At least 20 locations were investigated during the whole field trip, of which 15 outcrops (4 in EB area, 11 in UK area) were observed and described in detail (Figure 2). For each outcrop, orientation data, as well as photos and some quick sketches, for all the strata, sedimentary structures, fault planes, joints, and fractures were collected. At each outcrop, we sampled different types of rocks and recorded their GPS coordinates, occurrences, and contact relationships with their country rocks.

About 120 samples were collected in the field and then examined in the laboratory. Samples were named after outcrop and sampling number. For example, EB3-2 is a sample that was collected at the second sampling location at the EB-3 outcrop.

Around 100 thin sections from a total of 80 hand specimens were studied. Petrology and mineralogy were systemically described in terms of rock types, micro-textures, structures, mineral assemblages, mineral species, mineral contents, and mineral alternation.

Zircon geochronology study was conducted on five representative samples from five outcrops, including zircon internal structures, Th and U contents, rare earth elements (REE), and U-Pb dating on more than 200 zircons. Cathodoluminescence (CL) images were taken on these zircons before analysis to reveal the internal structure and determine the analytical positions. All these obtained data were combined to determine the origin of zircons.

The single-zircon U-Pb dating method was utilized to estimate the ages of these rocks. Sample preparation and age dating were conducted in the State Key Laboratory of Continental Dynamics at Northwest University, China. The equipment used was a Laser Ablation Inductively Coupled Plasma Mass Spectrometry (LA-ICP-MS) produced by Agilent with a GeoLas200 M laser ablation system produced by MicroLas. The laser ablation system consisted of a Compex102 ArFexcimer laser (operated at a wavelength of 193 nm) from Lambda Physik, with a MicroLas optical system.

Single spot laser ablation was used for sampling, with a spot diameter of 30 μm and a frequency of 10 Hz. The carrier gas for the ablated materials was high-purity argon and helium. Standard zircon 91,500 and synthetic glass NIST610 were used as standards, with one analysis on zircon 91,500 every five spots and one analysis on NIST610 every 10 spots. The background acquisition time for each spot was 30 s and the signal acquisition time was 40 s. GLITTER software (ver. 4.0, Macquarie University) was used for data processing. The zircon standard 91,500 was adopted as the external standard for isotopic ratio fractionation correction during age calculation.

The outcrops can be roughly divided into two types, covered, and uncovered. The former type of outcrops is usually covered by plants, sparse vegetation (Figure 3A), or overlying sedimentary strata such as Quaternary flood deposits (Figures 3B,C). Uncovered outcrops have no vegetation cover, and the rocks are completely and directly exposed, appearing as a single giant granite body (Figure 3D) or multiple continuous low-amplitude hills (Figures 3E,F). Many outcrops have clear broken zones, fractured belts, or joint systems (Figures 3E,F).

At least dozens of rock types, either igneous or metamorphic, were identified in the field at different outcrops. The igneous rocks include granite, syenite, diorite, and monzonite and they are relatively fresh (Figures 4A–C,F). The metamorphic rocks contain metasandstone, quartzite, gneiss, and mylonite (Figures 4D,E). Most outcrops are dominated by igneous rocks (85 vol%), followed by gneiss and quartzite (around 10%), and minor mylonite (less than 5%).

In this study, the UK-7 outcrop was selected to represent the rock types and their occurrences of the studied outcrops. The rocks at the UK-7 outcrop are directly exposed to the surface (Figure 5A) and are typically granite and syenite (Figure 5A). Spheroidal weathering is widespread, with hundreds of ball-shaped granites distributed randomly. Fractures often transect veins, joints, and other fractures (Figures 5A,B). The rocks are usually flesh red, dense, and massive, with a medium-coarse-grained texture. Aggregates of K- feldspar or hornblende crystals are visible in hand specimens (Figure 5C).

Sample UK7-1 is medium-to-coarse-grained syenite, with a subhedral granular texture and grain sizes from 1.0 to 4.0 mm. It is composed of alkali feldspar, plagioclase, and quartz. The dominant alkali feldspar (69%) occurs in both subhedral-tabular and wide-tabular shapes, within which plagioclase commonly appears as metasomatic relicts, suggesting intensive alkalization (Figures 6A,B). The plagioclase (10%) is tabular, with some polysynthetic twinning. The hornblende (20%), sometimes occurring as aggregates, are subhedral, euhedral granular, and columnar grains, with good cleavages, partial chlorination, and magnetite precipitation. The quartz, with varied contents, is xenomorphic-granular and infilled between tabular feldspar crystals (Figures 6C,D). Except for some metasomatic minerals, almost all the mineral crystals are in good shape with sharp boundaries, indicating a weak degree of hydrothermal alteration.

We studied all the collected samples from the 15 outcrops and found that they are more commonly composed of granite, syenite, and diorite, with a minor volume of metasandstone, mylonite, gneiss, and quartzite (Table 1). They have similar mineral assemblages of alkali feldspar, plagioclase, quartz, and amphibole (or hornblende). In igneous rocks, alkali feldspar and plagioclase are principal minerals, and their contents could reach as high as 75%. Residual plagioclases are usually developed within alkali feldspar, indicating strong alkalization. The contents of quartz vary from 10% to 25%. And the hornblende is usually with magnetite. These minerals seem fresh without significant alternation, but the micro-fractures and veins are usually developed among them. Metamorphic rocks have various mineral assemblages, and serious mineral alteration. Metasandstone and quartzite are mainly composed of quartz and lithic fragments. Similarly, the mylonite is also composed of the quartz-dominated matrix, showing strong recrystallization and orientation, while gneiss is constituted by biotite, feldspar, and quartz. Most of these minerals are underwent partly to completely alternation (Table 1).

Zircon grains from the sample UK5-2 display low Th (29–131 ppm; mean = 61 ppm) and U contents (76–204 ppm, mean = 131 ppm), with a limited Th/U ratio range of 0.3–0.5 (mean = 0.5). In sample UK6-1, zircons show variable Th (24–4,968 ppm; mean = 939 ppm) and U contents (385–1,460 ppm, mean = 835 ppm), with largely variable Th/U ratios of 0.1–3.4 (mean = 0.9). The Th and U contents of zircon grains from sample UK6-2 are much higher with values of 257–3,281 ppm (mean = 817) and 342–1,688 ppm (mean = 1,076 ppm), respectively. The Th/U ratios of zircons from this sample (mean = 0.8) are mostly higher than 0.5, similar to those of typical igneous zircons. Zircon grains from sample UK11-3 show extremely variable Th (54–5,399 ppm; mean = 1796 ppm) and U contents (106–1,277 ppm; mean = 501 ppm), resulting in largely variable Th/U ratios of 0.5–8.5. All these zircon Th-U data indicate the zircons from these samples are of magmatic origin.

The zircon REE pattern diagrams were also generated. The chondrite-normalized REE patterns of zircons from sample UK5-2 display enrichment of heavy REEs relative to light REEs, with significant positive Ce and negative Eu anomalies (Figure 10A). Comparatively, in sample UK6-1, the rising slopes from LREE to HREE are much weaker. However, the positive Ce and negative Eu anomalies remain obvious (Figure 10B). For sample UK 6-2, the zircons also display a steeply rising slope but relatively weak positive Ce anomalies and negative Eu anomalies among other samples (Figure 10C). Sample UK 11-3 also show significant positive Ce and negative Eu anomalies (Figure 10D). All these characteristics of zircon trace elements indicate the zircons are of magmatic origin (Hoskin and Schaltegger, 2003).

Using the same dating method and following the same procedure as those of sample UK7-1, we tested another 144 individual zircon grains from samples UK5-2, UK 6-1, UK6-2, and UK11-3. The sample ages were obtained either from the Concordia graph, weighted mean age, or both. Sample UK5-2 has a Concordia age of 217.8 ± 1.7 Ma, consistent with the weighted mean age of 218 ± 2 Ma (Figures 11A,B). The Concordia diagram of UK 6-1 revealed an age of 224.8 ± 3 Ma (Figure 11C), while the sample UK6-2 shows an intercept age at 207.2 ± 5.9 Ma (Figure 11D). The ages from sample UK 11-3 are a little dispersed and give a mean age of 195 ± 6 Ma (Figures 11E,F). All these ages were summarized in Table 2.

Before the basin-rifting stage, from Precambrian to Late Jurassic, the CARS rarely experienced large-scale intraplate magmatic events, so the Precambrian granite intrusions are very limited in the basin basement. However, since the beginning of basin-rifting in the Cretaceous, the CARS have undergone frequent and intensive magmatic events due to intraplate extension. At least three regional magmatic events took place, i.e., the Early Cretaceous (147–106 Ma), Late Cretaceous (95–75 Ma), and Paleogene (66–52 Ma) events. The main igneous rocks are basalt and diorite, and they are common in the Bongor, Termit, and Doba Basins (Table 3; Wilson et al., 1992; Guiraud and Maurin, 1992, Guiraud et al., 1992; Janssen et al., 1995; Maluski et al., 1995; Coulon et al., 1996; Kamdem et al., 2002; Lu et al., 2009).

The identification of the Late Triassic-Early Jurassic intrusions in addition to the Later Cretaceous intrusions indicates that some local thermal or magmatic events also occurred in the Early Mesozoic. These intrusions resulted in changes in rock association and basement morphology. These Early Mesozoic intrusions are therefore a new member to the basement of the Muglad Basin in addition to the Precambrian rocks.

As mentioned above, the lithology of the basement has fundamental impacts on the development of an unconventional reservoir (Hu et al., 1981; Dou et al., 2015; 2018). The primary minerals of the Late Triassic-Early Jurassic igneous rock in the Muglad Basin are feldspars (both potassic and plagioclase) and quartz, with various amounts of dark minerals (mainly hornblende and biotite). These mineralogical compositions of felsic rocks with inherent susceptibility to alteration and diagenesis makes them better candidates for basement reservoirs. Adjacent to these outcrops, drilled wells and three-dimension seismic data (3D seismic) in Block-6 have revealed a double-layer structure for the basement. The shallow layer is the weathered crust, and the middle-deep layer is mainly fractured zones (Zhao et al., 2020). The crust is mainly constituted by these granitoids and has very good porosity and permeability. Its thickness varies from area to area and is usually less than 50 m (Zhao et al., 2020). Similarly, in the Bongor basin (Chad), the thickness of the weathered crust could reach up to hundreds of meters with excellent porosity and permeability, and they are also the product of granitic basement weathering (Dou et al., 2015; 2018; Dou et al., 2018).

In addition, these intrusive rocks were formed as early as in the Early Mesozoic. After that, they experienced regional uplifting, mechanical breaking, physical weathering, and leaching for as long as 65 Ma – 95 Ma during and after the basin formation. This long-time duration of weathering and leaching would significantly improve their porous and permeable properties and form weathering-leaching zone or fissures-fracture zone of considerable scale and thickness. The exploration practice in the Bongor basin (Chad) revealed that the weathered crust in high buried hills is much thicker than that in low buried hills. This is mainly because the weathering time of high buried hills is longer than that of low buried hills before they were buried, indicating the impact of weathering time on basement reservoir development (Dou et al., 2018).

Finally, the outcrops are adjacent to the CARS, with strong tectonic activities and frequent changes in the tectonic stress regime, which provide external conditions for the formation of fractures and the improvement of reservoir properties (Genik, 1992; Fairhead et al., 2013). The geological field investigations found multi-stage fractures in these outcrops. Meanwhile, millimeter-sized fractures can also be seen in thin sections (Zhao et al., 2020).

Although the Mesozoic intrusions are first discovered and reported in this study, we do not know their regional distribution due to the limited study area compared to the Muglad Basin. Based on our field investigation and laboratory analysis, we intend to believe that these intrusive bodies are sills exposed at these outcrops. They may come from the same magma chamber in deep layer because they all share similar mineral assemblages, ages, and chemical compositions. It is anticipated that these intrusions may be developed in a much larger area, and more and more discoveries will be made in the future. To clarify their types and scales, additional geological surveys and geophysical work should be conducted in the future. Then their representativeness should be further demonstrated across the basin.

Besides, magmatism and its implication for tectonic evolution would be one major part of the next-stage research. Because magmatism is probably related to regional rifting. The volcanic activities in Chad basins were reported as the products of the intraplate extensional stress regime, corresponding to the tectonic setting of the whole West and the Central during the Cretaceous (Lu et al., 2009). A comprehensive study of these topics will help understand the scientific significance of this episode of magmatism and shed light on future regional geological surveys and oil and gas exploration.