3.a. Cretaceous–Tertiary unconformity (CTU)

The Qulqula Radiolarian Formation consists of more than 2000 m of alternating intervals of bedded cherts, calcareous shales (Fig. 6) and limestones (mostly siliceous). The thin-section of the chert shows only recrystallized circular radiolarians (Fig. 6a), while the limestone consists of pelletal-bioclastic limestone (Fig. 6b). According to Buday (Reference Buday1980) and Karim (Reference Karim2003), the precise age and stratigraphic position of the Qulqula Radiolarian Formation is unclear because of its complex structures. Waddington (Reference Waddington1955) proposed that the formation is older than Jurassic. However, Karim (Reference Karim1975) inferred an age of Albian to Cenomanian from foraminiferal data, while Buday (Reference Buday1980) and Jassim & Goff (Reference Jassim and Goff2006) proposed that the formation is Lower Cretaceous.

Figure 6.

(a) Siliceous limestone of the Qulqula Radiolarian Formation at the Qulqula Gorge section showing content of radiolarian fossils (circular white spots). (b) Bioclast-peloidal limestone of same formation and the same section. (c) Alveolina ovulum (U. Paleocene), equatorial section, Qulqula Gorge section. (d) Nummulites perforatus (L. Eocene), axial section, Qulqula Gorge section.

The Paleocene–Eocene Red Bed Series of the second unconformity consists of more than 1500 m of red claystones, sandstones and conglomerates. This conglomerate contains about 10% limestone clasts (pebbles), which contain Late Paleocene and Eocene foraminifera. The depositional trend and direction of the sediment transport is tested by palaeocurrent analysis of imbricated pebbles, which indicate a SW sediment transport direction (Fig. 4). The Red Bed Series is interpreted to have been deposited in a foreland basin (K. H. Karim, unpub. Ph.D. thesis, Univ. Sulaimani, Iraq, 2004; S. T. M. Al-Barzinjy, unpub. Ph.D. thesis, Univ. Sulaimani, Iraq, 2005). Al-Barzinjy (unpub. Ph.D. thesis, Univ. Sulaimani, Iraq, 2005) considered that both the Red Bed Series and the Paleocene Kolosh Formation were deposited in a single large-scale foreland basin, which covered most of Iraq during Tertiary time. In the Chwarta, Mawat and Qandil areas (Kurdistan Region) (Figs 1, 2), the Red Bed Series is underlain by the Qulqula Radiolarian Formation (Lower Cretaceous) (Figs 3, 5); while to the SW, it is underlain by the Tanjero Formation (Maastrichtian). The contact between the Red Bed Series and the Tanjero Formation is gradational (conformable) (Lawa, Al Karadakhi & Ismail, Reference Lawa, Al Karadakhi and Ismail1998). However, Karim (unpub. Ph.D. thesis, Univ. Sulaimani, Iraq, 2004) described this contact to be gradational in some places and sharp (unconformable) in others. Thus, the contact is gradational at an outcrop SE of Mawat (35°45′47.5″N, 45°29′50.5″E), whereas the contact is unconformable 3 km south of Chwarta (Fig. 1b).

Field observations, in the present study, showed that an angular unconformity between the formations is present at two localities. The first (36°20′35″N, 45°10′11″E) is at Qulqula Gorge near Qandil Mountain, about 20km NE of Raniya (Figs 7, 8). The second location (35°55′10″N, 45°34′36″E) is 15 km NE of Chwarta and 3 km north of Basine (Figs 5, 9, 10, 11). The difference in bedding dip between the Qulqula Radiolarian Formation and Red Bed Series ranges between 17° and 21° at the two localities (Figs 5, 7, 8). At both localities, the Upper Cretaceous Kometan, Shiranish and Tanjero formations are entirely absent.

Figure 7.

Outcrop photographs showing the angular unconformity between the limestones (a) and cherts (b) of the Qulqula Radiolarian Formation and the overlying conglomerates of the Red Bed Series. Location: left side of Qulqula Gorge, Kurdistan Region, Iraq.

Figure 8.

Schematic cross-section of the Qulqula Gorge along the left side of Marado valley, showing the angular unconformity between the Lower Cretaceous Qulqula Formation and the overlying Eocene Red Beds Series. See location of A–A′ in Figure 1.

Figure 9.

Field photograph of the angular unconformity between the Qulqula Radiolarian Formation and the overlying Red Bed Series in the Basine area (location in Fig. 2).

Figure 10.

Geological cross-section of the Basine area, passing through Suradiza and Dere villages showing the angular unconformity.

Figure 11.

General geological map of northeastern Iraq showing outcrops of the Campanian–Maastrichtian units (Shiranish and Tanjero formations) (modified from Sissakian, Reference Sissakian2000).

In most places, the unconformity surface is covered, but where observable, it is sharp and more or less undulating (Fig. 5). Efforts to observe or photograph a clear surface were not successful owing to the relative softness or friability of the stratigraphic units that bound the surface as mentioned above. The Qulqula Radiolarian Formation contains some siliceous limestone, which is resistant to karstification; therefore, signs of this process are not observed.

3.b. The Campanian–Maastrichtian unconformity (CMU)

The Campanian Shiranish Formation consists of about 200 m of bluish-white marls and marly limestones. It consists of deep marine facies (hemipelagite), which contain abundant Late Cretaceous planktonic foraminifera (see Bellen et al. Reference Bellen, Dunnington, Wetzel and Morton1959). It does not contain any sedimentary structures except for bedding; therefore, the depositional trend cannot be known directly. But it can be estimated by palaeocurrent direction from the Tanjero Formation (see below). This is because the contact between the two formations is conformable in all sections (see Buday, Reference Buday1980; Jassim & Goff Reference Jassim and Goff2006). However, recently Karim & Surdashy (Reference Karim and Surdashy2005b) and Karim (Reference Karim2006a) concluded that the contact is conformable except for in a few sections in coastal areas of the Late Cretaceous Zagros foreland basin, which are unconformable. Two of these sections are the Do Awan and Suwais village sections where the CMU is located and studied.

The Maastrichtian Tanjero Formation consists of about 400 m of sandstones in distal areas to the south of the Sulaimani and Dokan areas (Bellen et al. Reference Bellen, Dunnington, Wetzel and Morton1959; Buday, Reference Buday1980; Jassim & Goff, Reference Jassim and Goff2006), while it consists of about 500 m of indurate conglomerates in more proximal areas around the Chwarta, Mawat and Qandil areas (Fig. 1b) (Karim, Reference Karim2006a, unpub. Ph.D. thesis, Univ. Sulaimani, Iraq, 2004). In these areas, the conglomerate was tested by palaeocurrent analysis of the imbricated pebbles, and the palaeocurrent direction was found to be towards the SW.

Field observations show that this unconformity is more subtle than the Lower Cretaceous–Tertiary unconformity, and there is an angular difference of only 3–6° in bedding attitude of the over- and underlying formations. The unconformity surface is sharp and undulating (Fig. 12) on which there are no signs of tectonic polishing, brecciation, striation and mineralization. Based on these observations, the angular relationship is interpreted to be most probably a sedimentary rather than tectonic (i.e. thrusting) feature.

Figure 12.

Field photographs of the angular unconformity between the conglomerates of the Maastrichtian Tanjero Formation and the hemipelagites of the Campanian Shiranish Formation at (a) Qandil area (36°20′31.4″N, 45°07′17.6″E) (directly to the east of Suwais village); (b) at Do Awan (35°59′4.68″N, 45°24′36″E), with line interpretations.

The angular unconformity was observed at two localities close to the Lower Cretaceous–Tertiary unconformity. The first locality was near Do Awan, about 10 km NW of Mawat (35°59′4.68″N, 45°24′36″E). The second was a few kilometres north of Suwais (36°20′31.4″N, 45°07′17.6″E) (Fig. 1b) and about 30 km NE of Ranyia. Across this angular unconformity, about 50 m of the Shiranish Formation is missing at the unconformable contact. The estimation of this removed thickness is based on the fact that the thickness of the Shiranish Formation is about 200 m, while its thickness, in the mentioned locality, is about 150 m.

4.a. Tectonic significance of the CTU

The stratigraphic position of the CTU is interpreted to indicate a major tectonic phase in the Imbricate Zone. First, the absence of Upper Cretaceous rock units (Kometan, Shiranish and Tanjero formations) across the CTU is interpreted to indicate uplift and erosion during ophiolite–radiolarite obduction. This may be interpreted in two ways: if deformation had taken place in post-Cenomanian–Turonian time, then the Upper Cretaceous rock units, i.e. Kometan, Shiranish and Tanjero formations, had no chance to form. In other words, their absence is not attributed to erosion, but to their non-deposition. However, if deformation had occurred after the Upper Cretaceous rocks were deposited, i.e. Kometan, Shiranish and Tanjero formations, then the absence of these units is owing to their erosion earliest in Late Cretaceous time if not later. This erosion continued until Paleocene time when the area subsided to form a shallow basin where the conglomerates of the Paleocene–Eocene Red Bed Series were deposited. It can be seen that at this time the orogenic wedge (Qulqula Radiolarian Formation) is thrust over the Kometan Formation.

Facies distribution maps (Buday, Reference Buday1980) indicate that the palaeogeographic slope direction of the basin (in the study area) was towards the NE during Early Cretaceous to Middle Turonian times. This means that during these times the sediment transport was towards the NE. Later, during Coniacian and Santonian times, the palaeoslope direction was reversed (i.e. towards the SW) in the study area, which is also indicated by sedimentary structures such as flute casts and cross-bedding (Karim & Surdashy, Reference Karim and Surdashy2005a). This reversal corresponds approximately to the period of ophiolite obduction in Oman (Coleman, Reference Coleman1981), while Karim & Surdashy (Reference Karim and Surdashy2005b) interpreted this reversal to indicate the continental collision of the Iranian and Arabian plates. According to Karim (Reference Karim2003), prior to collision, the Qulqula Radiolarian Formation, as a thick sequence of bedded cherts and calcareous shales, transformed into an accretionary prism in the trench that was associated with the convergence of the two plates. During later collision, these rocks were intensely deformed and uplifted and thrust (as an orogenic wedge) onto the NE margin of the Arabian plate. In front of this wedge, a large-scale foreland basin developed on the Arabian plate margin where the Shiranish, Tanjero and Kolosh formations (in addition to the Red Bed Series) were deposited (Karim & Surdashy, Reference Karim and Surdashy2005a, Reference Karim and Surdashyb, Reference Karim and Surdashy2006).

Such an unconformity is reported from the Iranian Zagros as well. Farzadi (Reference Farzadi2006c) recognized channel-like features in the uppermost part of the Sarvak Formation in western Iran. Farzadi (Reference Farzadi2006b) suggested that this subtle channel-like feature near the top of the Sarvak Formation on the NE flank of the Parsi anticline probably indicates the Turonian unconformity. These channels are located close to the Kuh-e Bangestan anticline where the Turonian unconformity is well developed. Homke et al. (Reference Homke, Vergés, Serra-Kiel, Bernaola, Sharp, Garcés, Montero-Verdú, Karpuz and Goodarzi2009) showed that at the base of the Paleocene (within the Gurpi Formation) there lacks NP1 and NP2 zones, implying a hiatus of ~2 Ma at c. 65.5 Ma. They inferred that this hiatus corresponds to an early folding phase near the Cretaceous–Paleocene boundary.

The difference in bedding dip of the Qulqula Radiolarian Formation relative to the Red Bed Series, and the absence of Upper Cretaceous rocks at the CTU, indicate a significant phase of orogenic deformation. Uplift and erosion of Lower Cretaceous (and/or younger) units may have occurred at any time between their deposition and the deposition of the Red Bed Series in Paleocene–Eocene time. If it had occurred in Mid to Late Cretaceous time, the unconformity may be attributed to the obduction phase of the orogeny. However, if it had occurred during Late Cretaceous–Early Tertiary time, it is likely to be owing to the early stages of continental collision rather than obduction since the latter could not have occurred in the northwestern segment of the Zagros later than that in the southeastern extension of the belt (i.e. Oman ophiolites).

Sometime between Mid Cretaceous and Paleocene time, owing either to obduction in Mid–Late Cretaceous time or to collision (Oligo-Miocene), deformation was transferred southwestward, resulting in a series of structures interpreted as fault-related folds upon which the Red Bed Series was unconformably deposited. The fact that the Red Bed Series is nearly flat-lying in these two areas indicates that little post-depositional deformation took place in this part of the Zagros. This in turn means that the deformation front had propagated further to the SW, and new structures were forming towards the distal area of the foreland after the deposition of the Red Bed Series in the study area.

These observations are not inconsistent with previous estimates of the timing of obduction of the oceanic crust on the Arabian continental crust (e.g. Buday, Reference Buday1980; Buday & Jassim, Reference Buday and Jassim1987; Numan, Reference Numan1997; Jassim & Goff, Reference Jassim and Goff2006). However, they suggest that orogenic deformation in the Imbricate Zone consisted of separate phases which began in Santonian time (obduction of ophiolite) with southwestward advance of the orogenic front in Paleocene–Eocene time, and continued with other phases in Oligo-Miocene time (continental collision).

The presence of these unconformities has significant tectonic implications. In addition to the presence of many ophiolite bodies in the east and west of the study area, clasts of ophiolite (gabbro and peridotite) are present in the conglomerates and sandstones of the Upper Cretaceous Tanjero Formation in many places (Karim, Reference Karim2006b, unpub. Ph.D. thesis, Univ. Sulaimani, Iraq, 2004). It is therefore assumed that the CTU formed after ophiolite obduction.

4.b. Tectonic significance of the CMU

As indicated by the studies of Al-Barzinjy (unpub. Ph.D. thesis, Univ. Sulaimani, Iraq, 2005), Karim (Reference Karim2006a) and Karim & Surdashy (Reference Karim and Surdashy2005a, Reference Karim and Surdashyb, Reference Karim and Surdashy2006), the areas where the unconformities now occur were coastal areas of the Late Cretaceous (?) and Paleocene foreland basins. The deposition of 500 m of boulder and gravel conglomerates of the Tanjero Formation indicates a high-relief source area (a mountain belt), which underwent intense erosion and transportation. It is possible that the areas where the CMU is located were the first areas that were bulged up and eroded as deformation reached the coastal area of the foreland basin during the ophiolite obduction.

Erosion of part of the Shiranish Formation across the CMU is interpreted to reflect the advance of the deformation front southwestward, beyond the location of the CTU. This is because the former unconformity (CMU) is located a few kilometres to the south of the latter one (CTU). Now, this distance is short, but it is possible that originally the distance was more during Late Cretaceous time and shortened by folding and thrusting.

The unconformity between the Shiranish and Tanjero formations is characterized by the presence, in some places, of up to 500 m of conglomerate resting on the lower part of the Shiranish Formation (Fig. 8). It seems that the Shiranish Formation, in the coastal area of the Campanian foreland basin, was tilted gently and partially eroded before being unconformably overlain by the conglomerates and sandstones of the Tanjero Formation during Maastrichtian time. The CTU occurs along the strike of (and on) closely spaced anticlines (Fig. 13) of the Qulqula Radiolarian Formation, which may have formed in an accretionary prism during Early–Mid Cretaceous time.

Figure 13.

Bedded cherts and siliceous shale of the Qulqula Radiolarian Formation showing medium-scale folds which are exposed along the side road-cut at 4 km to the east of Chwarta town inside the Imbricate Zone.

The presence of these two unconformities at different stratigraphic levels and times could be used to argue that deformation along the c. 2000 km long ZFTB was variable both in style and timing. It is argued here that it is likely that ophiolite obduction, propagation of the deformation front, nature, temporal and spatial location and distribution of the foreland basins and collision were variable along the belt. However, it is worth underlining here that if obduction (or continental colliding) had started simultaneously along the entire length of the suture, the only way to fit a scissoring-style (protractor-style) closure for the Neo-Tethys would be to assume that the obducted slab in the NW was shorter, and that collision in the NW occurred earlier than in the SE where obduction lasted longer since the obducted slab was longer. This can partly be supported by the far-reaching obduction of the Semail ophiolite in the Oman Mountains, which was emplaced onto the Arabian continental margin during Late Cretaceous time (Searle & Cox, Reference Searle and Cox1999). However, the observations in this paper do not provide any evidence about the size of the ophiolite bodies in the NW Zagros, nor how far they have been emplaced.

An anticlockwise rotation of Arabia (and Africa) relative to Eurasia has been suggested by many workers (Kahle & Muller, Reference Kahle and Mueller1998; Sella, Dixon & Mao, Reference Sella, Dixon and Mao2002; Vernant et al. Reference Vernant, Nilforoushan, Hatzfeld, Abbassi, Vigny, Masson, Nankali, Martinod, Ashtiani, Bayer, Tavakoli and Chéry2004). And as stated above, based on this rotation, a scissoring-style closure of the Neo-Tethys Ocean has been proposed with the collision starting in the NW and propagating southeastward (Dercourt et al. Reference Dercourt, Zonenshain, Ricou, Kazmin, Le Pichon, Knipper, Grandjacquet, Sbortshikov, Geyssant, Lepvrier, Pechersky, Boulin, Sibuet, Savostin, Sorokhtin, Westphal, Bazhenov, Lauerm and Biju-Duval1986). However, Peynircioglu et al. (Reference Peynircioglu, Langereis, Kaymakci and Van Hinsbergen2010) used palaeomagnetic data to argue that observed rotation in the western part and northeastern edges of the Arabian plate is owing to local fault block rotations and does not represent the rotation of the entire plate. They argued further that large regions in the western part and northeastern edges of the Arabian plate (e.g. Urfa, Kocak and Mardin regions) show no rotations since Eocene time, and were likely part of the rigid Arabian plate. Whether Arabia was rotating until Eocene time or not is beyond the scope of this article. However, our field data indicate that deformation due to ophiolite emplacement has been variable in the northwestern segment of the ZFTB. This variation in propagation of the deformation front is likely to have occurred along this c. 2000 km long belt. Any rotation of Arabia relative to Eurasia would have likely added to this variation in deformation along the belt.