusions

  1. 1.

    Depending on the content of Al2O3% and SiO2% and the purity of limestones, the Middle-Miocene Govanda Formation was subdivided into massive limestone (MSL) and marly limestone (ML), composed predominantly of calcite.

  2. 2.

    The total rare earth element (∑REE) content in the limestones of the Govanda Formation is lower compared to the Arabian Sea carbonate sediments. Light rare earth element depletion (LREE) and enrichment of heavy rare earth element (HREE) and (La/Yb)n ratio in the limestone samples indicate that the seawater REE pattern was preserved despite the presence of a degree of detrital contamination that affected the REE content and element ratios of these samples.

  3. 3.

    The limestone beds have a consistent seawater-like REE pattern, indicating the retention of seawater characteristics in the limestones, slightly negative Ce anomalies (Ce/Ce* = 0.79 in MSL and = 0.89 in ML), relatively positive Eu anomalies (Eu/Eu* = 1.18 in MSL and = 1.14 in ML), and Y/Ho ratios are moderately high (37.69). The REE + Y pattern of samples mainly retains its original characters, but the positive correlation of ΣREE with Th, Zr, and Y confirms the detrital inputs source of REEs.

  4. 4.

    The Y/Ho vs. Nd diagram of the studied carbonate shows a low concentration of Nd suggesting that the samples lean more toward pure chemical precipitates.

  5. 5.

    The weak association of Mn/Sr with Th/U ratios and Mn/Sr values of < 2 indicates that these limestones were not affected by diagenetic alterations.

  6. 6.

    Authigenic U, trace element indices (V/Cr, U/Th, Ni/Co, and V/(V + Ni)), and negative Ce anomalies reveal the suboxic to anoxic conditions of deposition.

  7. 7.

    The microfacies revealed a shallow marine reef-fore-reef environment for the Govanda carbonates.

  8. 8.

    Al2O3% contents display a positive correlation with Fe2O3% and are mostly related to marine limestones rather than inland limestones. The Rb–Sr-Ba ternary diagram, and Sr/Ba vs. Sr/Rb, Al2O3% vs. Fe2O3%, and Ce/Ce* vs. Sm/Yb bivariate analyses imply that the studied limestones originated in the passive margin tectonic setting.