INTRODUCTION AND GEOLOGICAL BACKGROUND

1 Fabrics defined by the alignment of clay particles are widely reported from mudstones, and generally attributed, if they are bedding-parallel or sub-parallel, to compaction during dewatering and diagenesis (see review in Kisch 1991 and papers by Oertel and Curtis 1972, Moon 1972, Krizek et al. 1975, Curtis et al. 1980, and Maltman 1981). Discordant fabrics are generally attributed to early tectonic deformation and considered to represent incipient cleavage development under conditions transitional from high-temperature diagenesis into anchizone and greenschist-facies metamorphism, and in such cases, fabric geometry can be related to finite strain (e.g., Dunoyer de Segonzac 1966; Kubler 1967a, 1967b; Piqué 1975, 1982; Weber 1976; Siddans 1977; Dandois 1981; Kisch 1989; Glasmacher et al. 2004). What is described in this paper is a pair of cleavages, highly discordant to bedding, in a mudstone where diagenetic history and thermal maturation data (vitrinite reflectance and spore coloration) indicate conditions that were well below the anchizone-diagenesis transition. The geometric relationship of these cleavages to bedding and regional structures (folds and thrust faults) suggests that they formed during a very early phase of bedding-parallel shortening, and were progressively obliterated by layer-parallel slip and shear. Not only were these cleavages formed under exceptionally low-grade conditions, but they are also ephemeral features of these mudstones, largely destroyed by subsequent deformation.

2 The Lower Carboniferous sedimentary rocks in the area south of Moncton, New Brunswick, are contained in a series of northeast-southwest anticlines and synclines in the Hillsborough area, that are part of a fold-thrust belt along the southern edge of the Moncton sub-basin (Figs 1, 2; Park and St. Peter 2005; Park et al. in press). These features all lie beneath the unconformity at the base of the late Tournaisian to Visean Hillsborough Formation (Park and St. Peter 2005; Park et al. in press; St. Peter and Johnson 2009). The pre-Visean (Fammenian–Tournaisian) rocks below the Hillsborough Formation are included in two stratigraphic units: Horton Group and Sussex Group. The Horton Group is subdivided into a basal Memramcook Formation (absent locally) and an overlying Albert Formation. An unconformity separates the Horton Group from the overlying Sussex Group. The Sussex Group is in turn subdivided into a conglomeratic Round Hill Formation, a red mudstone-dominated Weldon Formation, and a grey-red shale - evaporite Gautreau Formation. The three formations are partly lateral equivalent, but in the area of this study the Round Hill Formation is overlain by, and interbedded with, the Gautreau Formation, and both are overlain by the Weldon Formation (Park and St. Peter 2005; Park et al. in press; St. Peter and Johnson 2009).

Display large image of Figure 1

Display large image of Figure 2

3 The Weldon Formation is a thick succession of mudstone interbedded with red siltstone and red to brown sandstone, for which a thickness of at least 300 m is proven from boreholes. Mudstone is the dominant lithology, and locally comprises intervals in excess of 50 m with no siltstone or sandstone inter-beds and with very poorly defined bedding. Gypsum rosettes are present in much of this mudstone, and iron-reduction spots and fracture-related iron-reduction zones are ubiquitous. Siltstone and sandstone beds are generally less than 10 cm thick, and are typically laminated or cross-laminated on a sub-centimetre scale. Channels, scours and rippled surfaces are common, reflecting the floodplain to lake-margin depositional environment that is inferred for these rocks (see, for instance, St. Peter 2006).

CLEAVAGES IN MUDSTONE OF THE WELDON FORMATION

4 Early cleavages are best preserved in the mudstone that is exposed in a large section along Weldon Creek near Isaiah Corner (Albert County). The progressive destruction of early cleavages is best observed in the cliff section and tidal platform along the Petitcodiac River estuary at Belliveau Village (Westmorland County). These two localities are described in detail below.

SCANNING ELECTRON MICROSCOPY OF CLEAVAGES

17 Scanning Electron Microscopy (SEM) was undertaken on samples from Weldon Creek. The mudstone is exceptionally difficult material to sample on account of its fissility, friability and the presence of vivianite, which is hygroscopic and decrepitates in moist air. Samples had to be diamond-sawn out of outcrop and encased in surgical plaster before removal, then oven-dried and impregnated with polyvinyl acetate prior to cutting in order to prevent complete disintegration. Optical examination of thin sections revealed a close association between veinlets and both the S_a and S_b cleavages (Figs. 4a–c), but aside from confirming the presence of hematite and vivianite along the S_a and S_b cleavages, iron-staining and extremely fine grain-size precluded more detailed analysis. SEM images reveal more detail of both fabrics, and confirm the presence of veinlets that reinforce diffuse clay mineral septae (Figs. 4d, e). Both deformed detrital illite grains, and authigenic clay grains with preferred orientation are present, defining both S_a and S_b. In both cases, veinlets of an irresolvable fine mixture of clay, gypsum, and calcite are present (Figs. 4d–f). Cross-cutting relationships are also made apparent by the transposition of clay septae, where S_b-related veinlets cut the S_a cleavage (Fig. 4f). Qualitative data (EDAX spectra) imply that the detrital clays are largely illite (with negligible peaks for Ca, Na or Mg), whereas the authigenic orientated grains, and the material that is mixed with ultra-fine calcite, have a more montmorillonitic composition (Ca and Na peaks appear alongside K peaks in qualitative EDAX spectra). The presence of such clay with swelling properties may account for why the two cleavages form such good partings near weathered surfaces. SEM images did not reveal any clay mineral fringes on the detrital quartz or feldspar grains that are sporadically present in these mudstones.

18 SEM is the only microtextural analytic technique practically applicable to this material. Attempts to assess illite crystallinity by X-ray diffraction (XRD) proved ambiguous as the mudstones are still dominated by detrital clays. Difficulties in obtaining standard thin sections from such friable material precluded attempts to prepare wafers and apply transmission electron microscopy (TEM).

TIMING OF CLEAVAGE FORMATION

19 Timing of cleavage formation in these rocks relates to: (1) formation of the cleavages relative to the formation of folds, and (2) formation of cleavages in relation to the diagenetic history of these rocks. Three observations in the immediate environment of these cleavages pertain to the relationship with the diagenetic history. First, the mudstones contain gypsum rosettes with no trace of anhydrite (a development similar to that seen in the cleavage micro-veins), implying that conditions (especially temperature) have been consistent with gypsum stability ever since the rosettes (and micro-veins) formed. Second, the first evidence for reducing fluids percolating through the mudstones is iron-reduction haloes around shear planes that all cross-cut the two early cleavages and the conjugate pair of shear planes. There is no iron-reduction associated directly with either S_a or S_b (both the hematite and vivianite associated with S_a and S_b at Weldon Creek are oxidized phases). Elsewhere in the area these iron-reduction features are associated with hydrocarbon migration, implying a later stage of maturation-diagenesis. Third, the sandstone layers within the mudstone sections have their primary carbonate cement intact with no evidence of secondary porosity formation or silica overgrowth. These three characteristics imply that an early stage of diagenetic history is preserved in these rocks, and that the highest temperatures they have experienced lie in the range 60–100 °C. Depending on when the gypsum rosettes grew, the lower end of this range is more probable. This observation places severe constraints on depth of burial. It also places these rocks below the range for conditions consistent with ‘anchimetamorphism’ (see Kisch 1989, 1991).

20 Such conditions for maximum burial depth and thermal maturation do not rule out the formation of cleavage. Indeed, the appearance of a fissile fabric in claystone particularly is well documented (Kisch 1989, 1991 for reviews; see also Oertel and Curtis 1972; Curtis et al. 1980, Krizek et al. 1975; Maltman 1978, 1981). However, these early cleavage-like fabrics are produced by compaction and dewatering, and the resultant single fabric is always bedding-parallel, or close to being in this orientation. This situation is in marked contrast to the case described at Weldon Creek, where the two cleavages are both perpendicular to bedding. Furthermore, the fabrics that are created by compaction and dewatering are generally described as ‘fissility’ (a penetrative fabric) rather than as a spaced, domainal foliation (Moon 1972). There may well be a continuum between the two, but the end-members of shaly fissility and domainal cleavage are morphologically quite distinct.

21 Bedding-parallel fissility forms during compaction and may be associated with deformation as well (Curtis et al. 1980), but this deformation generally resolves to a shortening perpendicular to bedding. Deformation associated with S_a and S_b in these mudstones is quite distinct in this respect. Veinlets parallel to S_b, and also perpendicular to bedding, show no evidence of shortening along their lengths, strongly suggesting that these veinlets were emplaced after the compaction associated with dewatering. Similarly, neither cleavage is deformed in a fashion that could be attributed to compaction, placing another constraint on timing with respect to diagenesis. The most marked deformation features associated with these cleavages are the development of the slickenside lineation on S_a (L_s), and the subsequent rotation of L_s and curvature of S_a when it was transposed into S_b, none of which can be resolved by simple loading with a maximum principal stress that would be perpendicular to bedding. Here the fabric geometry suggests a bulk shortening parallel to bedding.

22 In summary, two cleavages formed perpendicular to bedding in these mudstones after compaction and dewatering, but prior to folding-related deformation associated with the Weldon and Belliveau synclines. Indirect evidence suggests that both cleavages formed during the early stages of diagenesis (prior to iron-reducing fluid and hydrocarbon migration), probably at temperatures less than 100 °C in rocks that were never deeply buried. Geometry of the fabrics does imply a relationship to bulk shortening of the layers parallel to bedding, possibly an early response to the same shortening that generated the folds. Critically, both cleavages predated the folds themselves.

CONDITIONS OF CLEAVAGE FORMATION

23 The conditions under which the S_a and S_b cleavages formed are constrained by two sets of independent data from either the Weldon Formation itself, or from underlying oil shale of the Albert Formation. Diagenetic minerals in the red mud-stone of the Weldon Formation include gypsum, occurring as veinlets within the second cleavage (S_b), or as rosettes in the matrix with an ambiguous relationship to either cleavage (S_a and S_b). Petrographically, there is no evidence of the presence or former presence of anhydrite, indicating that these red mud-stones have never been buried sufficiently since an early stage of diagenesis to pass through the gypsum - anhydrite transition. Given that this reaction involves a very large molar volume change, it is pressure sensitive, but occurs at around 50 °C at atmospheric pressure. This temperature will fall with increasing pressure through burial, placing a severe constraint on depth of maximum burial (certainly less than 1.5 km).

24 Direct temperature estimates are also available from studies of the oil shales in the Albert Formation (Chowdhury 1995). Equivalent % vitrinite reflectance (%R) data from around Weldon Creek and Belliveau Village give values between 0.65 and 0.75 (with a maximum of 0.77 from a borehole at the albertite mine site, Albert Mines, 2 km from the Weldon Creek site). Thermal Alteration Index (TAI) from spore coloration has also been determined for rocks at Albert Mines, with values averaging 2.00 (Chowdhury 1995; MacIntosh and St. Peter in press). These data translate to a maximum temperature in the 70 to 90 °C range, and critically, the lower figures come from the top of the Albert Formation typically 100 to 150 m below the level of the exposed Weldon Formation in which these cleavages are observed. These data imply that maximum temperatures in the Weldon Formation of Weldon Creek and Belliveau Village were no higher than 50 °C – consistent with the observations regarding diagenetic gypsum as rosettes or micro-veins in the cleavages.

25 It is useful to compare these conditions to those generally associated with early cleavage formation, namely the higher temperature diagenetic anchizone regime transitional to prehnite-pumpellyite metamorphic facies (Dunoyer de Segonzac et al. 1966; Weber 1976, 1981; Piqué 1975, 1982; Siddans 1977; Dandois 1981; Kisch 1989, 1991, 1998; Gutiérrez-Alonso and Nieto 1996; Glasmacher 2004). These authors described conditions well above %R of 1.0 (as high as 3.0) for conditions of early cleavage formation and incipient metamorphism in the anchizone.

26 Although they are interpreted as tectonic features, these two cleavages and the resultant pencil structure cannot be related to any measurable strain (cf., Reks and Gray 1982). This is due to a number of factors, not least of which is the absence of reliable strain markers in these red mudstones. Thin section and SEM examination of the mudstones has not revealed any pressure fringes on detrital grains, despite the abundance of quartz and feldspar grains in some samples. Veinlets parallel to the cleavages (S_b mainly) show no evidence of shortening or elongation, though in the case of S_a there is some evidence of quartz dissolution along the cleavage septae. If silica from detrital grains was being mobilized during S_a formation, then the absence of fringes around other detrital grains may indicate very low strain, or at least strain that is well below the threshold defined by Reks and Gray (1982), and certainly less than the 40–60% shortening generally cited for conditions of incipient slaty cleavage formation (Kisch 1989, 1991). Though there is overwhelming evidence in these rocks that strain cannot be constant volume, fluid-mediated mass transfer on a scale necessary to accommodate 40–60% shortening is highly unlikely.

27 Finally, though now somewhat scattered, the two cleavages may have originated as a conjugate pair. Unlike the disruptive shear planes seen in the Belliveau section, such a pair would have no obvious relationship to the axial plane of any recognized fold. As an element of shear rather than simple shortening is implied by this orientation, ‘fracture’ cleavage is the preferred term for these features, rather than ‘slaty’ cleavage or incipient ‘slaty’ cleavage.

DISCUSSION AND CONCLUSIONS

28 The presence of cleavages in the mudstones of the Weldon Formation is significant for a number of reasons.

1. Timing relationships demonstrate that these cleavages formed later than compaction-dewatering but very early in the diagenetic history of these rocks (prior to hydrocarbon migration). Their geometry suggests that they are related to a local deformation that affected these rocks shortly after deposition and by the end of the Tournaisian (prior to the unconformity beneath the overlying Hillsborough Formation; Park and St. Peter 2005; Park et al. in press). The demonstration of a tectonic relationship for a very early pair of cleavages here implies that mechanisms other than dewatering-compaction may be important elsewhere.
2. Relationships of these cleavages to folding are particularly revealing: the rotation of beds during folding was partly accommodated on the very shear planes responsible for the progressive obliteration of both early cleavages. Although related to the deformation that ultimately produced the folds (a bedding-parallel bulk shortening), these two cleavages predate the folds themselves. S_a and S_b formed as an early response to bedding-parallel bulk shortening, and were subsequently disrupted and progressively obliterated as folding began and bulk shortening gave way to layer-parallel shear and rotation.
3. Both cleavages (S_a and S_b) are essentially ephemeral features of these mudstones that were preserved locally where favourable factors prevailed. They were probably far more widespread than current preservation suggests. Several authors have indicated the importance of pre-existing fabrics in the nucleation of new cleavages (Williams 1972; Cosgrove 1976; Gray and Durney 1979; Maltman 1981; Craig et al. 1982; Ho et al. 1996), though this is tacitly assumed to be an early fissility, caused by compaction-dewatering. The presence of cleavages in the Weldon Formation mudstones that are not compaction-dewatering features, and yet still ephemeral (and not axial planar) suggests that fabrics other than compaction-dewatering related fissility may form earlier than folding, and that they too provide anisotropies that can be recycled as new cleavages nucleate in an evolving rock mass during deformation.
4. The pencil structure created by the intersection of S_a and S_b is also unusual (see review in Reks and Gray 1982) in so far as such a structure is generally considered to be an intersection of a cleavage or incipient cleavage with a primary shale fissility, therefore an L¹₀ intersection lineation, close to, if not perfectly parallel to a fold axis. At neither locality in which this pencil structure is preserved is it bedding-parallel, or anything other than close to being perpendicular to bedding. It also bears little relationship to the geometry of the large-scale folds.
5. The ephemeral nature of S_a and S_b in the Weldon Formation is critical. They are preserved locally where conditions permit, but the evidence from the Belliveau Village section suggests that once folding began layer-parallel slip and shear obliterated these features. This is in marked contrast to most anchizone cleavages described in the literature (see Kisch 1991), which are regarded as the precursors to the slaty cleavage that is more fully developed in higher grade anchizonal and greenschist facies pelites. The early cleavages described here are fracture cleavages with evidence of shear, rather than true ‘slaty’ cleavages or their precursors. Other than providing another planar anisotropy at the scale of clay grains, on which subsequent cleavages may nucleate, the cleavages in the Weldon Formation mudstones cannot be considered precursors to anything seen at higher grades of recrystallization.