The Early Permian tectonic history of eastern Australia led to the formation of several orogenic curvatures termed the New England oroclines. How these oroclines formed is a controversial issue that is crucial for understanding the paleo-Pacific subduction dynamics at the Gondwanan margin and the formation of curved orogenic belts in general. Here we present new constraints on the role of vertical-axis block rotations in the New England oroclines using paleocurrent indicators from the core of the oroclinal structure (the Nambucca block). Focusing on the lower sedimentary succession within the Nambucca block (Kempsey beds), we recognize two facies associations. Facies association A comprises conglomerate and gravelly sandstone with minor sandstone, collectively interpreted as the deposits of coastal to subaqueous marine fans. Facies association B is made of heterolithic intervals of sandstone and mudrock that are interpreted as the products of deposition on a marine continental slope. Younging directions suggest that facies association A represents the basal part of the succession that is overlain by the more heterolithic association. The paleogeographic position of the Nambucca block, in conjunction with its stratigraphy and geochronological provenance, suggests that it formed as part of a large, deep-marine backarc basin. Paleocurrent and paleoslope directions are north to northeast, inconsistent with the present understanding of the Permian paleogeography that involved an approximately north-south–oriented continental margin (in present coordinates) and an eastward-deepening marine surface. This supports previous paleomagnetic interpretations of counterclockwise rotations of adjacent blocks. In conjunction with recently published structural, paleomagnetic, and geochronological constraints, our data suggest that counterclockwise rotations occurred between 285 and 275 Ma in the course of the formation of the southern segment of the New England oroclines (Manning orocline). The rotation incorporated both continental and marine plate margin segments of eastern Gondwana, thereby deforming the deep backarc basin that is partially represented by the Nambucca block. Our data thus provide constraints both on the kinematics and on the timing of the much-debated southern segment of the New England oroclines.

The New England Orogen is the youngest and easternmost component of the Tasmanides in eastern Australia (Fig. 1A; Glen, 2005; Cawood, 2005). The structural grain of the southern New England Orogen constitutes a set of orogenic curvatures (oroclines) that includes the Texas, Coffs Harbour, Manning, and Nambucca oroclines (Figs. 1B–1E). The major evidence for the oroclinal structure includes (1) a curved aeromagnetic anomaly (Fig. 1C); (2) a curved distribution of Devonian–Carboniferous forearc and accretionary complex rock units (Fig. 1B) (Korsch and Harrington, 1987; Rosenbaum, 2012); (3) a contorted belt of Early Permian S-type granitoids (Fig. 1D) (Rosenbaum et al., 2012); and (4) curved structural and magnetic fabrics (Fig. 1E) (Korsch and Harrington, 1987; Aubourg et al., 2004; Li et al., 2012; Li and Rosenbaum, 2014; Mochales et al., 2014). The existence of the orogenic curvature is well established; however, interpretations of paleomagnetic data are more ambiguous (Schmidt et al., 1994; Geeve et al., 2002; Cawood et al., 2011b; Shaanan et al., 2015a; Pisarevsky et al., 2016), thus leading to uncertainties in regard to the process of oroclinal bending. The timing of oroclinal bending is constrained to the Early Permian (299–272 Ma; Shaanan et al., 2015a, and references therein), and based on the available structural, geochronological, and paleomagnetic evidence, it has recently been hypothesized that, similar to modern examples of oroclinal bending (e.g., Rosenbaum, 2014), the New England oroclines formed in a backarc region associated with a retreating subduction zone (Shaanan et al., 2015b).

The Manning orocline, which is located at the southernmost part of the oroclinal structure, is particularly controversial (e.g., Lennox et al., 2013; Li and Rosenbaum, 2014, 2015; Offler et al., 2015; White et al., 2016). This orogenic curvature appears to be structurally more complex (Li and Rosenbaum, 2014; White et al., 2016) and its recognition is therefore somewhat more ambiguous (Figs. 1C, 1E). Nonetheless, a number of independent lines of evidence, such as the curved shape of the Early Permian granitoid belt (Fig. 1D; Rosenbaum et al., 2012), support the existence of the orocline. The role of vertical-axis block rotations, however, is yet to be tested.

Paleomagnetic results from Devonian–Carboniferous forearc blocks (Rouchel, Gresford, and Myall; Fig. 1B) are generally consistent with counterclockwise block rotations by as much as 120° (Geeve et al., 2002) prior to the Middle Permian (Shaanan et al., 2015a). However, the link between apparent block rotations and oroclinal bending remains inconclusive because of the possibility that angular relations of paleomagnetic vectors represent large displacements across medium to high latitude rather than vertical-axis block rotations (Cawood et al., 2011b). To address this shortcoming, we have taken an alternative approach that focuses on sedimentological data from Lower Permian rocks. The data inform us on paleocurrent directions and the paleoslope of the continental margin. Combined with a new synthesis of geochronological provenance data, and available paleomagnetic and structural information, these results are used to infer that vertical-axis block rotation may have played an important role in the formation of the Manning orocline.

The New England Orogen (Fig. 1) is predominantly composed of Devonian–Carboniferous subduction-related rocks, which are in part intruded and/or overlain by Permian–Triassic magmatic and clastic sedimentary rocks (Leitch, 1974, 1975a). Lower Permian sedimentary successions are mainly found in the Sydney-Gunnedah-Bowen basin system (Fig. 1A), but also in outliers throughout the southern New England Orogen (Figs. 1B and 2) (Roberts and Engel, 1987; Korsch et al., 2009). The origin of these sedimentary basins has been attributed to a phase of crustal extension, which occurred throughout the margin of eastern Gondwana in the Early Permian (Veevers et al., 1994) and was possibly driven by trench retreat (Jenkins et al., 2002; Shaanan et al., 2015b).

Among the lower Permian outliers that are within the oroclinal structure (Fig. 1B) are (1) several discrete outcrop belts in the Texas region (Silver Spur, Alum Rock, Terrica, Bondonga, Pikedale, and Ashford; Donchak et al., 2007; Campbell et al., 2015), (2) a structurally complex zone in the area of the Manning orocline (Manning Basin; White et al., 2016), (3) the Cranky Corner outlier in the Gresford block (Fig. 1B; Facer and Foster, 2003), and (4) the more extensive outcrop belts of the Nambucca (Leitch, 1978, 1988; Johnston et al., 2002; Shaanan et al., 2014, 2015b) and Dyamberin blocks (Korsch, 1978; Shaanan and Rosenbaum, 2016) (Fig. 1F). Detrital zircon provenance studies of the Nambucca and Dyamberin blocks indicate that both successions are correlative and were likely deposited in one basin (Shaanan and Rosenbaum, 2016). Moreover, early Paleozoic and Precambrian zircon U-Pb age populations in the succession of the Nambucca and Dyamberin blocks indicate that detritus was transported and recycled from cratonic Gondwana, thus suggesting that sedimentation occurred in a backarc environment (Shaanan et al., 2015b; Shaanan and Rosenbaum, 2016).

The Lower Permian succession of the Nambucca and Dyamberin blocks is fault bound against older Devonian–Carboniferous accretionary complex and forearc basin units (Tablelands Complex and Hastings block, respectively; Fig. 1F). Rocks of the Nambucca block vary from moderately deformed in the south to intensely deformed further north and west (with multiple, superimposed structural fabrics including slaty cleavage) and are in part regionally metamorphosed (Leitch, 1975b, 1978, 1988; Offler and Brime, 1994; Shaanan et al., 2014). Leitch (1975b, 1978) estimated a total thickness of at least 5 km for the succession, and Shaanan et al. (2014) showed a broad younging to the northwest and a stratigraphic tendency associated with crude upward fining. The Kempsey beds, which occur in the southeastern part of the Nambucca block (Fig. 1F), are characterized by well-preserved conglomerate and sandstone-dominated rocks (Brunker et al., 1970; Shaanan et al., 2014). The rocks of the Kempsey beds are significantly less deformed and less metamorphosed than the metapelitic rocks in the northwestern part of the Nambucca block (Nambuccca slate), although the deformation history of both parts of the succession appears to be consistent (Shaanan et al., 2014).

Earlier studies have recognized four phases of folding and associated deformation in the Nambucca block (Leitch, 1978; Johnston et al., 2002; Shaanan et al., 2014), the second of which was dated as 275–265 Ma by 40Ar/39Ar geochronology on muscovite (Shaanan et al., 2014). Both the first and second phases of deformation have been attributed to the process of oroclinal bending (Offler and Foster, 2008; Rosenbaum et al., 2012; Shaanan et al., 2014). Furthermore, in Shaanan et al. (2015b) it was proposed, based on the large fraction of craton-derived Precambrian detrital zircons in the succession, that deposition occurred in a backarc basin that developed in response to trench retreat, and that the same driving mechanism was responsible for the formation of the New England oroclines. Shaanan et al. (2015b) showed that the youngest detrital zircon age populations in the Nambucca block (including the Kempsey beds and the Nambucca slate) are 299 ± 1.4 (n = 22) and 285.5 ± 2.3 (n = 7) Ma. These ages are consistent with smaller detrital zircon U-Pb age populations from the same rocks of 297 ± 6 (n = 5) and 293 ± 7 (n = 7) Ma (Adams et al., 2013) and with a U-Pb SHRIMP (sensitive high-resolution ion microprobe) age of 292.6 ± 2.0 Ma from a dacite at the base of the succession (Halls Peak Volcanics; Cawood et al., 2011a).

To date, no stratigraphic or sedimentological analysis has been carried out on the Kempsey beds. In the following we provide a facies analysis of the putative lower part of the succession. We evaluate the geochronological provenance relationship of the Kempsey beds with the overlying strata of the Nambucca slate, and establish constraints on the depositional environment and sediment dispersal directions. We then place the succession in the regional context of Early Permian extension and orocline evolution.

In order to compare the geochronological provenance of the Kempsey beds and Nambucca slate, we compiled previously published data sets of detrital zircon U-Pb ages from the Nambucca block (Table 1; after Adams et al., 2013; Shaanan et al., 2015b). The relative probability and cumulative proportion curves of the data sets of the Kempsey beds (n = 179), and of the Nambucca slate (n = 398), predominantly overlap and consist of similar age populations (Fig. 3). The close resemblance implies that the Kempsey beds and the Nambucca slate, despite the clear variation in grain size and existence of several distinct drainage systems (Shaanan and Rosenbaum, 2016), shared similar provenance and received input of detritus from the same drainage systems. The consistent scatter of polymodal detritus in the Nambucca block (Shaanan et al., 2015b; Shaanan and Rosenbaum, 2016) suggests that the different drainage systems remained constant during the accumulation of the succession of the entire block.

Petrographic investigation of sandstones and conglomerates from the Kempsey beds reveals a mineralogically immature composition dominated by lithic fragments, with lesser quartz and minor feldspar (Fig. 4). Most samples are lithic arenites according to the classification of Dott (1964), or feldspathic litharenites and litharenites according to Folk (1980). A significant component of sandstone grains is present in most samples (and of sandstone clasts in conglomerates), with lesser pyroclastic, crystalline igneous, and chert lithic grains (Fig. 4). The detrital composition suggests that the Kempsey beds formed by significant reworking of older sedimentary and igneous detritus. The interpretation of polycyclic sedimentation is consistent with previously suggested recycling of detritus from the Devonian–Carboniferous forearc units of the New England Orogen (Shaanan et al., 2015b). Samples plot in the recycled orogenic field according to the provenance classification of Dickinson et al. (1983).

The moderately deformed exposures of the Kempsey beds examined herein represent two distinct facies associations. Facies association A consists principally of conglomerates with subordinate gravelly sandstones, whereas facies association B comprises mainly sandstones with interbedded mudrocks (Table 2). All of the outcrop areas examined exposed one or other of the two facies associations. None of the exposures shows interbedding, or vertical juxtaposition, of the two facies associations. Accordingly, there is no direct means of establishing whether the two associations represent lateral facies variations, or an upward stratigraphic change from one association into the other. The establishment of general northwestward younging in coastal outcrops of the Nambucca block led Shaanan et al. (2014) to suggest that the Permian succession preserves a gross fining-upward character. Nonetheless, exposures of the two facies associations alternate along the coast (from south to north, Crescent Head, Hat Head [Korogoro], Smoky Cape, and Southwest Rocks; Fig. 1F). From regional structural orientations (Shaanan et al., 2014), it is clear that the coarser grained lithologies of the Kempsey beds as a whole pass northward and westward (stratigraphically upward) across the Nambucca block into more strongly deformed, mudrock-dominated lithologies of the Nambucca slate.

Facies Association A


Facies association A (Table 2; Fig. 5) comprises an array of three facies that are intimately interbedded with each other, ranging from coarse-grained sandstone (facies A1), through gravelly sandstone and matrix-supported conglomerate (facies A2), to clast-supported, pebble to boulder conglomerate (facies A3). The coarse fraction comprises typically subrounded to well-rounded clasts, as much as 40 cm in a-axis length, but typically <10 cm (Fig. 5C), and mostly extraformational with a local component of angular and complexly shaped, intraformational siltstone clasts (Fig. 5D). Dominant clast types are quartzite, sandstone, vein quartz, chert, and lesser felsic intrusive lithologies, granites, and volcanic rock types (including fine-grained tuff). Overall, the clast population is poorly sorted, but individual beds show moderate to good sorting (Fig. 5). A sand matrix is dominant in all association A facies.

Continuous vertical intervals of association A reach at least 50 m in thickness at Hat Head (Korogoro) (Fig. 5A). Facies A1 occurs both as the upper portions of sharply based, fining-upward units several meters thick (Fig. 5B) and as discrete intervals to at least 20 m thick. Trough cross-beds are locally visible in both this and facies A2 (Fig. 5E), both of which are typically planar stratified (Fig. 5F). Stratification is in all cases defined by grain-size differentiation among successive strata. Facies A3 comprises thick intervals of continuous, clast-supported, pebble- to boulder-grade conglomerate (Fig. 5B). Clast imbrication is common in this facies, particularly in outsized and platy clasts (Fig. 5C), and is both of a-axis and b-axis varieties. No trace or body fossils were encountered in association A facies during the course of this study. Paleocurrent directions measured from clast imbrication (n = 120) and cross-bedding (n = 8) indicate a strong northeast mode (Fig. 6).


The coarse-grained, stratified nature of much of facies association A suggests that it formed in the proximal reaches of depositional fans. A purely fluvial origin can be discounted because no channel forms were noted in the 10–50-m-high and laterally extensive cliff exposures. Nonetheless, it is unclear whether facies association A accumulated in a subaerial or subaqueous setting, because no trace or body fossils were found in these rocks. The lack of soft-sediment deformation and mudrock partings (together with the lack of fossils) within association A facies argues against deposition in substantial depths of standing water (cf. Nemec and Steel, 1984). The dominance of stratified beds over massive beds and a lack of any relationship between maximum clast size and bed thickness (Heward, 1978; Nemec and Steel, 1984) argue for deposition from turbulent, frictional flows rather than from debris flows. This interpretation is also supported by the common preservation of trough cross-bedding. The fact that the most abundantly preserved physical sedimentary structure is planar stratification, however, suggests that formative flows were powerful and often within the stability field of plane beds, and/or that they were carrying high concentrations of sand in traction and near-bed suspension, leading to suppression of bedforms (Sumner et al., 2008; Baas et al., 2011, 2016). Overall, the lithology, texture, fabric and preserved sedimentary structures suggest that facies association A was formed in shallow-water fans or fan deltas that issued northeastward into the formative basin.

Facies Association B


Facies association B (Table 2; Fig. 7) comprises an array of clastic lithologies ranging from interbedded sandstones and siltstones (facies B1), through discrete sharp-bound sandstones beds (facies B2), to amalgamated sandstone intervals (facies B3), intraformational siltstone clast breccias (facies B4), and a single occurrence of intraformational sandstone clast breccia (facies B5). Although some of the sandstones in this facies association are coarse to very coarse grained and locally contain small extraformational and intraformational gravel, there is no palpable overlap with facies of association A. Vertical intervals of up to at least 75 m are continuously exposed in coastal cliffs near Crescent Head (Fig. 1F). A graphic log of the best-exposed portion of the section at Crescent Head is presented in Figure 8.

Thinly interbedded sandstones and siltstones (facies B1) contain a variety of interlamination and soft-sediment deformation structures (Fig. 7B), and some are diversely bioturbated (Fig. 7C; Table 1). Sandstone beds (facies B2 and B3) are sharply bounded (Figs. 7A, 7B, 7F), and many are normally graded, with small gravel abundant in the basal portion of beds. The most common structures are planar and low-angle lamination locally with a convex-upward component (Fig. 7B), and scour surfaces are common within composite beds (Fig. 7D). Soft-sediment deformation structures are common, including small clastic intrusions, some of which are tilted in the direction of paleoflow (Fig. 7E). Bioturbation is sparse, and where present is composed of vertical traces (especially fugichnia; escape burrows) in the uppermost parts of beds. Paleocurrent indicators (n = 40) suggest northeastward sediment dispersal (Fig. 6).


The presence of trace fossils associated largely or exclusively with marine settings indicates that facies association B formed subaqueously in a marine environment. The absence of any wave- or combined flow-generated sedimentary structures from facies association B places the depositional environment below (storm) wave base. The trace fossil assemblage, most diversely preserved in facies B1, is interpreted as an expression of the Zoophycos Ichnofacies (MacEachern and Bann, 2008). The composition of the trace assemblage is also distinctively different from Permian trace fossil suites in nearshore marine successions of the Sydney Basin (Thomas et al., 2007; Bann et al., 2008; Rygel et al., 2008; Fielding et al., 2006, 2008) in preserving few if any vertical domicile traces and more extensive records of surface dwellers and infaunal miners. This also suggests that the depositional environment was relatively deep, in the range of 100–500 m water depth. The medium gray color of mudrocks in facies B1 suggests moderate organic carbon contents, which in turn may point to periodically dysoxic to anoxic seafloor conditions that led to restricted trace diversity, another characteristic of the Zoophycos Ichnofacies.

The dominance of planar and low-angle stratification in facies B2 and B3, with some convex-upward stratification and common scour surfaces, suggests that sand was deposited from turbulent (frictional) flows that were transcritical (within the stability fields of plane beds, antidunes, and even chutes and pools; Cartigny et al., 2014). Some flows may have been high-concentration flows transitional to debris flows on the basis of their virtually massive character with floating siltstone clasts (Mulder and Alexander, 2001). The presence of internal erosion surfaces and vertical changes in sedimentary structure within composite sandstone beds suggests sustained flows of temporally varying strength. The general absence of trace fossils from all but the very tops of some sandstone beds is suggestive of rapid rates of sand deposition from formative flows. The occurrence of locally derived (from the ragged clast shape) siltstone clast breccias (facies B4) attests to the erosional potential of these flows. Tilting of load structures, small sandstone dikes, and some trace fossil shafts in the direction of paleoflow (determined independently from physical sedimentary structures) suggests that deposition occurred on a northeastward slope.

The occurrence of facies B5 at Crescent Head (Fig. 7F) also indicates that the depositional surface was a slope, subject to failure with downslope transportation of (apparently) cohesive slabs of sandstone. In the absence of any evidence for cyclic wave loading of the depositional surface, the most likely triggers for such slope failures are oversteepening due to sediment buildup, and seismic shaking of the seafloor. Whatever the trigger, facies B5 indicates that the slope failed en masse, causing coherent slabs of sandstone to slide and perhaps locally be entrained into debris flows, and then accumulate a short distance downslope in an imbricate stack. Such failures are common on modern submarine slopes, where they may reach considerable size and are termed mass-transport complexes (e.g., Sawyer et al., 2007; Moscardelli and Wood, 2008; Khani and Back, 2012). The facies B5 occurrence at Crescent Head closely resembles a series of Middle Permian submarine debris flow deposits in the Moah Creek Beds of east-central Queensland, which were interpreted (Fielding et al., 1997) to record the onset of Middle Permian to Triassic multiphase contractional deformation.

The range of lithologies and the physical and biogenic sedimentary structures in facies association B are entirely consistent with deposition on a submarine slope, below storm wave base. The surface sloped north to northeastward, based on paleocurrent measurements including the attitudes of imbricated sandstone slabs in facies B5 (Fig. 6). This interpretation is consistent with what little has been published previously on the topic (Leitch, 1988).

Depositional Setting and Time-Space Distribution of Permian Strata

The surface exposures of the Kempsey beds disclose two different depositional environments: shallow fans or fan deltas (facies association A) and submarine slopes below storm wave base (facies association B). The lack of any information regarding the relative stratigraphic position of the two facies associations, and the lack of any interfingering or interbedding of the two facies associations, renders resolution of the overall depositional setting a challenging task. Nonetheless, it is clear that the two facies associations are intimately associated, given the alternation of one with the other in successive coastal outcrops and the consistency in sediment dispersal direction between them (Fig. 6). The local trend of bedding and the idea of a large-scale (megasequence) fining-upward trend in the Nambucca block suggest that facies association A underlies facies association B and that the succession records upward deepening of the depositional setting. The regional paleoslope in the study area was toward the northeast (current orientation), suggesting a large, deep-marine basin along the present-day coast and immediate offshore easternmost Australia, outboard of the other, more localized Early Permian structural outliers.

Detrital zircon (U-Pb) ages from the Nambucca block (n = 577; after Adams et al., 2013; Shaanan et al., 2015b), including two samples from the Kempsey beds (n = 179; after Shaanan et al., 2015b), indicate that sediment accumulation occurred in the Early Permian ca. 299–290 Ma, with a later phase of sediment accumulation ca. 285 Ma (Shaanan et al., 2015b). These sediments were subjected to deformation, with muscovite from the second foliation phase (S2) dated as 275–265 Ma by 40Ar/39Ar geochronology (Shaanan et al., 2014). The rocks therefore must have been deposited in the Early Permian. Similar depositional ages are recorded from other structural blocks within the southern New England Orogen (Fig. 9), including the Werrie syncline (Roberts et al., 2006), Manning Basin (White et al., 2016), Dyamberin block (Shaanan and Rosenbaum, 2016), and Texas region (Roberts et al., 1996; Campbell et al., 2015). Data from these areas appear to indicate two major periods of Early Permian sediment accumulation, the first from ca. 299 to 290 Ma, and the second from ca. 290 to 280 Ma. Although there is only modest separation between these modes, individual areas typically include radiogenic isotope ages from both periods and a clear time gap between (Fig. 9).

The earliest phase of sedimentation at 299–290 Ma (Asselian to late Sakmarian) is consistently represented in all samples from the Nambucca and Dyamberin blocks (Adams et al., 2013; Shaanan et al., 2015b; Shaanan and Rosenbaum, 2016; Fig. 9). The rocks, which occupy a zone in the center of the main embayment of the New England oroclines, differ both in lithology and in inferred depositional environment from rocks of comparable age that are preserved elsewhere around the trace of the New England oroclinal structure. Included in this latter category are rocks of the Texas region (Bondonga beds, Terrica beds, and Alum Rock Conglomerate, Fig. 1B; Roberts et al., 1996; Donchak et al., 2007; Campbell et al., 2015). The lower parts of all these Permian successions are dominated by diamictites, with associated sandstones and mudrocks. The successions are also correlative to some of the earliest formations preserved in the northern Sydney Basin (Lochinvar Formation; McClung, 1980) farther southwest (Figs. 1 and 2). Those formations preserve trace and body fossils suggesting shallow-marine depositional environments, and the presence of outsized clasts and other features that imply deposition under glacial influence (Fielding et al., 2008). The glaciogenic facies preserved in all these successions are considered part of the glacial epoch P1 of Fielding et al. (2008). These formations have all been interpreted to have formed during a period of Early Permian extension (Scheibner, 1973, 1993; Jenkins et al., 2002; Korsch et al., 2009; Shaanan et al., 2015b).

The second phase of sedimentation at 290–280 Ma (Artinskian) is recorded in detrital zircons from sedimentary successions in most outliers around the Texas region (Silver Spur beds, Pikedale beds, Terrica beds, and Bondonga beds; Campbell et al., 2015), detrital zircons from the Manning Basin (White et al., 2016), a U-Pb SHRIMP age from the Werrie syncline (Woodton Formation; Roberts et al., 2006), and detrital zircons from the Nambucca block (Shaanan et al., 2015b) and the Dyamberin block (Shaanan and Rosenbaum, 2016; Figs. 1 and 9). The rocks that represent this age range comprise both diamictites and other probable glaciogenic facies with interbedded sandstones and mudrocks of shallow-marine origin. They are coeval with successions in the Gunnedah Basin (Leard and Goonbri Formations; Tadros, 1993) to the west of the New England Orogen and in the northern Sydney Basin (Allandale, Rutherford, and Farley Formations, McClung, 1980; Fig. 2) that were deposited in shallow-marine and lacustrine environments under the influence of glacial ice (Fielding et al., 2008). Like the earlier deposits, these formations have all been interpreted to have formed during a period of Early Permian extension (Scheibner, 1973, 1993; Jenkins et al., 2002; Korsch et al., 2009; Shaanan et al., 2015b).

Tectonic Reconstruction and Implications for Oroclinal Bending

Our results provide new constraints on the Early Permian paleogeography in the southern New England Orogen and specific information on paleocurrent directions in the Nambucca block. The implications of these results to tectonic reconstructions (Fig. 10) are discussed in the following.

All of the lower Permian basins and outliers mentioned here, from the Gunnedah and Sydney Basins to the west, eastward to the coastal exposures of the Nambucca block, are interpreted to have formed in response to crustal extension, most likely in a backarc context (Scheibner, 1973; Korsch et al., 2009; Jenkins et al., 2002; Shaanan et al., 2015b). Furthermore, a number of studies have shown that Early Permian extension occurred approximately simultaneously with oroclinal bending in the southern New England Orogen (Cawood et al., 2011b; Rosenbaum et al., 2012; Shaanan et al., 2014, 2015a, 2015b). We therefore think that the spatiotemporal record associated with the depositional environment of the lower Permian successions contributes to the understanding of the process of oroclinal bending.

The depositional environment of the Gunnedah and Sydney Basins, as well as the Texas region, Werrie syncline, and Manning Basin, was mainly lacustrine and shallow marine during the early Permian. In contrast, the lower Permian rocks of the Dyamberin and Nambucca blocks were initially deposited in a shallow-marine setting that progressively became deep marine over time. Altogether, the rocks of the Dyamberin and Nambucca blocks show distinct sedimentological features indicative of deposition in deeper water environments relative to the other lower Permian rocks farther west. Within the Nambucca block, samples from the Kempsey beds and Nambucca slate show remarkably similar geochronological provenances (Fig. 3), which also correspond to the geochronological provenance of the Dyamberin block (Shaanan and Rosenbaum, 2016), indicating that the two formations, and the two blocks, received detritus from similar sources. This means that the paleocurrent directions from the Kempsey beds likely represent the general paleocurrent direction in the Nambucca block. However, we note that the observed north- to northeast-directed paleocurrent measurements do not seem to conform with typical drainage orientations, which are expected to be orthogonal to the plate boundary. The inconsistency between the measured paleocurrent directions and the expected approximately north-south orientation of the continental margin suggests that the Nambucca block may have been rotated around a vertical axis. Such block rotations commonly occur in backarc regions during the progressive curvature of the plate boundary that is driven by trench retreat (e.g., see Lonergan and White, 1997; Rosenbaum, 2014).

In order to account for the paleocurrent orientations, we suggest that the Nambucca block was subjected to a counterclockwise rotation around the inferred hinge of the Manning orocline (Fig. 10). A number of independent arguments support this reconstruction. (1) Rocks from the Nambucca block, Cranky Corner outlier, and northern Sydney Basin (Fig. 1B) share similar sedimentary facies, thus suggesting connectivity during the early Permian. (2) Paleomagnetic studies from Devonian–Carboniferous forearc basin rocks of the Rouchel, Gresford, and Myall blocks record counterclockwise rotations by 80°, 80°, and 120°, respectively (Geeve et al., 2002), before 272 Ma (Shaanan et al., 2015a). It is therefore possible that the Nambucca block was subjected to a similar sense of rotation. The Hastings block father east (Fig. 10) has also been subjected to rotations and translations, although the sense of kinematics and timing of emplacement are debated (Schmidt et al., 1994; Roberts et al., 1995; Lennox and Offler, 2009; Cawood et al., 2011a, 2011b; Pisarevsky et al., 2016; Phillips et al., 2016). (3) A post–288 Ma counterclockwise rotation was also inferred for the eastern part of the Manning Basin, based on a recent structural investigation (White et al., 2016).

The timing of block rotations is constrained by several lines of evidence. The first phase of deformation in rocks of the Nambucca block is represented by penetrative vertical east-west slaty cleavage, and asymmetric, rounded, open mesoscopic folds (Leitch, 1978; Johnston et al., 2002; Offler and Foster, 2008; Shaanan et al., 2014), which were suggested to result from north-south contraction of the Nambucca block in the course of oroclinal bending (Offler and Foster, 2008; Shaanan et al., 2014). This deformation phase and thus the rotation of the block are constrained to have taken place after sediment accumulation (post–285 Ma; Shaanan et al., 2015b) and before deformation (pre–275 Ma) (Shaanan et al., 2014). These constraints overlap with the 298–288 Ma ages of a bent belt of granitoids that provide a maximum constraint for oroclinal bending (Rosenbaum et al., 2012), and a minimum age constraint from paleomagnetic data that indicate that no significant block rotations occurred after 272 Ma (Shaanan et al., 2015a).

A sedimentological study, complemented with a synthesis of detrital zircon geochronological data, was conducted on the lower Permian Kempsey beds of the Nambucca block, eastern Australia. The Kempsey beds, comprising a lower conglomeratic association and an overlying heterolithic sandstone-mudrock–dominated association, are interpreted to have formed in initially shallow water or even emergent fans and fan deltas, evolving over time into deeper marine slope environments. This implies the existence of a large, deep-marine basin along the present-day eastern Australian coast during the Early Permian that contrasted markedly with more localized, continental to shallow-marine extensional depocenters farther west in the New England Orogen. North to northeastern paleocurrent and paleoslope measurements suggest postdepositional counterclockwise northward rotation of the Nambucca block around the inferred hinge of the Manning orocline. The proposed kinematic reconstruction is supported by structural evidence from other blocks in the eastern limb of the Manning orocline, including the Devonian–Carboniferous blocks of the southern Tamworth Belt, Hastings block, and the Early Permian Manning Basin. A compilation of paleomagnetic and geochronological constraints places the rotation of the blocks, and the formation of the Manning orocline, between 285 and 275 Ma. These results provide robust indications for the much-debated existence of the Manning orocline, place time constraints for its formation, and contribute toward a fuller understanding of the development of the New England oroclinal structure and the late Paleozoic Gondwanan margins.

This research was supported by Australian Research Council Discovery Grant DP130100130 to Rosenbaum, S.A. Pisarevsky, Fielding, and F. Speranza. We thank two anonymous referees for their constructive reviews of the submitted manuscript.