'Middle' Cambrian stratigraphy of the rifted western

2 downloads 0 Views 6MB Size Report
Monarch Cirque–South. Rifted Laurentian Margin and Castle Mountain restored to ... lithological interval characterized by large dome-shaped bodies of weakly-bedded to ...... Colorado School of Mines Quarterly, vol. 80 (3). 70 pp. Jeary, V.A. ...
Palaeogeography, Palaeoclimatology, Palaeoecology 277 (2009) 63–85

Contents lists available at ScienceDirect

Palaeogeography, Palaeoclimatology, Palaeoecology j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / p a l a e o

Reinterpretation of ‘Middle’ Cambrian stratigraphy of the rifted western Laurentian margin: Burgess Shale Formation and contiguous units (Sauk II megasequence), Rocky Mountains, Canada C.J. Collom a,⁎, P.A. Johnston b, W.G. Powell c a b c

Department of Chemical & Petroleum Engineering, The University of Calgary, 2500 University Drive, Calgary, Alberta, Canada T2N 1N4 Department of Earth Sciences, Mount Royal College, Lincoln Park Campus, Calgary, Alberta, Canada T3E 6K6 Department of Geology, Brooklyn College of CUNY, 2900 Bedford Avenue, Brooklyn, NY 11210, USA

a r t i c l e

i n f o

Keywords: Cambrian Burgess Shale Seeps Mud mounds Cathedral Escarpment Sequence stratigraphy

a b s t r a c t Until recently, research on the renowned fossil animals of the Burgess Shale has advanced to a greater degree than an understanding of the rocks in which they are found. Studies addressing lithostratigraphy and hydrothermal petrography of the so-called ‘inner carbonate belt’ and adjacent Chancellor Group, however, have begun to re-evaluate long-standing hypotheses on the Middle Cambrian of Western Canada. Tectonic activity along the Kicking Horse Rim (buried remnants of Neoproterozoic rifting) during the Cambrian Period had more influence on local sedimentation than previously thought. Notably, large-scale collapse of the Cathedral carbonate platform margin at ∼ 509 Ma BP is evidence of reactivated basement faults. These failures produced listric Megatruncation Surfaces, having near vertical escarpments (N 150 m in height) where they terminate against the platform. The majority of what have been interpreted as shed olistoliths from the Cathedral platform margin is herein shown instead to be carbonate mud mounds. These grew in situ along the face of the Cathedral Escarpment, and are associated with fossiliferous intervals in overlying basinal mudstones at three distinct stratigraphic horizons (two within the Burgess Shale) during the Delamaran and Marjuman stages. The mounds nucleated where deep-seated normal faults intersected the seafloor, building primarily during periods of relative sea level transgression. Mound growth is associated with syndepositional exhalative activity, inferred from stratigraphic relations, sedimentology, and geochemistry. The new name Monarch Formation is proposed for the initial post-collapse beds and mounds deposited against the Cathedral Escarpment (early Glossopleura Zone); the variably-defined term ‘Takakkaw Tongue’ is thereby confined to pre-collapse slope deposits coeval to and correlative with the Cathedral Formation. © 2009 Elsevier B.V. All rights reserved.

Nature has a habit of placing some of her most attractive treasures in places where the average man hesitates to look for them. • Charles D. Walcott, 1911 1. Introduction Discovery of the exceptionally well-preserved Cambrian fossils within the Burgess Shale on Fossil Ridge, British Columbia by Charles Walcott a century ago have been justifiably touted as among the most important paleontological sites on Earth (Gould, 1989). Not only did the study of these N500 Myr old fossil organisms (often retaining soft-tissue details) forever change the way we view early animal evolution, but it also introduced the scientific community to a poorly understood suite of

⁎ Corresponding author. E-mail address: [email protected] (C.J. Collom). 0031-0182/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.palaeo.2009.02.012

mixed carbonate and siliciclastic rocks in the Kicking Horse Pass region, of which the Burgess Shale is part. During the decades since its discovery, a vast literature — with its associated controversies — has amassed steadily on the Burgess Shale, but almost exclusively with respect to the taxonomy and paleoecology of the nonbiomineralized fossils (Coppold and Powell, 2000). Following recent lithostratigraphic studies (e.g., Stewart et al., 1993; Fletcher and Collins, 1998), we herein present new observations of these otherwise poorly-scrutinized strata from throughout the southern Canadian Rockies and propose emendations to existing stratigraphic schemes of previous studies (Figs. 1,2). Relying on field observations made during six helicopter-assisted seasons (1998– 2003), extensive collections of paleontological and lithological samples, and sequence biostratigraphy concepts, this study addresses these marine facies from a multi-disciplinary perspective. Lack of modern sedimentological analysis of the marine facies in which these invertebrate fossils are entombed has done a disservice to this most familiar of fossil lagerstätten. The reinterpretations set forth in this

64

C.J. Collom et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 277 (2009) 63–85

Fig. 1. Map of study area in Yoho, Kootenay, and Banff National Parks. Northwest–southeast trend of the Cathedral Escarpment after Stewart (1991). Isopach contours (modified from Aitken, 1997) indicate thickness of: A. upper Gog Group (late ‘Early Cambrian’), B. the Mt Whyte/Cathedral Grand Cycle: note that the edges of isolated thickened strata correspond closely to two prominent embayments (within inferred grabens) along the Cathedral Escarpment. Numbered localities are: 1. Fossil Ridge (Walcott Quarry), 2. Mt Field, 3. Mt Stephen (northwest ridge), 4. Mt Stephen (Trilobite Beds), 5. Mt Odaray, 6. Park Mtn., 7. Natalco Lake, 8. Monarch Cirque-North, and 9. Monarch Cirque–South. Rifted Laurentian Margin and Castle Mountain restored to their position prior to Cordilleran overthrusting. C. A–A’ seismic transect through the Canadian Cordillera (from Van der Velden and Cook,1996); note the position of the Kicking Horse Rim (KHR) relative to the thickened intervals of Grand Cycle #1 sediments (Mt Whyte–Cathedral). Features illustrated are Slide Mountain Fault (SMF), Dog Tooth High (DTH), Rocky Mountain Trench (RMT), and Lewis Thrust (LE); numbered crustal elements include: 1. and 2. — Middle Proterozoic (Purcell Group; Windermere Grp absent), and 3. — Paleozoic and Mesozoic (E. Cambrian to L. Cretaceous). Positions of the westward-thinning Moho from Van der Velden and Cook (C: 1996) and estimated from Price (P: 1986).

study have wider implications, touching on a broad spectrum of Burgess Shale topics — including subsurface structural geology, benthic paleoenvironments, seafloor exhalative processes, and relative sea level change in platform margin settings. 1.1. Background An appropriate starting point for any stratigraphic discussion of the Burgess Shale is with the Cathedral Escarpment, because it is in the shadow of this prominent feature that these dramatic series of geological events unfold. Aspects of the Cathedral Fmn resemble

closely carbonate platforms and ramps known from throughout the Paleozoic (Wilson, 1975; Enos, 1983; Greenlee and Lehmann, 1993), particularly with respect to facies and geometry. An important, albeit not unique, characteristic of this widespread and thick limestone and dolostone unit was inadvertently re-discovered by Ney (1954) during a mining survey of the Kicking Horse Pass region. Ney described that “a striking change occurs at the top of the Cathedral Formation. Here there is a steep west-facing precipice of dolomite nearly 400 ft high, against which shales on the west terminate abruptly…it seems to be an original feature of deposition…” (ibid., p. 123–124). From this brief report was born the recognition of an ancient submarine cliff, the

C.J. Collom et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 277 (2009) 63–85

65

Fig. 2. Lithostratigraphic and biostratigraphic framework of ‘Middle’ Cambrian platformal carbonates and adjacent basinal mudstone facies belts addressed in this study. The controversial interval previously assigned to the upper Takakkaw Tongue is herein named the Monarch Formation, representing the earliest post-margin collapse deposition; corresponding sections from Stewart (1991) and Fletcher and Collins (1998). Of the ten members of the Burgess Shale Fmn introduced by Fletcher and Collins (1998), the lower five members can be recognized only near the Kicking Horse Pass region. Consequently, the available name Wapta Mbr (sensu Aitken, 1997) is used for the pisoid-rich interval above the Walcott Quarry Mbr. Placement of the boundary between Grand Cycle #1 (Mt Whyte/Cathedral) and #2 (Stephen/Eldon) at the top of the Cathedral Fmn (early Glossopleura Zone) follows Aitken (1997). Cambrian stage names from Palmer (1998); MTS = Megatruncation Surface.

origin of which would remain a geological enigma for another forty years. Interestingly, some four decades prior to Ney's account the same abrupt transition from platformal carbonates to basinal siliciclastics in the Yoho National Park region had been interpreted as a Laramide fault. Representing the Geological Survey of Canada, Allan (1914) stated “The southwestward limit of these Middle Cambrian beds is defined by a fault called the “Stephen-Dennis Fault,… This break was traced for 15 miles southeast of the Kicking Horse valley” (ibid., p. 67). However, nearly all subsequent work has confirmed that this abrupt facies change is in fact an original geological feature, not one produced by structural processes and orogenic activity. Varied studies during the ensuing decades (Brown, 1948;Rasetti, 1951; Aitken, 1971; Fritz, 1971; McIlreath, 1977a; Whittington, 1985; Aitken, 1989; Stewart et al., 1993; Fletcher and Collins, 1998) have greatly improved the working model of the Cathedral margin. Much of the current understanding of the Cathedral margin has been gained from comparison to Recent carbonate-dominated analogs, particularly those with relatively steep (N70°) margins that experience failure frequently (Cook and Mullins, 1983; Hine et al., 1992; Grammer et al., 1993; Masson et al., 1998). In addition to such comparisons, the nature of the unusual strata described herein necessitates inclusion of other depositional environments associated with continental margins — such as ‘passive margin’ tectonism and seafloor regions prone to fluid and/or brine exhalation (Struik, 1987; Paull and Neumann, 1987). This study focuses on the fine-grained rocks deposited during the Delamaran Stage and lower Marjuman Stage — traditionally assigned to the ‘Middle Cambrian’ — from approximately 509 to 503 Myrs ago

(Palmer, 1998). The Delamaran Stage spans the Plagiura, Albertella, and Glossopleura trilobite zones in Western Canada, an interval of ∼6.0 Myrs, from 512 to 506 Ma BP (Palmer and Geissman, 1999). Within the Delamaran–Marjuman stages of Western Laurentia are three of the most prolific invertebrate fossil lagerstätten of the Early Paleozoic (Figs. 2,4) — including the horizons associated with the ‘Phyllopod Bed’ on Fossil Ridge. Each is characterized by preservation of nonbiomineralized or ‘soft-bodied’ fossils directly above an unusual lithological interval characterized by large dome-shaped bodies of weakly-bedded to massive micritic calcium carbonate. These are indistinguishable from carbonate ‘mud mounds’ described elsewhere from the Early Paleozoic (Pratt, 1995; Belka, 1998; Somerville, 2003), their origin represents one of the most controversial elements of the Delamaran–Marjuman lithological succession. As many aspects of the Burgess Shale are inherently controversial and prone to widely-differing interpretation, in part owing to their complex geology and remote exposure, much of this paper is revisional in nature. With recent advances in stratigraphy and carbonate shelf/platform edge processes, we have benefited immeasurably from information previously unavailable to Burgess Shale workers. Despite these advantages, key questions remain; thus this study should only be considered a ‘next step’ in the effort to resolve the geological and biological events that transpired along the Laurentian margin half a billion years ago. In summary, this paper will: (1) address important lithostratigraphic issues, (2) define mud mound-bearing intervals in Cambrian strata of the Columbia Basin (including naming of a new formation), (3) establish a sequence stratigraphic framework for the study interval, and (4) suggest

66

C.J. Collom et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 277 (2009) 63–85

Fig. 3. Stratigraphic features of the Glossopleura Zone, Yoho and Kooteney National Parks; location of Cathedral Escarpment [‘e1’] indicated by arrow. A. South side of Monarch Cirque illustrating effect of extensional rifting on Early Paleozoic sedimentary cover. The Kicking Horse Rim is represented by a sharp basinward pinch-out of the Gog Group (Lower Cambrian), interpreted to be the fault-bounded edge of a half-graben with its footwall block to the lower right (not exposed). Mt Whyte facies thinned towards the edge of this block (lower black arrow), where Gog Grp quartzites are separated by an erosional contact with dolomitized carbonates of the Cathedral Fmn (Middle Cambrian). The Cathedral Escarpment separates bedded mudstones of the Monarch and Burgess Shale formations (on right) from highly dolomitized, sulfide-bearing megabreccia formed by hydrothermallydriven dissolution collapse prior to formation of the megatruncation surface. Inset photo: undolomitized carbonate mud mounds of the Monarch Fmn (M) occur adjacent to Cathedral Megabreccia; persons for scale are circled. B. Cross-section of the Cathedral Escarpment on northwest ridge of Mt Stephen. The upper carbonate-dominated portion of the Monarch Fmn [Fmf: consisting of anastemosed mud mounds] terminates abruptly against dolomitized platformal facies of the Cathedral Fmn [surface ‘e1’, on left], and sharply overlies the Takakkaw Tongue of the Cathedral Fmn. Overlying mudstones are of the Kicking Horse Mbr of the Burgess Shale Fmn [Fbkh], which includes a large olistolith of early dolomitized Cathedral Limestone (refer to Figs. 4,9). C. Monarch Fmn on near-vertical north-facing cirque wall of Mt Odaray; Cathedral Escarpment indicated by black triangles. The e1 Megatruncation Surface has a two-part headwall and a ‘bench’ at this locality. D. Cathedral, Monarch, and Burgess Shale formations at Natalco Lake. Carbonate mud mounds [‘M’] located on distally-sloping surface of the Cathedral Escarpment. These are correlative to Monarch Fmn mounds exposed on Mt Stephen and Mt Field. Glossopleura-Bathyuriscus Zone boundary (= contact between Monarch and Burgess Shale formations) indicated by dashed line. Arrow at bottom of photo indicates direction of ∼ 35 m thick exhalite deposits of lower Monarch Fmn (Powell et al., 2006).

emendations for the correlation and outcrop interpretation of the Burgess Shale and related units on Mt Field and Mt Stephen. 1.2. Rift tectonism & lithostratigraphic overview The northern margin of Laurentia is a so-called “passive margin” or rifted remnant of the Neoproterozoic break-up of the supercontinent Rodinia (Struik, 1987). Subsequent counter-clockwise rotation of Laurentia has changed the orientation of the craton by nearly 90°; due north during the Cambrian is now west. Additionally, Laurentia has

tectonically drifted from some10° south of the Cambrian paleoequator to N50° north at present). Active extension appears to have declined from ∼ 700 Ma BP to the Ediacaran, then reactivated during deposition of the Gog Group in the Early Cambrian (Kubli and Simony, 1994; Lickorish and Simony, 1995). During this ∼ 20 myr interval several hundred metres of predominantly quartzose sandstones were deposited along the Laurentian margin. No consensus exists regarding the provenance of these prolific siliciclastic sediments, but they imply a nearby source area having subaerially-exposed, granitic uplands (to account for the volume of quartz). These sediments were draped over

Fig. 4. Lithostratigraphic and sequence stratigraphic correlation between the platform-top detrital sequences (right) and their basinal equivalents (on left). Note the difference in scale between the two columns. Carbonate mud mound intervals (upper Monarch Fmn, Yoho River Mbr, Wash Mbr) in the deep basin represent deposition during general conditions of sedimentary starvation and rising sea level (Transgressive System Tracts: TST). In contrast, thicker successions of siliclastic mudstone correspond to intervals of sea level fall (Regressive System Tracts: RST). Accordingly, the Burgess Shale was deposited during two and a half successive third-order sequences: RST of DEL-5, DEL–MAR Transitional Sequence, and MAR-1. T–R cycle nomenclature after Beauchamp and Henderson (1994); member names for the Stephen Fmn from Aitken (1997). Carbonate bedding shown in blue, gray background indicates the Bathyuriscus Zone. Abbreviations: Y.R. = Yoho River Limestone Mbr, W = Wash Limestone Mbr, Camp Cliff = Campsite Cliff Mbr.

C.J. Collom et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 277 (2009) 63–85

67

68

C.J. Collom et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 277 (2009) 63–85

a continental margin and slope that have been compared to that of the Atlantic Ocean: en-echelon, normal fault-bounded half grabens stepping-down into the widening central basin. It is above the distal ‘cusp’ of the most landward of these half-graben salients (more accurately: a NW–SE trending series of tilted half-grabens), commonly referred to as the Kicking Horse Rim (Aitken, 1971) that the events described in this study occurred. Post-Early Cambrian sediments deposited in the vicinity of the Kicking Horse Rim are dominantly fine-grained, consisting of

carbonate micrite and/or clastic mudstone. These facies are consistent with low-energy, deeper water environments (N100 m), typical of continental margin settings. All Middle Cambrian stratal units along the Laurentian margin eventually grade into homogeneous finegrained mudstones of the Chancellor Group, that exceed 1.0 km in thickness and are generally interpreted to have been deposited in a deep-water setting (Stewart, 1991). Micrite production on the Cathedral platform via the so-called “carbonate factory” (Bathurst, 1975) appears to have become significantly curtailed during the early

C.J. Collom et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 277 (2009) 63–85

Glossopleura Zone. The transition from carbonates to clastics at this stratigraphic horizon is sufficiently pronounced, that Aitken (1978) chose it as the boundary between the first and second “Grand Cycles” of the Cambrian succession in the Canadian Rockies (Fig. 2). Within the study area little evidence can be found for catastrophic depositional processes during the Cambrian. Storm deposits and associated fans of platform-top sediment at the base of the escarpment are uncommon, and mass wasting events even rarer. Large-scale disintegration of extensive portions of the Cathedral margin is thought to have proceeded as a series of pervasive, short-term episodes of dissolution, brecciation, and eventual collapse (Stewart et al., 1993). The last of these processes may have affected at least a 75 km-long section of the Cathedral platform margin along the Kicking Horse Rim (Fig. 1). We refer to this collapse event as the Cathedral Megatruncation Event, and present evidence that it occurred at the termination of carbonate platform growth (early Glossopleura Zone). The listric Megatruncation Surface (MTS) generated is the first of two documented platform margin collapses during the Cambrian; the other is associated with the Eldon Formation. Both mass wasting events can be linked to dolomitization– mineralization events involving exhalative processes (Powell et al., 2006). The Cathedral Escarpment is therefore labeled as e1 in Figs. 3–8. The younger Eldon Megatruncation Event (or “e2”) will be addressed in a separate paper. The margin collapse paradigm of Stewart (1991) has largely supplanted the previously-held model of the Cathedral Escarpment as a vertically-accreted Epiphyton reef edge (Aitken, 1971; McIlreath, 1977a). Field observations indicate that the maximum amount of material removed by collapse of the Cathedral margin did not scallop beyond several 10s of m inboard of the platform edge, or below approximately the level of the Trinity Lakes Mbr. But this may simply reflect the lack of well-defined Trinity Lake calcareous mudstone facies within several hundred metres of the pre-MTS Cathedral platform margin, where both shalier members tend to pinch out. At its type section, the Trinity Lakes Mbr is 98.1 m below the Cathedral top, north of Sunwapta Pass in Jasper National Park. Yet the Cathedral escarpment was at least 200 m in height in places, based on exposures in Kootenay and Yoho parks. After the Cathedral Megatruncation Event, the scalloped platform margin and adjacent basin began to accumulate relatively more siliciclastic fine-grained sediments. We have re-named the initial 50–95 m of this complex interval the Monarch Formation in lieu of earlier informal nomenclature, such as “thin Cathedral” or “Takakkaw tongue of the Cathedral” (McIlreath, 1977a): see discussion below and formal description in Appendix A. 2. Mud mound-bearing intervals 2.1. Monarch Formation — lower portion The initial deposits in the Columbia Basin (adjacent to the newlyformed escarpment) after the Cathedral Megatruncation Event are

69

among the most controversial strata in the North American geosciences literature. At issue are the lithostratigraphic assignment of these beds, their thickness, relation to adjacent units, lateral extent, and paleo-depositional setting (relative to formation of the adjacent escarpment). Ideally, this interval should have been included in the definition of the Burgess Shale Formation (Fletcher and Collins, 1998), but there can be no doubt that there is a significant interval of basinal deposits below the Burgess Shale Formation (sensu stricto). We herein name these deposits the Monarch Formation, after a mounain of the same name in Kootenay National Park (see Appendix A). As is the case with overlying non-platformal Cambrian strata, these deposits thin dramatically away (basinward) from the Laurentian margin. The ‘lower’ facies of the Monarch Fmn are predominantly calcareous mudstones that are well-bedded in places (Aitken, 1997: Figs. 45,46; Figs. 3C,7A herein), yet significant amounts of limestone are absent. The lower Monarch Fmn is 48 m thick at The Monarch type section (Johnston et al., 2009-this volume). Rasetti (1951) interpreted the stratigraphic interval herein named the Monarch Fmn as a shaley continuity of the Cathedral Fmn, not having had the benefit of Ney's (1954) critical reexamination of this apparent abrupt facies change. In contrast, facies of the ‘upper’ Monarch Fmn are massive, partly-dolomitized carbonate and prominent carbonate mud mounds (McIlreath, 1977a: pl. 6–1; Fig. 7A,B herein). Stewart (1991, pl. 11) referred to the lower, relatively monotonous interval as ‘ribbon calcilutite lithofacies’ —characteristic for much of the N1.0 km thicknesss of the basinal Chancellor Group (Fig. 5F). Intraclastic-peloidal rudstone/grainstone (‘chipstone’) is also present in the ‘lower’ Monarch Fmn at southern localities including Natalco Lake and The Monarch (Stewart, 1991: pl. 22). These distinctive rocks are composed of a micritic limestone matrix in which abundant tabular, buff-weathering, micritic dolostone clasts, each 2–10 mm in length, are set. The abundant dolomitic intraclasts of this lithofacies were likely derived from the adjacent platform, where shallow water environments conducive to their formation were at least locally developed (Moore, 1994; Jeary and Spencer, 2002). 2.2. Monarch Formation — upper portion Overlying these calcareous mudstones is a relatively more calcareous interval (4–28 m thick) of mixed thin-bedded calcilutite, ooidal packstone, plus a mud mound-bearing interval that Rasetti (1951, pl. 6) simply referred to as the “anomalous lithology.” During the period from the early 1970s (Fritz, 1971) to late 1990s (Stewart et al., 1993) the predominant view regarding these carbonate structures in the basinal Lower Chancellor Group (“Thick Stephen” Fmn of Fritz, 1971) was that they were randomly shed talus from the adjacent vertically-accreted Cathedral reef. This model did not explain why the majority of these ‘talus blocks of Epiphyton boundstone’ lack internal bedding and dolomitized facies like that of the Cathedral platform, or why they occurred in distinct stratigraphic horizons — rather than being randomly distributed throughout the basinal deposits. Brecciated lithology of all

Fig. 5. Geological history of the northern margin of Laurentia: Glossopleura Zone to early Bathyuriscus Zone interval. A. Prior to Cathedral platform collapse, hydrothermal dissolution and brecciation combined with down-dropping of basement half-grabens lead to destabilization and eventual listric Megatruncation Surface [MTS]. Dolomitizing fluids migrated upward into platform along basement faults (major brine source) and downward via gravity-driven infiltration (minor source). B. Initial collapse, located over edge of tilted basement salients of the Kicking Horse Rim, locally produced an vertically continuous N 200 m submarine escarpment (e.g., Natalco Lake locality) and a stepped escarpment in other areas (Monarch Cirque, Mt Odaray). Megatruncation event occurred towards end of relative sea level rise (TST). Interval of carbonate-dominated agradation, that began with deposition of the upper Mt Whyte and Naiset formations [Plagiura Zone], comes to an end. C. Onset of more siliciclastic-rich sedimentation; initial deposition along base of escarpment are calcareous mudstones of the Monarch Fmn, and the lower shoaling-upward succession of the Narao Mbr, Stephen Fmn (not shown). Subsequently, carbonate mud mounds form adjacent to Cathedral Escarpment; some slide downslope and disintegrate in response to gravity and ongoing seismic disturbances. D. Non-calcareous mudstones of the Kicking Horse Mbr (lowest portion of the Burgess Shale Fmn) are deposited, blanketing Monarch Fmn mounds. Scattered pods of calcareous ooids indicate periodic storm winnowing of platform-top facies. The overlying Yoho River Mbr records a gradational transition to thin-bedded, and then massive carbonates — including mud mounds several metres in height. Upper shoaling-upwards succession of the Narao Mbr is correlative to the Kicking Horse–Yoho River sequence [DEL-5]. E. Yoho River Mbr mounds are overlain sharply by noncalcareous mudstones of the Campsite Cliff Shale Mbr. The basal few meters are densely fossiliferous, dominated by the large trilobite Ogygopsis klotzi. The base of the Campsite Cliff Shale Mbr marks the boundary between the Glossopleura and Bathyuriscus trilobite biozones; separated from essentially identical facies of the Walcott Quarry Mbr by an overlying interval of fossiliferous, mound-bearing, breccia-rich carbonate of the Wash Limestone Mbr (∼6 m thick). Succeeding 10–15 m of Walcott Quarry Shale Mbr contain abundant and diverse soft-bodied metazoan fossils. Non-bioturbated, laminated mudstones of the Walcott Quarry Mbr are correlative to the lower Waputik Mbr, Stephen Fmn on the adjacent platform. F. Subsequent basinal facies of the Lower Chancellor Group consist of interbedded, non-fossiliferous mudstones, and pisoidal packstone and wackestone beds of the Wapta Mbr. Thickening upsection, these pisoid beds grade into the cyclically-bedded peritidal carbonates of the Eldon Fmn. Limestones and dolostones of the Eldon Fmn, representing a return to carbonate-dominated deposition along the northern margin of Laurentia.

70

C.J. Collom et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 277 (2009) 63–85

Fig. 6. Delamaran and Marjuman platform-to-basin successions on the Kicking Horse Rim (KHR). A. Interpretation of Stewart (1991) and Stewart et al. (1993); ‘F’ denotes position of Walcott Quarry, above the so-called ‘Boundary Limestone.’ The Monarch Fmn is the interval above MTS-‘0’ included in theTakkakaw Tongue (sensu lato). Four megatruncation surfaces (MTS) are inferred, including two within the younger Duchesnay Unit, basinward of the Eldon platform. B. Interpretation presented herein; ‘e1’ indicates the Cathedral Escarpment (‘e2’ is a nearly identical escarpment of the Eldon Fmn, not discussed herein). White colour in Cathedral Fmn indicates dolomitized carbonates, contact with platformal limestones (light blue) is simplified. Correlations between Stewart's MTSs and those identified herein: only MTS-‘0’ and MTS-2 are considered to be megatruncation surfaces; the others being sequence stratigraphic surfaces and associated facies changes. Wash Limestone Mbr too thin to be shown at this scale (see Figs. 2,4,5E, and 11).

size classes generally have sharp edges and corners; the carbonate bodies in question are nearly always dome-shaped. Evidence, presented elsewhere, indicates a seep-related origin for the majority of these ‘anomalous’ carbonate bodies (Collom and Johnston, 2000b; Collom, 2000; Tremaine et al., 2001); some workers, however, continue to refer to them as shed Cathedral olistoliths (Fletcher and Collins, 2003). Pratt (1995, p. 54) interpreted Monarch Fmn mounds as patch reefs, noting that “in situ and tumbled patch reefs occur seaward and downslope of ooidal grainstone shoals of the Cathedral Fmn of the southern Rocky Mountains,” and envisaged the mounds to have grown in water depths of ∼50 m. As the Cathedral Escarpment was initially up to 250 m in height, and that the ‘lower’ portion of the Monarch Fmn

averages 50 m in thickness, it is more likely that Monarch mounds formed well below the photic zone (∼100 m water depth). Thus, their status as patch reefs (formed mostly by photosymbiotic organisms) is suspect. Aitken (1971, Fig. 3b) described the carbonate bodies within the basin succession adjacent to the Cathedral ‘reef margin’ as generalized bioherms. This was the first interpretation of these “allochthonous” blocks of limestone as something other than spalled-off pieces of the Cathedral platform, suggesting instead that they were in situ aspects of the basin. However, despite categorizing them as mound-shaped bioherms, Aitken (1971) did not describe or discuss these in detail. Despite a paucity of large organisms in these mounds, petrographic

C.J. Collom et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 277 (2009) 63–85

71

Fig. 7. Carbonate mud mounds within the Monarch and Burgess Shale formations. A. 95 m-thick complete exposure of the Monarch Fmn [Fmf] on the south face of Mt Field; underlying calcareous facies are the Takakkaw Tongue of the Cathedral Fmn [Ftt], overlying mudstones are of the Kicking Horse Mbr of the Burgess Shale Fmn [Fbkh]. Two large, fractured, partly overturned mounds (indicated by ‘M’) are visible near the middle of the Monarch Fmn. B. ∼10 m wide carbonate mud mound within the upper Monarch Fmn, 2nd gully northwest of the Trilobite Beds on Mt Stephen (person for scale: circle). Close-up of external surface in Fig. 9B. C. Mound M-2a/2b of the Yoho River Mbr, approximately 100 m from the renowned Trilobite Beds (to lower left): this ∼ 7.0 m tall structure is formed from two connected mud buildups and has steeply-dipping flanking beds (N55°). Basinal strata in the vicinity of the Mount Stephen Trilobite Beds have been further inclined (to 45–60°) by Cordilleran regional uplift and deformation. D. Mound M-4 of the Wash Limestone Mbr, ∼40 m from the Cathedral Escarpment, northwest ridge of Mt Stephen. This stratigraphic horizon is correlative to the exposed Wash Mbr below the abundantly-fossiliferous Walcott Quarry on Mt Field (4.3 km from this site).

analysis reveals abundant microscopic carbonate-secreting biota (Powell et al., 2006: Fig. 7D) including micrite-producing Epiphyton in the Monarch Fmn by McIlreath (1977a: pl 7–4), Pratt (1995: Fig. 5), and Aitken, (1997: Fig. 42). Stewart (1991: pl. 32A) illustrates typical Renalcis in Monarch Fmn carbonates from Mt Stephen. Bija sibirica bacterial microfossils in ‘Cathedral Escarpment debris’ have been mistaken for bryozoan colonies (McIlreath, 1977a: pl. 7–5E,F,G); the earliest bryozoans do not appear in the fossil record until the Ordovician. The presence of Epiphyton and Renalcis in mounds of the Monarch and Burgess Shale formations is significant, particularly since these microfossils are absent in typical shallow-water bioherms and patch reefs. Mud mounds of the Monarch Fmn vary in size, from 2.5 m to nearly 10.0 m in height (Fig. 7A,B; Table 1). Approximately one-third appear to be overturned from their initial orientation, as indicated by the presence of their rounded tops on the underside of the mound in outcrop. Numerous mounds within the Vermilion Subunit and Duchesnay Unit of the Bolaspidella Zone (associated with the Eldon Escarpment) also exhibit capsized orientations. Detailed descriptions of the Monarch Fmn mounds have been made at the following localities, and the number of discrete mounds and isolated mound fragments N2.0 m and the maximum size recorded at the following localities. While micrite comprises the bulk of Monarch Fmn mounds, dendritic microbial structures are common in thin-section. These most closely resemble Epiphyton and Renalcis-like structures, and are likely biological in origin rather than diagenetic growths. They have been reported in Recent ‘towers’, ‘chimneys’, and other carbonate buildups associated with hydrothermal fluid seepage in freshwater lakes (Laval et al., 2000, Fig. 4; see also Buchardt et al., 2001). Mat-like features consisting of these and related microbes have also been recorded in

authigenic carbonate mud buildups on serpentinized basalt, some 15 km from the active Mid-Atlantic spreading ridge (Kelley et al., 2001; Von Damm, 2001). Though these analogs have considerable morphological differences with mounds from the Monarch Fmn, the facies and isotopic signature of both Recent and Cambrian mounds share too many similarities to be merely coincidental. Flat-bottomed cavities indistinguishable from reports of similar ‘Stromatactis’ in the literature (Devonian examples in Aitken et al., 2002) also occur in Monarch Fmn mounds. These cavities are filled with either rimmed cements that vary in colour from white to bright orange (McIlreath, 1977a: pl. 7–3C,D,E,F,G), or flow debris including ooid sand and dolomitic ‘chipstone’ (Stewart, 1991: pl. 22e). These shallow-water sediments are interpreted to be platform-derived, transported to the basin floor by tempestites and gravity-flow. Where there is abundant ooidal material (2nd Gully, Mt Stephen: Fig. 9B, same mound as shown in Fig. 7B), it fills typically 1–3 cm wide and 1–2 cm high stromatactislike cavities. The presence of ooidal sediments to depths of ∼1.0 m inside mounds suggests these cavities formed an open network prior to burial by basal Burgess Shale Fmn mudstone. Though the genesis of Stromatactis cavities remains controversial, evidence from the Monarch Fmn mounds (and two mud mound-bearing horizons in the Burgess Shale Fmn) supports an origin via dissolution. Hydrothermal brines associated with mound formation and precipitation of other authigenic deposits (e.g., ‘exhalite’: Powell et al., 2006) may have been implicit in the partial disintegration of the host micrite-dominated mounds. 2.3. Monarch Formation biostratigraphy Biostratigraphic control of the Monarch Fmn is established by uncommon, but not rare, trilobites of the Glossopleura boccar Zone

72

C.J. Collom et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 277 (2009) 63–85

Fig. 8. Olistolith of Cathedral Fmn in the basinal Burgess Shale, Mt Stephen (see also Fig. 4B). White line denotes the margins of the block, the exposed height of which is ∼40 m. Black lines indicate orientation of peritidal bedding surfaces within the block; note these are oriented perpendicular to the relatively horizontal bedding of the surrounding Kicking Horse Shale Mbr. This is the only known example of shed megabreccia of Cathedral platformal facies, having toppled into the basin N1.0 Myrs after the megatruncation event that formed the Cathedral Escarpment. Person in left photo, circled, for scale.

(Rasetti, 1951; Fritz, 1971). Few reports of trilobites from the Monarch Fmn (= ‘upper’ Tak Tongue) have been published in the half-century since Rasetti (1951). Stewart (1991, p. 444) reported the occurrence of Glossopleura sp., Kootenia sp., and the articulate brachiopod Nisusia sp. (GSC C-167278) from the lower part of the Monarch Fmn at the type locality (DS-27 section). Although these species occurrences establish a specific biozone, the zonation scheme applied to the Canadian Cambrian has long been without a stage-level framework. Palmer (1998) has argued that a significant portion of what has traditionally been referred to as ‘Middle Cambrian’ should now be assigned to the Delamaran Stage (Albertella and Glossopleura Zones) and subsequent Marjuman Stage (Bathyuriscus/Ehmaniella and Bolaspidella Zones). Formalization of stage-level names within the Cambrian on an international scale has been an elusive goal for decades, but we use these names here to recognize the trilobite faunas of the Laurentian marine realm and to further stabilize such chronostratigraphic

nomenclature. At present there are numerous locally-applied regional biozone schemes for the Cambrian of Laurentia, but none has yet been consistently applied to the Burgess Shale and contiguous strata of Western Canada. As indicated above, corynexochidid trilobites of the Glossopleura lineage occur sparsely within the ‘lower’ portion of the Monarch Fmn. Represented by G. boccar (Walcott), G. templensis Rasetti, and G. mckeei Resser (Rasetti, 1951), these age-diagnostic species permit detailed correlation between the platform and basin. Argillaceous beds of the lower Monarch Fmn are correlative to the lower cycle of the Narao Mbr, Stephen Fmn, among the most ichnofossil-rich sedimentary units of the Sauk II Megasequence. At Castle Mountain, in Alberta (Figs. 4,9D), the Narao Mbr contains well-preserved walking tracks of trilobites, arthropods and vermiform burrowing organisms (Keiran, 2003, p. 22–23). Ichnofossils are considerably rarer in the basinal facies; low-relief and relatively simple Cruziana

C.J. Collom et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 277 (2009) 63–85 Table 1 List of carbonate mud mounds from the Monarch Formation (new), British Columbia. Locality

# of mounds

Max. height

The Monarch (cirque: south side) The Monarch (cirque: north side) Natalco Lake (east-facing cliff) Mount Stephen (north gully to fossil gully) Mount Stephen (1st gully to 5th gully) Mount Field (escarpment to Fossil Gully Fault) Mount Field (‘shoulder’ section)

3 mounds 2 mounds 2 mounds N8 mounds N5 mounds N5 mounds N3 mounds

largest = 7.0 m largest = 2.5 m largest = ∼ 4.0 m largest = ∼ 5.0 m largest = ∼ 10 m largest = ∼ 4.0 m largest = ∼ 4.0 m

Note that ‘maximum size’ refers primarily to height, not diameter at base.

Assemblage traces were documented, but are not illustrated herein. Paleoenvironmental control on the spatial distribution of key trilobite species cannot be ruled out, but their temporal longevity is demonstrably not diachronous. Biostratigraphic zones and zone boundaries within the study interval, including the overlying Bathyuriscus-Elrathina Zone, correspond closely to facies changes and transgressive sequence stratigraphic surfaces. 2.4. Takakkaw Tongue — historical perspective The Takakkaw Tongue of the Cathedral Fmn is an inconsistently applied geological name, despite its history of usage and association with the Burgess Shale. Since its introduction by Fritz (1971), the Takakkaw Tongue (a.k.a. “Thin Cathedral”; McIlreath, 1977a) has been

73

variably applied to both the slope equivalents of the Cathedral Fmn (Albertella bosworthi Zone) and fine-grained basinal facies of the early Glossopleura boccar Zone (Stewart, 1991; Aitken, 1997). This latter application of the Takakkaw Tongue (hereafter abbreviated as ‘Tak Fmn’) implies deposition at the foot of the N200 m high Cathedral Escarpment, or after events leading to its formation. Until 1991, Tak Fmn strata had only been described from the Kicking Horse River valley near Field, British Columbia (McIlreath, 1977a), but Stewart (1991) expanded use of the term southwards to exposures at The Monarch and Natalco Lake. The age of Tak Fmn is established by sparse trilobite fossils of the late Albertella Zone and lowest Glossopleura Zone (Fig. 2). In the Canadian Lexicon (Glass, 1990), Aitken characterized the ‘Tak Tongue’ as: (1) having a type locality (242.0 m thick) on the lower north slopes of Mt Stephen, Yoho National Park; (2) consisting of cliffforming limestone (mainly ribbon-bedded lime mudstone, pellet and skeletal wackestone, minor grainstone), and derived dolomite; and (3) representing a “westward projection of the lower part of the Cathedral Fmn.” Furthermore, the Lexicon definition (p. 625) specified that the lower Cathedral was indeed “…the part pre-dating the Cathedral Escarpment (reef)” and “represents slope deposits.” Stewart (1991) recognized strata on either side of a distinctive stratigraphic contact (or megatruncation surface) as comprising the Tak Fmn, partly as a compromise to earlier confusion regarding the relation of slope facies to the escarpment. Use of Takakkaw terminology for deposits above the ‘e1’ MTS by Aitken (1989) may have influenced Stewart

Fig. 9. Sedimentary features of the Monarch Formation and coeval facies. A. Monarch Cirque (type locality): thin-bedded calcareous mudstones of the lower Monarch Fmn [Fmf] covering in situ cavern-filling megabreccia of the Cathedral Fmn [Fcl-mb]. Person at lower right is standing next to a ∼10 m wide ‘olistolith’ of dolomitized Cathedral facies cemented into indurated breccia that formed before platform margin collapse. B. Ooidal cavity-fill on the outer surface of Monarch Fmn carbonate mud mound shown in Fig. 7B. Fill material likely shed from nearby Cathedral platform. Abundant Stromatactis-like cavities (arrows indicate margins) formed via dissolution of authigenic mound micrite by caustic brine fluids. Coin is 20 mm in diameter. C. Detail of Wash Limestone Mbr, Walcott Quarry, Fossil Ridge. Light coloured limestone intraclasts lack fossil content, are in darker bioclastic carbonate matrix, and have isotopic signatures similar to those of mud mounds. Darts indicate larger irregular shaped pieces. D. Platformal facies coeval to the Monarch and Burgess Shale formations; Castle Mountain, Alberta. Narao Mbr of the Stephen Fmn [Sn] contains a lower ‘shoaling-upward’ sequence interpreted as coeval to the Monarch Fmn of the rifted margin, ∼42 km to the west. In contrast to deeper-water correlatives, Narao Mbr calcareous facies (Glossopleura Zone) are bioturbated and contain shallow-water ripples and cross-bedding. The Waputik Mbr of the Stephen Fmn [FSw] is primarily recessive mudstone, but has a thin concretionary zone (“Alpineclub Bed”; red dashed line) correlative with level of Wash Limestone Mbr (Bathyuriscus Zone) in the deep basin. Contact with overlying Eldon Fmn is gradational.

74

C.J. Collom et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 277 (2009) 63–85

(1991, Figs. 54,55), who labeled bedded and purportedly brecciated facies on either side of the MTS as ‘Takakkaw Tongue’ and associated with the Cathedral Escarpment. More recently, Fletcher and Collins (2003) recommended formally that the ‘Takakkaw Tongue’ be raised to formational status, which we support and follow herein. The Takakkaw Fmn is characterized by deep water slope facies, whereas the Cathedral Fmn is characterized by shallow water platformal facies. In areas where no transitional contact can be observed easily between the Cathedral and Takakkaw formations (due to dolomitization – e.g., Mt Field and Mt Stephen), an arbitrary vertical line extended down from the Cathedral Escarpment shall be considered sufficient (with all strata basinward of this line being assigned to the Takakkaw Fmn). A similar artificial ‘dividing line’ extends upwards from the vertical e1 surface, separating the otherwise continuous Stephen Shale (platform) from the coeval Burgess Shale (basin). Geographical and nomenclatural separation of the Cathedral Fmn from Tak Fmn below the e1 MTS coincides with the outboard margin of a major un-named half-graben salient of the Kicking Horse Rim. Accordingly, facies changes and corresponding lithostratigraphic changes occur along essentially the same “divide” or boundary in both underlying strata (e.g., Mt Whyte Fmn versus Naiset Fmn) and overlying strata (e.g., Eldon Fmn versus Tokumm unit). 2.5. Takakkaw Tongue — Bio- and lithostratigraphic framework Basinal facies of the Monarch Fmn, in contrast to those of the underlying Tak Fmn, do not contain Albertella Zone guide fossils. Dominant trilobites of the Monarch are Glossopleura Zone indicators (e.g., Polypleuraspis insignis), the biozone succeeding the Albertella Zone. This poorly-recognized biozone boundary dichotomy has lured numerous workers to extend upwards the depositional history of the Cathedral Fmn to at least mid-Glossopleura Zone time (fossils of which are restricted to the Trinity Lakes Mbr and highest Cathedral; Fig. 2 herein). Older concepts of the so-called “Boundary Limestone” (McIlreath, 1977a; now known as the Yoho River Mbr, Burgess Shale Fmn) argued that these carbonates were a syndepositional basinward extension of the youngest Cathedral. However, it can be conclusively demonstrated that both the Yoho River and Wash Limestone Members of the Burgess Shale Fmn terminate against the Cathedral Escarpment, and thus post-date its formation (Stewart, 1991). This is further confirmed by the differing trilobite assemblages in each unit. In order for the Kicking Horse Shale Mbr (underlying the Yoho River Mbr) to be coeval to upper Cathedral Fmn carbonates, these heterogeneous units would necessarily represent lateral facies of each other. But the Kicking Horse Mbr is not composed of slope debris or fine-grained ‘peritidal’ carbonate deposits — as would be expected if they were adjacent age-equivalent to the Cathedral Fmn. The expected transition between these units would be rather gradational (were they coeval) instead of sharp, as is observed against the e1 MTS on Mt Stephen. Such abrupt facies changes over narrow transition zones are seldom (if ever) observed in Recent analogs, although vertical escarpments bordering carbonate atolls and platforms are not uncommon (Hine et al., 1992). Slope deposits of the Tak Fmn bear little resemblance to basinal mudstones of the Monarch and Burgess Shale formations. 2.6. Burgess Shale Formation — Yoho River Limestone Member The lowest carbonate interval or ‘wedge’ of the Burgess Shale Fmn has thus far only been identified within the Kicking Horse Pass area (Fletcher and Collins, 1998). In exposures on both Mt Stephen and Mt Field the Yoho River Limestone Mbr extends for a relatively short distance from the escarpment (b500 m). We have identified and documented no fewer than four carbonate mud mounds in the Yoho River Limestone (Collom, 2000; Johnston and Collom, 2001), and an equal number of large mound facies fragments, measuring up to 2.5 m in length, within the basal Campsite Cliff Shale Mbr on Mt Stephen. At least two of these mounds are located b75 m from outcrop of the

world-renowned Ogygopsis Trilobite Beds (Fig. 7C), now a UNESCO World Heritage Site, and b15 m from the Cathedral Escarpment (Powell et al., 2006: Fig. 6a). Yet these mounds remained unnoticed as such for 112 years — since R.G. McConnell became the first geologist to visit and study the site in 1886. It was during the 1998 field season that we first recognized their presence. Fletcher and Collins (2003) incorrectly attribute Cathedral megabreccia status to these same mounds, though their internal facies bear no resemblance to typical Cathedral bedding. We discovered closely-spaced to anastemosed Yoho River Limestone Mbr mounds in a measured section on the ‘northwest shoulder’ of Mt Stephen in 2002 and 2003 (Fig. 3B); an excellent but remote section accessible only by helicopter. Fragments of Yoho River Mbr mounds exceeding 50 cm in size are also present at the ‘shoulder section’ of Mt Field, across the Kicking Horse Valley. 2.7. Burgess Shale Formation — Wash Limestone Member The second carbonate unit of the Burgess Shale Fmn is between the monotonous and irregularly fossiliferous mudstones of the Campsite Cliff and Walcott Quarry members. The Wash Limestone Mbr does not extend as far from the Cathedral Escarpment and is relatively thinner than the older Yoho River Mbr. Here again the remote ‘northwest shoulder’ section on Mt Stephen proves important: it is here — not at the better-known Walcott Quarry on Fossil Ridge — that we found mud mounds within the Wash Mbr (Fig. 7D). At least four discrete mounds were documented; the largest being 3.5 m in height, and ∼4.0 m in width at the base. Across the valley, brecciated angular clasts and flat intraclasts of microbial micrite of the typical mound lithology are present in the lowest beds exposed in the R.O.M. excavations of the Walcott Quarry (Fig. 9C), ∼3.0 m below the Phyllopod Bed. Even though no mounds are visibly exposed on Fossil Ridge, facies consistently associated with mounds are preserved both here and across the valley on Mt Stephen, indicating that mud mounds were present at the time of Wash Limestone deposition along at least 2.5 km of the Cathedral Escarpment. Exceptional preservation of soft-tissued fossils, such as worms (e.g., Ottoia) and non-calcified sponges (e.g., Vauxia), occur in profusion within the dark mudstones directly above the Wash Mbr on both Mt Stephen and Fossil Ridge — suggestive of a link between the dense occurrence of fossilized invertebrates and unusual sedimentary features (mud mounds and Mg-rich exhalite; Powell et al., 2006). The repeated association of soft tissue preservation, Mg-rich precipitates (exhalite), and underlying mud mounds is observed in the Wash Limestone–Walcott Quarry Shale interval, and in the underlying Yoho River Limestone–Campsite Cliff Shale interval. A third such fossil and unusual facies association is herein documented for the new Monarch Formation–Kicking Horse Shale interval. Two additional such associations are known by us from overlying basinal deposits of the Vermilion and Duchesnay units (Bolaspidella Zone; Stewart, 1991; Motz et al., 2001), but will be described elsewhere. 2.8. Other mound-bearing units/Ancient and Recent analogs A total of seven discrete stratigraphic horizons having carbonate mud mounds exist within the Delamaran and Marjuman stages (Glossopleura to Bolaspidella zones) in the study area (Fig. 1). Their occurrence appears to be genetically linked to the abrupt, hydrothermally-characterized platform margins of the Cathedral and Eldon formations. The mound-bearing horizons are found in the upper slope deposits from the Takakkaw Tongue up to the Middle Chancellor Group (McArthur and Duchesnay units of Stewart, 1991), as shown in Fig. 6. With the possible exception of a single observed mound from Eldon Fmn carbonates above the Field Mbr (Stephen Cirque locality), no mud mounds having nucleated and grown directly upon Laurentian carbonate platforms are known to us. In addition, mounds have not been documented to date in the Mt Whyte or Naiset

C.J. Collom et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 277 (2009) 63–85

formations. Domal carbonate bodies indistinguishable from mud mounds have been observed in the Tak Fmn (Assiniboine side of The Monarch, B.C.), the Tokumm Subunit (slope equivalents of the Eldon Fmn), and just above the Field Mbr — in platform margin carbonates of the ‘upper’ Eldon Fmn (Stephen Cirque; near locality #4 in Fig. 1). These mound-rich intervals within the McArthur Unit and Duchesnay Unit (as defined by Stewart, 1991) are complex and deserve separate attention. Occurrences of carbonate ‘microbialite’ mounds and conical buildups in both marine (Kelley et al., 2001) and freshwater settings in Recent analogs (Pavilion Lake, BC: Laval et al., 2000) strongly argue against their origin being linked to geochemical or biological processes unique to either environment. Benthic exhalative environments (seeps) and associated carbonate/siliceous precipitates occur wherever fluid-bearing faults intersect either a lake bed, or the seafloor. Such faults can be demonstrated to be statistically random in distribution relative to potential water cover (as a function of geographic location), and are predictably concentrated at the base of subaqueous escarpments (e.g., Florida Escarpment: Paull et al., 1984; Paull and Neumann, 1987). Such deep-penetrating faults were also present at the base of the Cathedral Escarpment, contributing to both MTS-formation and subsequent mound growth (Fig. 1). Concurrently, the variable chemistry of brines can contribute to dissolution of carbonate sediments; Van Balen and Cloetingh (1993) describe flow of ‘meteoric’ water (= hydrothermal) inducing secondary porosity in mixed dolostone–limestone host rocks. At a sufficiently large scale the structural integrity of the carbonates being corroded by hydrothermal dissolution will eventually be compromised. Localized buildups of exhalite and authigenic carbonate proximal to paleo-escarpments suggest similar depositional settings (Powell et al., 2003, 2006). Perhaps it is no coincidence that all three Burgess Shale examples cited here (in addition to the Sirius Passet biota, Buen Fmn, N. Greenland: Conway-Morris et al., 1987; Ineson and Peel, 1997) were located along extensional margins of Laurentia. The Lower Cambrian Qiongzhusi Fmn, China has Burgess Shale-Type preservation of softbodied organisms, but is not associated with a paleo-escarpment (Steiner et al., 2001). These renowned metazoan fossils, collectively known as the ‘Chengjiang Biota’, share numerous taxonomic and preservational similarities to the Burgess Shale biota, and have been linked to the close stratigraphic occurrence of underlying barite-rich exhalative deposits (Steiner et al., 2001). Future geophysical research on the tectonic history and basement structures of Southwest China Platform of Yunnan Province will shed light on the inferred link between hydrothermal fluid flow and the presence of diverse and abundant Chengjiang fossil biota. 3. Sequence stratigraphic framework 3.1. Introduction Although sequence stratigraphy has been widely utilized as an interpretive tool in the earth sciences since the late 1970s (particularly the hydrocarbon exploration sector: Van Wagoner et al., 1988) these concepts have only been applied recently to Early Paleozoic strata of Laurentia (Glumac and Walker, 2000). The Cambrian Period of Western Canada, however, remains largely uncharted territory with respect to modern sequence stratigraphic methodology (Spencer and Demicco, 2002). Aitken (1966) applied the concept of “Grand Cycle” to the Middle Cambrian to Lower Ordovician interval of the Canadian Rocky Mountains, each defined as comprising a lower shale-rich and an upper carbonate-rich hemicycle. Of the eight cycles identified by Aitken (1966) the Mt Whyte and Cathedral formations comprise the oldest Grand Cycle, and the Stephen and overlying Eldon formations the next oldest cycle. These lithological oscillations are undoubtedly several millions of years longer in duration than typical third-order sequences (1.0–3.0 Myrs average), and may be more accurately compared to second-order cyclothems (Kauffman, 1979). Third-order

75

classic sequences may be best equated to ‘Rhythms’ of Aitken (1966, although none is defined from either the Mt Whyte–Cathedral or Stephen–Eldon Grand Cycles), and classic system tracts of third-order sequences to Aitken’s so-called ‘sub-cycles.’ At even larger chronological timescales are the first-order megasequences of Sloss (1963, named after North American native cultures), of which the “Sauk Sequence” encompasses the Cambrian Period. Because each ‘Half-cycle’ of a Grand Cycle contains more than a single third-order sequence, we find it impractical to continue use of this somewhat obsolete concept. Grand cycles are generally many hundreds of metres in thickness and may represent large-scale alterations between carbonate-dominated and siliclastic mudstonedominated intervals, during a time when the Earth's oceanic thermal and geochemical composition fluctuated wildly between extremes. But they do not reflect relative sea level change on a scale that permits high-resolution and meaningful correlation between strata deposited in different regions of western Laurentia. Nonetheless, the term ‘Sauk Sequence’ (of Sloss, 1963) is ingrained in the literature, and does refer to a recognizable (albeit rather thick) package of sedimentary rocks that form much of the Central Ranges within the contiguous national parks of the Canadian Rocky Mountains in Alberta and British Columbia. Sloss (1988) later refined his cratonic-wide megasequence scheme, dividing several of these into tripartite sub-sequences. The portion of the Sauk that includes the Monarch and Burgess Shale formations falls within the ‘Sauk II’ – from approximately the base of the Dyeran Stage (late Early Cambrian) to the Dresbachian–Franconian Stage boundary (early Late Cambrian). This concept was used in the title of this paper, as it is the only sequence of sufficient duration to include all of the Delamaran and Marjuman sequences addressed. 3.2. Discussion and new model We present here interpretations of the third-order sea level history of the ‘Middle’ Cambrian for the study area in Alberta and British Columbia (Fig. 1), addressing platformal and basinal strata deposited during the interval spanning the Albertella bosworthi Zone to Bathyuriscus rotundatus Zone. Concepts used are based primarily on the emerging sequence biostratigraphy paradigm (Beauchamp and Henderson, 1994; Collom and Kravec, 2000a; Aitken et al., 2002). Supported by a plethora of high-resolution biostratigraphic data from throughout the Phanerozoic, sequence biostratigraphy asserts that synchronous changes among key invertebrate species and formation of short-term disconformities (sequence boundaries) or condensed zones (maximum flooding intervals, separating transgressive and regressive deposits) are not coincidental (Brett, 1995; Brett, 1998); the fauna evolve in concert with sea level changes as a result of cyclical variation in sedimentation rates, nutrient influx, intra- and interspecific competition for niche space, and availability of sunlight (paleobathymetry). Although this study concentrates on the Delamaran and Marjuman successions in the southern Rocky Mountains, the sequences outlined below can also be recognized throughout the Canadian Rockies — as far north as the Kechika Trough in northern British Columbia (Post, 2001). All four biostratigraphic boundaries of the ‘Middle’ Cambrian succession discussed herein are located at, or very near, important facies changes, corresponding to relative sea level rise (transgression) or sea level fall (regression). These will be referred to as Transgressive System Tracts (TST) and Regressive System Tracts (RST); the two major building components of third-order sequences. ‘Lowstand System tracts’ (LST) deposits were not recognized or defined, although some of the massive debris-filled channels bordering the Cathedral platform (e.g., Mummy Lake, B.C.) may be examples, as they occur where there is abundant accommodation space and paleobathymetry necessary for lowstand fan/wedge preservation. In general, transgressive system tracts (TST) in the study area are characterized by deposition of carbonate facies, whereas regressive

76

C.J. Collom et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 277 (2009) 63–85

system tracts (RST) are predominantly comprised of argillaceous or siliciclastic facies. In the following definitions, the term ‘basal unconformity’ is used in lieu of “sequence boundary”, although they are considered to be equivalent. Nearly all of these unconformities are disconformities, representing sediment bypass and/or erosional intervals of relatively short duration. Although estimates vary considerably, Phanerozoic marine disconformities which juxtapose RST and TST deposits of consecutive sequences encompass less than 100 Kyrs — or a typical fourth-order Milankovitch eccentricity cycle (Posamentier and Vail, 1988). All absolute age dates used and assigned here are tentative, and require additional research to confirm. The terms ‘sequences’ and ‘cycles’ are used interchangeably, and parasequences (the building blocks of system tracts) were not delineated in outcrop, but have been for coeval strata in the Great Basin of the United States (Montañez and Osleger, 1993). Nonetheless, cyclical bedding at several levels and thicknesses is recognizable within much of the Middle Cambrian of Western Canada. 3.3. Delamaran third-order sequences Nearly all stages of the Phanerozoic have formal 3-letter abbreviations (Harland et al., 1989), with the exception of contentious periods such as the Cambrian. Little international agreement presently exists regarding the adoption of a single set of Cambrian stage names. Until such decisions are reached, we shall use the following Laurentian stage names and abbreviations (from Palmer, 1998): Delamaran = DEL, and Marjuman = MAR. Regional correlation of the DEL-4 (part), DEL-5, DEL– MAR, MAR-1, and MAR-2 (part) sequences are illustrated in Fig. 4. SB = sequence boundary (third order), MFI = maximum flooding interval. 3.3.1. DEL-1 sequence This eustatic cycle occurred during renewed rifting and normal faulting along the Laurentian margin (Fritz and Mountjoy, 1975; Lickorish and Simony, 1995), and comprises the poorly-understood Peyto Fmn/Hota Fmn interval (upper Gog Grp) and Mt Whyte/Naiset formations. These formations, below the Cathedral Limestone, are not shown in Fig. 4. The transgressive portion of DEL-1 (Peyto Fmn) is within the Bonnia-Olenellus Zone, so part of this eustatic cycle is Dyeran Stage in age. The regressive portion (Mt Whyte) is within the Plagiura-“Poliella” Zone, the lowest of three established zones of the Delamaran Stage. Stratigraphic matters requiring clarification, such as the spatial and temporal relations of the Mt Whyte and Naiset grabenhosted sediments, are beyond the scope of this study. Thus, this sequence is only mentioned insofar as (1) it is the initial cycle of the Delamaran Stage, and (2) the upper unconformity defines the base of the DEL-2 Sequence (during which time widespread carbonate platform deposition was initiated in the Kicking Horse Pass region). 3.3.2. DEL-2 sequence Definition: basal unconformity at or near the Mt Whyte Fmn– Cathedral Fmn contact, within a 15–25 m interval of pervasive ooidal packstones and increasingly common ribbon dolomicrite. The TST includes the lower portion of the Cathedral Fmn below the Ross Lake Mbr (latest Plagiura Zone; Aitken, 1997); the RST consists of the more argillaceous, thin-bedded carbonates of the Ross Lake Mbr (Albertella Zone). The Maximum Flooding Interval (a.k.a. the Condensed Interval) is tentatively placed near the base of the Ross Lake Mbr, as the MFI appears to delineate the transition from massive carbonates to relatively more siliciclastic deposits of the Ross Lake calcareous mudstones. As the lower of two easily-recognized mudstone ‘tongues’ within the Cathedral Fmn (Aitken, 1997: Figs. 43,52), the Ross Lake Mbr is also distinguishable within the Snake Indian Fmn in eastcentral British Columbia, in and near Jasper National Park. This shaly unit or ‘tongue’ represents a fall in relative sea level, and basinward progradation of the Laurentian paleoshoreline — then located near the Alberta–Saskatchewan provincial boundary.

3.3.3. DEL-3 sequence Definition: basal unconformity near the top of the Ross Lake Mbr; the TST includes the cliff-forming Cathedral Fmn platformal facies below the Trinity Lakes Mbr. The RST consists of the thin-bedded, argillaceous Trinity Lakes Mbr, which (much like the Ross Lakes Mbr) represents sea level fall and renewed shoreline progradation. The MFI is assigned near the base of the Trinity Lakes Mbr, as the MFI delineates the transition from carbonates to more siliciclastic deposits. The DEL-2 and DEL-3 sequences are both characterized by relatively thick transgressive system tracts that reflect high deposition rates, and considerably thinner regressive system tracts reflecting, in part, decreased carbonate sedimentation. Numerous authors (e.g., Sarg, 1988; Schlager, 1992) have stressed that terrigenous sediment input must be minimal or curtailed for carbonate micrite production, resulting in the growth of platforms, ramps, reefs, and mud mounds. Assigning regressive status to the Ross Lake and Trinity Lakes members simply recognizes their greater content of non-carbonate, continentally-derived sediments. These units extended into the study area from the east and north (Fritz and Mountjoy, 1975; Aitken, 1997) during relative sea level fall. Parasequence stacking during RST should exhibit offlap and basinwarddescending clinoforms, but this is difficult to establish with the dissected, remote, often vertical nature of exposures in the study area. 3.3.4. DEL-4 sequence Definition: basal unconformity located just above top of the Trinity Lakes Mbr; the TST includes the cliff-forming Cathedral platformal facies above the Trinity Lakes Mbr to the upper lithostratigraphic contact of the Cathedral Fmn (Fig. 4). The RST consists of well-bedded, calcareous basinal mudstones, carbonate mud mounds, and isolated lenses of coated grain facies of the ‘lower’ and basal ‘upper’ units of the Monarch Fmn (Fig. 3C), and lower cycle of the platformal Narao Mbr (Stephen Fmn: Figs. 2,7A,B,9D). The MFI is located at the top of the Cathedral Fmn (Cathedral–Stephen contact) and coeval slope facies of the Tak Fmn, and represents termination of platform build-up during the early Glossopleura Zone. During this interval of maximum flooding and sediment bypass, extensive dissolution and brecciation of the Cathedral platformto-slope transitional facies intensified (Stewart et al., 1993). A combination of localized tectonism and ascending fault-controlled hydrothermal brines associated with the Proterozoic rift margin initiated a series of processes (Fig. 5) culminating in the failure of the Cathedral margin along much of its length in the Yoho-Kootenay parks region. Thus, the Cathedral Megatruncation Event, from which the Cathedral Escarpment was formed at ∼509 Ma BP, occurred during the transition from thirdorder eustatic transgression to regression. Maximum flooding was essentially coincident with margin collapse. Accordingly, the initial regressive muddy sediments deposited on the MTS (e1) are geneticallyrelated to the youngest platformal carbonates (sensu Van Wagoner et al., 1988; Posamentier and Vail, 1988), having been laid down during the same sequence (Figs. 4,5). Penecontemporaneous co-occurrence of such events is not fortuitous; rather, it is accounted for and predicted by sea level fluctuation models which link plate tectonics and eustatic cyclicity (Cloetingh,1988; Beauchamp and Henderson,1994; Aitken et al., 2002). A nearly identical set of circumstances prevailed at the time the younger Eldon Fmn platform experienced failure, resulting from a megatruncation event that occurred at the Bathyuriscus-Bolaspidella Zone boundary (Stewart, 1991: pl. 53,54). In this younger analogue, new deposits of the basal Vermilion subunit (early RST) are genetically-related to the TST deposits of the highest Eldon Fmn and Tokumm unit carbonates (Fig. 6). Thus, both Middle Cambrian platforms had growth terminated and were subject to large-scale mass wasting during the same portion of separate tectono–eustatic sequences. 3.3.5. DEL-5 sequence Definition: basal unconformity within the ‘upper’ unit of the Monarch Fmn, marking the transition from predominantly bedded

C.J. Collom et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 277 (2009) 63–85

calcareous mudstones and argillaceous limestones to massive authigenic carbonate and mud mounds. As with the younger mud moundcontaining limestone and dolomite units of the Burgess Shale Fmn (Yoho River and Wash Limestone members), the position of the thirdorder sequence boundary near the top of the Monarch Fmn is cryptic and difficult to define precisely. No scours or pebble/cobble lags exist at this level, that would allow easy demarcation of maximum sea level fall. Regional-scale pervasive early dolomitization has obscured the contact between Monarch carbonates and the Cathedral escarpment against which they accumulated (Fig. 3B). Evidence of a hardground or secondarily lithified surface on the outside of Monarch mounds was not recognized; early dissolution in the form of cavities on carbonate mud mound surfaces (Fig. 9B) is stratigraphically located well above the SB at the top of the DEL-4 Sequence. Transgressive (TST) deposits of the DEL-5 Sequence include the majority of the large (N8.0 m) mud mounds of the ‘upper’ Monarch Fmn. Despite the large individual size of these carbonate build-ups, the overlying RST siliciclastic mudstones are proportionately thicker than the TST deposits of this same sequence. In fact, this pattern of relatively thicker RST deposits overlying thinner mound-bearing TST deposits distinguishes this (DEL-5) and the two overlying sequences. In places where the carbonate units of the Burgess Shale cannot be recognized (e.g., at The Monarch: Figs. 3A,5F, 10D), the basin-fill deposits are simply referred to as ‘Undivided Burgess Shale.’ Regressive deposits are represented by the lowest bedded mudstones of the Burgess Shale Fmn, constituting part of the “Amiskwi Mbr” (sensu McIlreath, 1977a; Aitken, 1997) of the nowabandoned ‘Thick Stephen Fmn.’ The lithostratigraphic Amiskwi Mbr concept encompasses the lower five members of the Burgess Shale Fmn, as defined by Fletcher and Collins (1998). Inasmuch as this interval can be readily subdivided, the name Burgess Shale Fmn is used herein to the top level of the Walcott Quarry Mbr (Fig. 2). This

77

scheme allows greater precision for discussion of various carbonate and clastic stratigraphic intervals of the basin succession adjacent to the Cathedral Escarpment. As such, the RST deposits of DEL-5 are represented by bedded mudstones of the Kicking Horse Mbr. Rare unique fossil arthropod(s) have been collected from the basal ∼ 5.0 m of the Kicking Horse Mbr, most notably the chelicerate-like Sanctacaris (Briggs et al., 1994, Fig. 131, p. 170). Additionally, we have collected Anomalocaris appendages, Olenoides trilobites, and filamentous vendotaenid bacteria from this interval on Mt Stephen. In all, the Delamaran Stage in Western Canada contains five thirdorder sequences which span three consecutive trilobite biozones (Plagiura, Albertella, and Glossopleura). The DEL-1 Sequence contains trilobites of the late Bonnia-Olenoides and Plagiura Zones. DEL-2 is entirely of early Albertella Zone age, and the DEL-3 Sequence spans the Albertella-Glossopleura Zone boundary. Sequences DEL-4 and DEL-5 are of Glossopleura Zone age. The youngest portion of the Glossopleura Zone is contained within TST deposits of the next third-order cycle, which continues into the subsequent Bathyuriscus Zone (thus spanning the Delamaran–Marjuman Stage boundary). This important interval, containing the Mount Stephen Trilobite Beds and the most extensive development of carbonate mud mounds in the Burgess Shale is termed the DEL–MAR Sequence, and addressed below. 3.3.6. ‘Transitional’ Delamaran–Marjuman sequence (DEL–MAR) Definition: basal unconformity within the upper third of the Yoho River Limestone Mbr, Burgess Shale Fmn, separating interbedded thin limestones and mudstones below from more massive, thicker-bedded limestone. TST deposits consist of the large, well-defined carbonate mud mounds of the Yoho River Mbr, best observed on Mt Stephen (Figs. 5D,7C). Mound breccias from this unit were also discovered in the ‘shoulder section’ on Mt Field (Section 2 in Fig. 1). These transgressive deposits are overlain sharply by RST mudstones of the

Fig. 10. Correlations of the Stephen and Burgess Shale formations, relative to the Cathedral Escarpment. Model of Deiss (1940) and Fritz (1971) identified an unconformity along the ‘rim’ of the cathedral platform, with the Narao Mbr locally removed by subaerial erosion. Stewart (1991) also included an unconformity, but thought that the Narao Mbr passed laterally into the “reef core” of the platform margin. Aitken (1997) followed this scheme, but did not recognize an unconformity separating the Cathedral and Stephen. According to this study, the Narao and Waputik members of the Stephen Fmn are correlative with the Monarch and Burgess Shale formations in the basin, containing no more than three minor shared disconformities. The Narao Mbr is entirely younger than the upper Cathedral Fmn, and does not interfinger with any portion of the underlying platform.

78

C.J. Collom et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 277 (2009) 63–85

Campsite Cliff Mbr. The basal ∼ 5.0 m of these laminated siliciclastic mudstones, among the most fossiliferous in the Cambrian of Laurentia, contain the ‘Mount Stephen Trilobite Beds’ (Yoho National Park). Many thousands of specimens of Ogygopsis klotzi and Olenoides serratus are known from these beds, often occurring in dense concentrations; ten or more trilobites on a 1.0 m2 bedding plane slab is common. Anomalocaris appendages, Byronia tubes, and inarticulate brachiopods are also common to abundant. These biota have compelling similarities with the Sanctacaris Beds below and the Phyllopod Bed above, in terms of taxonomic composition and stratigraphic/paleoenvironmental position within the Burgess Shale. The maximum flooding interval/condensed section of the DEL– MAR Sequence is coincident with the Glossopleura-Bathyuriscus biozone boundary (Fletcher and Collins, 1998). On Mt Stephen the DEL–MAR condensed interval comprises a layer of dense, black, finelylaminated to massive exhalite (Powell et al., 2003, 2004, 2006). The Mg-rich exhalite layer is less than 10 cm thick, and appears to coat all surfaces of the upper contact of the Yoho River Mbr carbonates. This deposit/precipitate also fills syneresis cracks on the outer surfaces of Yoho River mud mounds (Fig. 7C). On the Cathedral platform the correlative horizon is located at the lithostratigraphic contact between the Narao and Waputik members (Figs. 4,9D). Although the Campsite Cliff Mbr and coeval portion of the Waputik Mbr appear to be equal in thickness in our regional correlation (Fig. 4), the scale of the Mt Bosworth and Castle Mountain section (platformal) as shown is actually half that of Mt Field and Mt Stephen (basinal). This approximate 2:1 sediment thickness ratio between basin and more condensed platform is relatively consistent for nearly all of the Middle Cambrian mudstones in the study area (Fig. 6). 3.4. Marjuman third-order sequences 3.4.1. MAR-1 sequence Definition: basal unconformity in the upper one-third of the Wash Limestone Mbr, separating interbedded thin limestones and mudstones below from massive intraclastic limestone. TST deposits consist of debris flow deposits and carbonate mud mounds, best developed on Mt Stephen (Fig. 7D). Fletcher and Collins (1998, p. 426) considered the Wash Mbr “to lie conformably between the Campsite Cliff Shale Mbr and the Walcott Quarry Shale Mbr.” The predominance of marine debris flow, scour, and calcareous micrite lithology suggest instead a depositional hiatus and facies transition from regression (Campsite Cliff Mbr) to transgression (Wash Mbr). In addition, bedded carbonates are generally absent from siliciclastic mudstone basinal facies deposited during relative sea level fall. Thus, we propose a thirdorder disconformity near the base of the intraclastic and moundbearing facies of the Wash Mbr. MAR-1 transgressive deposits on the dormant Cathedral platform are relatively thinner (b1.5 m) than the coeval Wash Limestone, but are also characterized by rip-ups and carbonate intraclasts. This thin interval is well-indurated and contains angular debris showing clear evidence of shallow-water storm scouring and associated erosional activity. Located approximately near the middle of the Waputik Mbr (Fig. 4), this bed is sufficiently distinctive to merit a formal frame of reference. It is herein informally named the “Alpineclub Bed”, after the Alpine Club of Canada, that maintains a spartan climber's hut on Castle Mountain near the site of the measured section. Unlike the coeval Wash Limestone in the basin, the Alpineclub Bed and adjacent facies are bioturbated, predominantly by shallow infaunal organisms producing Planolites-like burrows. Regressive deposits of MAR-1 are represented by the Walcott Quarry Shale Mbr; regarded by authorities as the most important invertebrate fossil-bearing sedimentary rocks anywhere (Gould, 1989). The preserved soft tissues of these animals and associated sponges are an exceptional window into Early Paleozoic evolution, captured during a protracted transition of the major phyla into continental shelves from

deep water environments (Johnston and Collom, 2001). Within the basal 5.0 m of this laminated mudstone unit Charles Walcott and his Smithsonian crew began excavating the now legendary quarry. From this b2.0 m interval (referred to here as the “Phyllopod Beds”) at least 125 species of segmented animals, sponges, and algae have been described during the century following the initial discovery (Whittington, 1985). Above the Walcott Quarry are three other fossil quarries, the Raymond Quarry (developed by Harvard University) being the next largest. An increase in carbonate sedimentation rate towards the end of the MAR-1 Sequence is reflected in its greater thickness than the underlying DEL–MAR Sequence. In turn, the overlying MAR-2 Sequence represents even higher rates of calcareous sedimentation, particularly during the ‘carbonate half-cycle’ of the Stephen/Eldon Grand Cycle (Aitken, 1966). At the outset of RST deposition of MAR-1 the Cathedral Escarpment was still ∼140 m in height (e.g., at Mt Field). Thus, even during the interval that the world-famous Burgess Shale biota was living in the vicinity of the Walcott Quarry no solar radiation penetrated to the seafloor adjacent to the Cathedral Escarpment. The TST carbonates of the DEL-5 Sequence (Monarch Fmn), DEL–MAR Sequence (Yoho River Limestone), and MAR-1 Sequence (Wash Limestone) were each deposited in marine environments devoid of sunlight, or the aphotic zone. Mud mounds and associated carbonates, primarily authigenic limestones, were thus also precipitated in this dark seafloor setting. Typical Phanerozoic reefal carbonates, bioherms and ‘mounds’ (sensu lato), conversely, owe their origin to processes within the photic zone — or bathymetries shallower than 60–70 m. Conversely, the presence of coated grain facies at the base of and at discrete horizons within the Cathedral Platform succession indicates that this portion of the study area was within the photic zone and storm wave base throughout most of its history (prior to the megatruncation event). During falling relative sea level (RST of MAR-1), the first sediments deposited were dark olive to gray mudstone. These pass upwards into interbedded mudstone and thin (N10 cm) discontinuous beds of calcareous ooids and large ovoid coated grains (avg. 1.0 cm long axis), in which non-undulose concentric laminations are evident in thin sections (pisoids sensu Young, 1989). The laminae are uniform in thickness and mineralogy (calcite, often replaced by dolomite), and are analagous to expansive ooidal deposits of the present day outer Bahamas Bank (Harris, 1979). They resemble less algal/bacteriallymediated coated grains, such as oncoids — common in intertidal environments that have abundant microbial mats and even stromatolitic structures (e.g., Abu Dhabi coast, United Arab Emirates). Accumulations of oncolitic sediments in the Stephen Fmn near the platform margin have not yet been identified in the study area, or in the region to the north, near Jasper National Park. Therefore, use of the term ‘oncolite’ by Fletcher and Collins (1998, 2003) to describe these facies is incorrect, requiring that the lithostratigraphic name ‘Emerald Lake Oncolite Mbr’ is abandoned. 3.4.2. MAR-2 sequence Definition: basal unconformity at or within the basal ∼10 m of the Eldon Fmn. The sequence boundary at this level records the gradual shallowing of relative sea level that was initiated during deposition of the basal Walcott Quarry Mbr, inasmuch as the facies transition from the Burgess Shale to Eldon carbonates is transitional over a N20 m interval. Renewed CaCO3 sedimentation during TST, as for the DEL-2 Sequence (basal Cathedral Fmn), results in accumulation of considerable thicknesses of well-bedded platformal limestone and dolomite. No clastic units (such as the Ross Lake or Trinity Lakes in the Cathedral Fmn) have yet been documented from the lower Eldon Fmn; it appears as though this entire 150–175 m carbonate interval was deposited during a single Transgressive System Tract. RST deposits of the MAR-2 Sequence are represented by dark grey to black mudstones and argillaceous limestones of the Field Mbr, present upsection of the stratigraphic intervals illustrated in Fig. 4. Although the upper Eldon

C.J. Collom et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 277 (2009) 63–85

Fmn is assigned to the MAR-3 Sequence (not addressed here), the entire Eldon Fmn was deposited during the upper Bathyuriscus Zone. The presence of thick platformal carbonates above basinal facies of the Burgess Shale in the Kicking Horse Pass region near Field, British Columbia suggests that the Eldon platform locally extended farther into the Panthalassa oceanic basin than did the Cathedral Fmn platform. This observation is supported by the absence of sections having slope deposits of the Tokumm Unit (coeval to the Eldon Fmn) directly overlying the Wapta Mbr or equivalent undivided Burgess Shale Fmn. This relation between the Delamaran-age Cathedral platform and Marjuman-age Eldon platform can also be seen in other nearby areas such as Prospector's Valley, British Columbia. In general, however, the Cathedral and Eldon Escarpments are located near each other (Fig. 6), owing to tectonic control of the underlying Kicking Horse Rim (Lickorish and Simony, 1995). 3.5. Trilobite biostratigraphy Biozones of the Delamaran and Marjuman (‘Middle Cambrian’) are interval zones, based on the first appearance datum (FAD) of the index species. Trilobite lineages used most often in defining these zones include the Redlichiida, Corynexochida, Ptychopariida, and Agnostida (Briggs and Robison, 1984). Beginning with the lowest zone in the Delamaran Stage, the zonal guide fossils are: Albertella Helena Walcott/Paralbertella bosworthi Walcott (Albertella Zone), Glossopleura templensis Rasetti (Glossopleura Zone); Marjuman Stage guide fossils: Bathyuriscus rotundatus (Rominger) and Ehmaniella waptaensis Rasetti (Bathyuriscus Zone), and Bolaspidella aff. wellsvillensis Resser (Bolaspidella Zone, Eldon Fmn: not addressed herein). Agnostid trilobites appear to undergo more rapid evolutionary turnover during the Cambrian than coeval Ptychopariida, Redlichiida, Asaphida, and Corynexochida (Rasetti, 1951; Fritz, 1991). The possibility of finer zone resolution should result, but such ‘subzones’ are best recognized in basinal mudstones of the Chancellor Grp and less so on the platform. The Bathyuriscus-Elrathina Zone, for example, has two such agnostid faunules: Pagetia bootes Subzone (base of Campsite Cliff Mbr to Raymond quarry level within the Walcott Quarry Shale Mbr), and Pagetia walcotti Subzone (level of Raymond quarry, Walcott Quarry Mbr to top of Wapta Mbr). These are preceded by P. praecurrens (Monarch Fmn to top of Yoho River Mbr), and succeeded by Ptychagnostus gibbus (Field Mbr, Eldon Fmn). Thus, subzones or “faunules” (McIlreath, 1977a) are delineated by these two biostratigraphic interval (Fig. 2). Each of these biozones corresponds closely with significant facies change, reflecting eustatically controlled changes in sediment supply and parasequence stacking patterns (landward or basinward). The base of the Albertella Zone is located near the top of the Ross Lake Mbr of the Cathedral Fmn; the overlying Glossopleura Zone begins very near the top of the Trinity Lakes Mbr, Cathedral Fmn (Aitken, 1997). Additionally, the basal Bathyuriscus Zone (characterized by the Ogygopsis klotzi Faunule: the most abundant species within the Mount Stephen Trilobite Beds) occurs at the relatively sharp contact between the Yoho River Limestone Mbr and Campsite Cliff Shale Mbr. The most regionally-persistent facies change, however, is the Cathedral– Stephen contact, located within the lower Glossopleura Zone. Biostratigraphic position of this lithofacies contact is fixed by the presence of G. templensis/G. boccar, and absence of Glossopleura and Polypleuraspis below the Trinity Lakes Mbr (within the upper third of the Cathedral platformal carbonates; Aitken, 1997: Figs. 17,28,52). Furthermore, the Cathedral–Stephen contact represents the boundary between the first and second ‘Grand Cycles’ of the platformal/basinal succession in the Rocky Mountains (Aitken, 1966: Fig. 4). Transition from RST to TST (sequence boundary formation), or from TST to RST (maximum flooding/condensed interval formation) is a primary control on the first appearance and often extinction of trilobite species. Considering the paleoceanographic changes that accompany

79

the rise and fall of sea level on the scale of 10,000s to 1,000,000s of years (parasequences to system tracts), it is not unexpected that the contacts between biozones are at, or very near, such transitions. The underlying premise is that frequent oscillations in sea level exert sufficient ecological stress, and select for populations or species more capable of adapting to changing marine environments (Beauchamp and Henderson,1994). The basal boundaries of at least four Cambrian trilobite zones can be correlated with maximum flooding (MFI), during the transition from transgression to regression. These include: • Albertella Zone — relatively sharp facies transition at or near base of Ross Lake Shale Mbr, Cathedral Fmn (Ross Lake Mbr = RST of DEL-2 Sequence): upward change from platformal limestone to calcareous mudstone; • Glossopleura Zone — relatively sharp facies transition at or near base of Trinity Lakes Shale Mbr, Cathedral Fmn (Trinity Lake Mbr = RST of DEL-3 Sequence): upward change from platformal limestone to calcareous mudstone; • Bathyuriscus Zone – relatively sharp facies transition at or near base of Campsite Cliff Shale Mbr, Burgess Shale Fmn (Campsite Cliff Mbr = RST of DEL–MAR Transitional Sequence): upward change from mound-bearing limestone to exhalite-bearing mudstone; • Bolaspidella Zone – [above study interval, not discussed herein] relatively sharp facies transition at or near base of Pika Fmn (correlative lower Vermilion subunit = RST of MAR-3 Sequence): upward change from platformal limestone to thin-bedded intertidal argillaceous carbonates. The examples above, representing four consecutive Laurentian trilobite biozones, make a case for the stratigraphic juxtaposition of zone boundaries and MFIs. Each flooding interval is characterized by a transition from carbonate deposition to an interval of more argillaceous or mudstone-dominated deposition. Sequence boundaries within the Delamaran and Marjuman stages are not as well delineated, in terms of biotic turnover and appearance of new biostratigraphically-useful species. All seven sequences considered herein lack significant paleontological response across the transitions from RST to TST: top of DEL-2 (Ross Lake Mbr); top of DEL-3 (Trinity Lake Mbr: near base of Glossopleura Zone); top of DEL-4 (‘upper’ Monarch Fmn); top of DEL-5 (Yoho River Mbr); top of DEL–MAR (Wash Mbr); top of MAR-1 (Wapta Mbr); and top of MAR-2 (Field Mbr; not shown in Fig. 4). Each records a transition from muddy clastic deposits to more calcareous or carbonate-dominated strata. We have, as yet, no explanation for this phenomenon; for example, in Mesozoic strata of the Western Canada Sedimentary Basin marine sequence boundaries are nearly always at the break between recognized biozones. This is particularly evident in third-order sequences of the Upper Cretaceous (Collom and Kravec, 2000a), with molluscan faunas changing in concert with sea level fall and basinward progradation of sandstone-dominated clastic wedges. It is possible that the biological response (e.g., trilobites) to eustatic changes at this early stage in metazoan evolution was still becoming established during the Ediacaran and Cambrian periods, considering that most lineages had 20 Myrs or less residency time in continental shelf and slope environments — having originated in deep water spreading ridge settings (Johnston and Collom, 2001). 3.6. Facies and systems tracts A fundamental component of assessing system tracts, particularly in the context of facies trends, is assessing the geometry of lithological units. The regional geometry of the Cathedral margin is relatively well known. Much of it was deposited during transgressive phases of the DEL-2 to DEL-4 cycles and was extensively modified during the Delamaran stage by mass wasting processes. Conversely, the geometry of TST system tracts in the basin adjacent to the Cathedral Escarpment is poorly understood and requires further study. Consideration of the

80

C.J. Collom et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 277 (2009) 63–85

sequence stratigraphic position of carbonate mud mounds has led to the interpretation (see discussion above) that these features built-up during relative sea level rise. They all appear to originate close to the escarpment, but can slide farther away down slope as a result of gravity and/or syndepositional seismicity (Pratt, 1998, 2002). Mud mounds that were capable of moving or sliding as isolated entities were identified within the upper Monarch Fmn (Fig. 7A,B) and also the Wash Limestone at the northwest ridge section on Mt Stephen (Fig. 7D). The Yoho River Limestone mounds at the Mount Stephen Trilobite Beds site (Fig. 7C) differs in that they form part of a laterally extensive bed of partially-dolomitized carbonate averaging about 2.0 m thick. Unlike mounds in the older Monarch Fmn and overlying Wash Mbr, the Yoho River mounds could only slide farther from the Cathedral Escarpment if then entire limestone/dolostone bed were to move as a coherent unit. This is unlikely, as the exposure in question is N100 m2 in area. Broken fragments of partly-disintegrated mounds are present in the Yoho River Mbr at Mt Stephen, but are imbedded to varying degrees in other mounds. At Mt Stephen the Monarch Fmn consists of anastomosed, partly-dolomitized mud mounds (Fig. 3B) that are difficult to differentiate from the underlying Tak Tongue carbonate deposits. The nature of Monarch mud mounds is thus not consistent between the type locality (Monarch Cirque) and Kicking Horse Pass. Intervals containing mounds in the Wash Mbr and parts of the Monarch Fmn mound were non-pervasively cemented. Being enclosed within mudstone (and rare coated grain facies) these mounds resemble those of the Upper Cretaceous Teepee Buttes in Pueblo County, Colorado (Howe, 1987; Collom, unpublished data). Abundant geochemical and paleontological data support an exhalative origin for the N500 isolated mounds within a relatively thin stratigraphic interval of the Pierre Shale Fmn (Arthur et al., 1982; Longman, 1997). The Wash Mbr (Cambrian) and Teepee Butte Mbr (Cretaceous) mounds both are comprised dominantly of micrite, the latter is host to huge numbers of the chemosymbiotic bivalve Nymphalucina. No such macrofauna are known from Monarch or Burgess Shale (or younger) mounds, and bivalves have yet to be documented from the Cambrian of Laurentia. Only rare trilobite fragments have been observed by us within mud mounds in the study area. Wash Limestone Mbr-type mounds bear little overall resemblance to the much larger Lower Devonian Kess Kess mud mounds at Hamar Laghdad, Morocco (Aitken et al., 2002, Fig. 5), which are more akin to the Yoho River Mbr mounds in overall shape and spatial distribution. The 35 or so discrete mounds of the Seheb el Rhassel Formation at this North African locality are also interpreted to be of exhalative or hydrothermal origin (Belka, 1998; Mounji et al., 1998). Much like the Yoho River Limestone mud mounds on Mt Stephen (locality #4, Fig. 1A), they appear to be firmly ‘rooted’ in a laterally-pervasive, but thin limestone bed (b2.5 m). Instead of an adjacent escarpment, the Devonian mounds are nucleated on a quasi-active volcanic rise. Both Paleozoic examples are intimately associated with deep-seated faults. The Kess Kess mounds built-up during relative sea level rise, and represent TST carbonate deposits (Aitken et al., 2002). We would not be surprised if the majority of seep-related mounds throughout Earth history were found to have grown during sea level transgression; their origin being inextricably linked to both processes (sea level fluctuation and hydrothermal activity; Beauchamp and Henderson, 1994). Third-order tectonoeustacy appears to have been ongoing during the entire Phanerozoic Eon, but mud mounds lacking biological frameworks preferentially accumulate during deposition of TSTs if there are seeps operating in an area with the prerequisite structural faults. 3.7. Olistoliths of the Late Delamaran stage Elliptical pods of ooidal packstone (up to 4 m in length) and an olistolith derived from the Cathedral Fmn occur in the monotonous RST mudstones of the DEL-5 Sequence. McIlreath (1977a, pl. 6–2:B,C: p. 117) was probably the first to record and illustrate a huge olistolith

on Mt Stephen, which he described as “a margin-derived megaclast (approximately 90 ft high by 30 ft [wide] at the top) interrupting basinal lime mudstones”, and he further noted “… vertical bedding within the block and the relatively horizontal bedding of the enclosing basinal lime mudstones.” Aitken (1997, Fig. 37) most recently identified it as a “giant talus block from the Cathedral Escarpment embedded in the Takakkaw Tongue” (p. 69). Having studied it in outcrop ourselves, we concur the giant block is rectangular-shaped and estimate its size at 40 m long × 10 m wide × 15 m thick (Figs. 3B,8). This Cathedral olistolith would technically be classified as ‘megagravel’ (32.8–65.5 m size class), and described as a ‘Very Coarse Block’ on the Udden–Wentworth scale. Whereas Aitken (1997) concluded that the large olistolith lies within the Tak Fmn, we reinterpret its position to be within the Kicking Horse Mbr based on the presence of the Monarch Fmn mound interval below and overlying calcareous beds and mounds of the Yoho River Limestone. Clearly the olistolith is not a remnant of the Cathedral Escarpment-forming collapse event. Exactly when this second Glossopleura Zone mass wasting event transpired is not clear, but that it occurred prior to deposition of the Yoho River Mbr is established by the presence of very common to abundant Ogygopsis klotzi at a stratigraphic level above the olistolith. Accordingly, we determine that the shedding occurred during the regressive portion of the DEL-5 Sequence. Interestingly, the block is oriented such that the long axis (∼40 m) is vertical, suggesting that it penetrated into the relatively soft, non-indurated basinal muds like a dart when it was shed, or some type of obstacle prevented it from sliding away into the deep basin (the fate of nearly all debris associated with the Cathedral Megatruncation Event; Stewart et al., 1993). In his ground-breaking dissertation study, Stewart (1991, plates 26d,41) illustrated another ‘periplatform-derived talus block,’ at The Monarch (locality #9 in Fig. 1, and on extreme right of Fig. 3A). He describes it as “projecting from the top of the Cathedral megabreccia … The megaclast is standing on end (steeply inclined stripes are bedding, as confirmed by remnant internal sedimentary structures), and is onlapped and draped by thin-bedded ribbon calcilucite of the upper Takakkaw Tongue” (p. 403). This olistolith was affectionately nicknamed “The Big Muthah” (D. Stewart, pers. comm.), for its impressive size; we estimate to be ∼70 to 75 m in length. With the exception of stratigraphic position (and thus relative timing of their shedding from the Cathedral margin), the Monarch and Mt Stephen olistoliths are rather similar in their size, shape, and vertical orientation in outcrop. The thin-bedded ‘calcilutite’ draped over the ‘Big Muthah’, and other large Cathedral in situ breccia megablocks, on the south side of Monarch Cirque (Fig. 9A) represents the lower Monarch Fmn as defined herein. Thus this olistolith directly abuts the Cathedral Escarpment, and appears to have been ‘shed’ during the late transgressive portion of the DEL-4 sequence. We interpret the highly dolomitized and brecciated platform margin material this large olistolith is embedded in as cavern-filling megabreccia formed by the hydrothermally-driven dissolution of the Cathedral margin (Johnston et al., 2009-this volume). The timing of its formation is enigmatic, and can only be reasonably approximated. As with the North Gully Fault olistolith on Mt Stephen (Fig. 8), there may have been an impediment preventing the megablock from avalanching away into deeper waters of the Columbia Basin (Fritz et al., 1991). In this case, though, the surrounding facies are chaotic and coarsegrained, rather than well-bedded and fine-grained. Abundant sulfides, anomalous precipitates, and euhedral quartz within the so-called “megabreccia” are evidence of extensive synsedimentary brine flow through these strata (Powell et al., 2006). The very presence of breccia of this type directly at the near-vertical escarpment is unexpected, as by comparison with Recent analogs (Hine et al., 1992) such loose, angular material should have slid a considerable distance downslope. Early cementation of void-filling collapse breccias during the final stages of Cathedral Fmn deposition (early Glossopleura Zone; TST of

C.J. Collom et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 277 (2009) 63–85

DEL-4 Sequence, see below) or within the Maximum Flooding Interval (during which the Cathedral Megatruncation Event occurred) seem most plausible. The “Big Muthah” olistolith likely arrived at its present position by either 1. becoming entrapped in a large dissolution cavern located at the shelf-slope break prior to creation of the MTS (Fig. 5), or 2. penetrating through the ceiling of a pre-extant void having formed above the basinward edge of a KHR half-graben — where fluid flow was relatively concentrated. Depending on which portion of the Cathedral the “Big Muthah” was derived from, this olistolith could have fallen as much as ∼ 200 m down the Cathedral Escarpment during the MTS-forming event (or at some time after the megatruncation event). Nonetheless, this Cathedral megablock and the basinal mudstonehosted one from outcrop on Mt Stephen were both derived from the Cathedral platform margin and constitute the only known examples of shed olistoliths. That there were many others can not be doubted, considering the height of the escarpment that resulted from the Cathedral Megatruncation event. These blocks, however, slid far enough from the escarpment to preclude their presence in the several outcrops that we have studied. All other basinal carbonate objects are mud mounds, despite assertions by Stewart et al. (1993) that the blocks are remnants of the collapse. 4. Emmendations 4.1. Nature of the Cathedral margin In a series of sedimentary facies maps, Aitken (1989: Figs. 8–11) illustrates the Takakkaw Fmn as a basinal deposit in relatively sharp juxtaposition against the Cathedral platform (along a NW–SE trending contact) suggesting that the deposition of these beds was penecontemporaneous to the existence of the Cathedral Escarpment, not prior to it, as in the Lexicon definition (1990). The assumption guiding this interpretation was that the near-vertical edge of the Cathedral

81

platform was a depositional artifact of a vertically-accreted Epiphyton reef (McIlreath, 1977a). Evidence presented by Stewart et al. (1993) established that the escarpment was instead the result of mass wasting resulting from multiple destabilizing factors, not ongoing carbonate deposition. We propose that production of lime and dolomitic mud had effectively ceased by the time the Cathedral Megatruncation Event occurred during early Bathyuriscus Zone time. The presence of ooidal packstones in the Monarch Fmn and basal Burgess Shale Fmn indicate that limited carbonate sedimentation was ongoing for approximately 1.5–2.0 Myrs after margin collapse. An underlying tenant of the Aitken (1989) facies maps mentioned above is that the ‘rim’ of the Cathedral platform represents a highlydolomitized reef margin. So intense is the dolomitic alteration of the ‘rim’ it is argued that the original algal bioherms and reefal features are said to have been completely obliterated — with essentially no surviving aspects of these original fabrics. Following McIlreath (1977b) and Aitken and McIlreath (1984), it was assumed that the original, pre-dolomitized platform edge resembled a typical Paleozoic reef-to-slope transition (e.g., Wilson, 1975, p. 25). Reef-rimmed platforms contain multiple succession of different encrusters and frame builders, as well as cavities with internal sediment and cement (James, 1983; James and Macintyre, 1985; Schlager, 1992). By contrast, shoal-rimmed platforms lack organic frameworks and barrier reefs along their seaward edges (where water depths abruptly increase), instead having a complex array of carbonate sand shoals (James and Macintyre, 1985, Figs. 44,45; Fig. 11 herein). Pervasive dolomitization and quartz replacement of the platformto-basin (Cathedral to Takakkaw) transition has indeed overprinted the original depositional fabrics of laterally-discontinuous coated grain facies and peritidal-bedded platformal micritic limestones. Both facies can be observed merging into massive dolostone where the Cathedral Escarpment and underlying rocks are well exposed (Mt Field, Mt Stephen, Natalco Lake, The Monarch). Thrombolitic, stromatolitic, and cryptalgal buildups or framework, however, are

Fig. 11. Generalized models of platform edge development. Lower illustration shows model used by nearly all workers to date, beginning with Aitken (1966). This interpretation invokes a reef rim to the Cathedral Fmn platform, despite the lack of any ‘fringing reef’ or frame-building organisms in Cathedral or Eldon carbonates within the study area. The upper illustration shows the model used in this study; that the Cathedral platform margin was similar in numerous respects to the western margin of the Bahamian platform. Our interpretation takes into account the predominance of bedded micritic limestone (relatively quiet water shelf deposits) and coated grain shoals (ooidal and pisoidal facies) observed in outcrop. Images modified from James (1983).

82

C.J. Collom et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 277 (2009) 63–85

wholly absent from the Cathedral margin at these localities. Domal structures up to 2.0 m in height consistent with thrombolites have been observed within the Waputik Mbr on the platform (Aitken, 1997: Figs. A68,70; Beauty Creek, AB), but these occurrences are N20 km inboard from the restored Cathedral margin and clearly post-date the termination of platform growth. The limited volumes of platformderived ooidal and pisoidal deposits documented herein and previously (Aitken, 1997) hardly constitute evidence for a thriving carbonate platform early in the Glossopleura Zone. Poorly-resolved depositional environments for the Cathedral margin have provided fertile grounds for other unfortunate miscorrelations (Ludvigsen, 1989). Slope deposits of the Tak Fmn are relatively more calcareous than post-MTS basinal mudstones of the Burgess Shale, casting further doubt on any correlations having Cathedral platform construction during or after deposition of the Monarch Fmn. Lastly, the presence of Glossopleura boccar and related trilobites within the Narao Mbr of the Stephen Fmn (e.g., Mt Bosworth, Castle Mountain: Fig. 9D) and the Monarch Fmn/lower two members of the Burgess Shale Fmn establishes that much of the Glossopleura Zone indeed post-dates the Cathedral Fmn (Fig. 2). Credible data is therefore lacking to establish that (1) a ‘reef’ was ever present between peritidal platformal facies and Tak Fmn equivalents on the slope, and (2) the so-called “rim facies” (Stewart, 1991: Fig. 54) or “core facies” (Fletcher and Collins, 1998: Fig. 3) of the Cathedral Fmn are coeval to the lower part of the “basinal” Stephen Shale Fmn. Both models have, at least via illustrations, suggested that buildup of the Cathedral platform continued well after the Cathedral Megatruncation Event during early Glossopleura time. Detailed regional-scale correlations presented herein (Figs. 2,4) suggest instead that the Stephen and Burgess Shale formations represent deposition that clearly post-date deposition of the Cathedral carbonates and their Takakkaw slope equivalents. Further, almost half of the lower member of the Stephen Fmn (Narao Mbr) is correlative with a basinal interval (referred to by some as the ‘upper Takakkaw Tongue’) that we have named the Monarch Formation. As such, the name ‘Takkakaw Fmn’ is restricted to slope deposits coeval to the Cathedral Fmn, most of which are older than the basal Glossopleura Zone. 4.2. Age and correlation of Stephen Fmn It is difficult to establish high-resolution stratigraphic correlations between the Stephen Shale Fmn (‘inner carbonate belt’) and the Burgess Shale Fmn (‘outer detrital belt’) that can withstand close scrutiny. Several published platform-to-basin correlations have been based primarily on trilobite biostratigraphic criteria (Ludvigsen, 1989; Fritz and Simandl, 1993; Aitken, 1997). However, there is no consensus among Laurentian trilobite workers as to the biozone scheme and zonal index species employed for these lithostratigraphic correlations. It is also important to bear in mind that the ∼ 6.0 Ma interval (507.5 to 501.5 Ma BP) during which both the upper Stephen Fmn and Eldon Fmn were deposited is entirely within the same biozone of the corynexochid trilobite Bathyuriscus. Lacking more precise zone control, it is prudent to correlate using all available lines of geological evidence. When interpreting a platform–basinal sedimentary succession that has experienced one or more collapse events, it is critical to determine whether significant depositional unconformities are present. Such bedding-parallel disconformities represent depositional hiatus and/or submarine erosion. Large-scale collapse of carbonate platform margins could represent intervals in excess of 1.0 Myrs, unlike disconformities that usually characterize third-order sequence boundaries. Accordingly, unconformities of this type will have implications for correlation, particularly from on-platform sequences to off-platform equivalents. Close examination of biostratigraphic and lithostratigraphic data for the interval between the Cathedral and

Eldon formations (in a sequence stratigraphic context) indicates that no major unconformities exist within either the Stephen Fmn (platformal) or coeval Burgess Shale Fmn (basinal). A lack of significant diastemas at the disconformable contacts between Grand Cycles was suspected by Aitken (1966, p. 424), who stated that “at several levels, faunal evidence reveals that the contacts marking the commencement of cycles are not strongly diachronous, and also that any hiatus at the contact is of short duration.” The fundamental differences in how the Stephen and Burgess Shale formations have been variously correlated are illustrated in Fig. 10. Deiss (1940), the earliest attempt, considered a massive unconformity to be present at the ‘rim’ of the Cathedral platform based on the apparent lack of the Glossoplera Zone fauna in the heavily-dolomitized Cathedral facies that is so common along the margin. Uplift and tilting of the carbonate platform were invoked as the mechanism that generated the eroded ‘rim.’ Later workers such as Rasetti (1951), Fritz (1971), and McIlreath (1977a) perpetuated the idea that a substantial unconformity was present at this stratigraphic level. Stewart (1991, Fig. 55) and Aitken (1997) suggested a correlative relationship between the lower, more calcareous half of the Stephen Fmn (Narao Mbr) and the dolomitized ‘rim’ of the Cathedral Fmn. Both stratigraphic intervals were deposited during the Polypleuraspis Subzone of the Glossopleura Zone, and Aitken (1997, Fig. 53) claimed their coeval nature could be seen at ‘Chalet Gully’ on Mt Field, b1.0 km from the Cathedral Escarpment. While not disputing the close age association of the uppermost Cathedral Fmn and Narao Mbr (the two being genetically-related portions of the same third-order sequence: DEL-4), our field observations support the Narao Mbr being younger than the topmost Cathedral Fmn everywhere the Stephen Fmn is present on the platform (Fig. 10). With no real carbonate shoal complex at its margin (as in Recent biological barrier reefs), the Cathedral and Eldon platforms were constructed almost entirely from peritidal bedded micritic limestone and ooidal shoals (Fig. 11). The degree to which these facies were biologically-derived by macroorganisms or their precipitation and accumulation mediated by microbiota (such as calcareous algae) remains to be determined. Localized hydrothermal ‘hot-spots’ and their unusual facies (Moore, 1994; Jeary, 2002) comprise notable exceptions, being unlike the typical banded micritic platform facies. These hydrothermally-influenced localities (e.g., Whirlpool Point, Alberta for the Cathedral: Jeary, 2002) also indicate the presence of exhalative processes that influenced many sedimentary aspects of the Kicking Horse Rim during the Delamaran and Marjuman stages (Powell et al., 2006). Aitken (1997, Fig. 40) acknowledged that the base of the Stephen Fmn was generally conformable with the underlying Cathedral Fmn, “except, possibly, strictly locally” (p. 72). However, within the basinalfill mudstones he illustrated (ibid.) the Glossopleura-Bathyuriscus biozone boundary rising diachronously through the ‘Boundary Limestone’ (= Yoho River Limestone, sensu Fletcher and Collins, 1998) implying that this carbonate unit was younger distally from the Cathedral Escarpment. This interpretation may have resulted from a miscorrelation of mound-bearing carbonates of the Yoho River Mbr (highest Glossopleura Zone) and Wash Mbr (Bathyuriscus Zone). The critical exposure of the Wash Limestone in the lowest part of the Walcott Quarry on Fossil Ridge was only uncovered by the Royal Ontario Museum in the late 1990s, and the corresponding exposure on the northwest shoulder of Mt Stephen requires access by helicopter, and as such, was probably unknown to all workers prior to Fritz (1971). Including data from these two localities, field relationships support a non-diachronous model for all beds within the Burgess Shale and Stephen Shale formations, with the Glossopleura-Bathyuriscus boundary separating the Yoho River and Campsite Cliff members in the basin, and the Narao and Waputik members on the platform (Fig. 9D).

C.J. Collom et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 277 (2009) 63–85

83

5. Conclusions

A.1. Monarch formation

Of the Cambrian sedimentary formations in North America, the Burgess Shale is perhaps the most widely known and recognized — principally because of the well-preserved fossil lagerstätten it contains. Less well known are the details of the strata within which these fossil invertebrates are entombed. This study attempts to resolve long-standing issues involving the geological setting and depositional environment of the Burgess Shale Fmn, in addition to previously unknown aspects of this fine-grained fossiliferous unit and the adjacent (and older) Cathedral Fmn. We have (1) defined and named a new formation — the Monarch Fmn — between the slope deposits of the Tak Tongue and the earliest Burgess Shale basinal shales; (2) established that the majority of anomalous carbonate bodies of Delamaran and Marjuman stage ages that have previously been referred to as olistoliths are autochthonous or parautochthonous mud mounds; (3) that these dome-shaped mud mounds are variablydolomitized and located in discrete stratigraphic horizons, having grown in situ at the base of the Cathedral Escarpment in conjunction with exhalative processes; (4) that mud mounds and associated basinal marine facies occur in repetitive, predictable sequences that originated via cyclical eustatic processes; and (5) the Narao Mbr of the Stephen Fmn is not coeval to the upper Cathedral Fmn, but instead represents younger, siliciclastic-dominated sedimentation during deposition of the Burgess Shale (upper Glossopleura Zone). The renowned fossil beds associated with these seep-related sediments and mounds require re-interpretation in light of new data presented here. Published paleoenvironmental reconstructions of the Burgess Shale invoking anoxia as the main factor in soft tissue preservation and distribution of organisms in a narrow strip at the seafloor-escarpment interface are shown to be inadequate. Steadily accumulating evidence support the hypothesis that at least a portion of the unusual biota of the Burgess Shale was chemosynthesis-based (Johnston et al., 2009-this volume), and that their isolated occurrences along the length of the Cathedral Escarpment represent “oases” at which life thrived briefly, owing to the brines expelled at the base of the Escarpment at these sites. It is entirely plausible that the faultcontrolled micro-environments of these localities served as “steppingstones” for early organisms migrating into continental shelves from deeper water settings. Guided by these new observations, our reinterpretation leads to a considerably different perspective of depositional events and the distribution of invertebrate organisms in the Burgess Shale along a ∼150 km length of the Laurentian margin approximately half billion years ago.

Age: Delamaran Stage (Glossopleura Zone), “Middle” Cambrian. Author(s): Collom, Johnston, and Powell (this paper). Type locality: South face of Monarch Cirque, British Columbia. UTM Coordinates: Zone 11: 580,350 E–5,654,560 N (WGS-84 Datum). Lithology: Lower portion consisting of dark, calcareous, thinbedded mudstone; upper portion characterized by bedded limestone, dolomite, coated grain facies, and conspicuous dome-shaped carbonate mud mounds. Thickness and distribution: All surface exposures are in British Columbia. Estimated to be ∼ 55 m thick at the type locality (47 m in Stewart, 1991: p. 453,511), although the exposure is partly on inaccessible vertical cliffs. Somewhat more covered, but readily accessible outcrops of the mounds and fossiliferous debrites are found on the north side of Monarch Cirque. The Monarch Formation is clearly exposed near Field, BC (Kicking Horse Pass area), on the south face of Mt Field (95 m thick) and the north face of Mt Stephen (∼ 85 m thick). The formation is present on vertical cliffs above remote Natalco Lake and at Odaray Pass. Relations to other units: Conformably overlies the Takakkaw Formation (formerly ‘Takakkaw Tongue’; also referred to as “Thin Cathedral”) — lateral equivalent of the Cathedral Formation, or the Monarch Formation abuts against the Cathedral Formation at the Cathedral Escarpment. Overlain conformably by the Burgess Shale Formation (Kicking Horse Mbr); grades distally (to the west) into the Lower Chancellor Group. Paleontology: Invertebrate metazoans preserved in lower, thinbedded unit (principally trilobites, such as Glossopleura and Olenoides); microbial microfossils (e.g., Epiphyton), and the tubeworm(?) Byronia are found within the upper, mound-bearing unit. References: McIlreath, 1977a; Stewart, 1991; Aitken, 1997; Fletcher and Collins, 1998; Powell et al., 2006; Johnston et al., 2009-this volume.

Acknowledgments We thank Robyn Pollock, Rosemary Powers, Kimberley Johnston, Jim McCabe and Kevin Aulenbach for assistance in the field and laboratory; Randle Robertson and the Burgess Shale Geoscience Foundation kindly allowed use of their guide hut in Field, BC on many occasions. The Royal Tyrrell Museum of Palaeontology (Drumheller, Alberta) provided logistical support. Funding provided through NSERC Grant #244503 to P. Johnston. Discussions with Alison ‘Pete’ Palmer on aspects of Cambrian biostratigraphy and Brian Pratt, on mud mound micrite, are appreciated and were helpful. Numerous improvements were made to the manuscript as a result of both reviewers. Appendix A The new lithostratigraphic name Monarch Formation is used in this paper for the first time, and is herein formally defined in accordance with Articles 1–15 of the North American Stratigraphic Code (1983). The format used follows that of the Canadian Lexicon of Stratigraphic Names (1990), to which the addition of the available name ‘Monarch Formation’ in its present use is proposed.

References Aitken, J.D., 1966. Middle Cambrian to Middle Ordovician cyclic sedimentation, southern Rocky Mountains of Alberta. Bulletin of Canadian Petroleum Geology 14 (4), 405–441. Aitken, J.D., 1971. Control of Lower Paleozoic sedimentary facies by the Kicking Horse Rim, Southern Rocky Mountains, Canada. Bulletin of Canadian Petroleum Geology 19 (3), 557–569. Aitken, J.D., 1978. Revised models for depositional grand cycles, Cambrian of the southern Rocky Mountains, Canada. Bulletin of Canadian Petroleum Geology 26, 515–542. Aitken, J.D., 1989. Birth, growth and death of the Middle Cambrian Cathedral carbonate lithosome, Southern Rocky Mountains. Bulletin of Canadian Petroleum Geology 37 (3), 316–333. Aitken, J.D., 1997. Stratigraphy of the Middle Cambrian platformal succession, southern Rocky Mountains. Geological Survey of Canada Bulletin 398 Ottawa, 322 pp. Aitken, J.D., McIlreath, I.A., 1984. The Cathedral Reef Escarpment. A Cambrian great wall with humble origins. Geos 13 (1), 17–19. Aitken, S.A., Collom, C.J., Henderson, C.M., Johnston, P.A., 2002. Sequence biostratigraphy, paleoecology, and origin of Lower Devonian carbonate mud buildups, Hamar Laghdad, eastern anti-Atlas, Morocco, Africa. Canadian Society of Petroleum Geologists Bulletin 50 (2), 217–243. Allan, J.A., 1914. Geology of Field map-area, B.C. and Alberta. Geological Survey of Canada Memoir 55 Ottawa, 312 pp. Arthur, M.A., Kauffman, E.G., Scholle, P.A., Richardson, R., 1982. Geochemical and paleobiological evidence for the submarine spring origin of carbonate mounds in the Pierre Shale (Cretaceous) of Colorado. Abstracts with Programs — Geological Society of America 95th Annual Meeting, New Orleans, LA, Oct. 18–21, 14(7), p. 435. Bathurst, R.G.C., 1975. Carbonate Sediments and their Diagenesis: Developments in Sedimentology 12, second edition. Elsevier, New York. 658 pp. Beauchamp, B., Henderson, C.M., 1994. The Lower Permian Raanes, Great Bear Cape and Trappers Cove formations, Sverdrup Basin, Canadian Arctic: stratigraphy and conodont zonation. Bulletin of Canadian Petroluem Geology 42 (4), 562–597. Belka, Z., 1998. Early Devonian Kess–Kess carbonate mud mounds of the eastern AntiAtlas (Morocco), and their relation to submarine hydrothermal venting. Journal of Sedimentary Research 68 (3), 368–377. Brett, C.E., 1995. Sequence stratigraphy, biostratigraphy, and taphonomy in shallow marine environments. Palaios 10, 597–616. Brett, C.E., 1998. Sequence stratigraphy, paleoecology, and evolution: biotic clues and responses to sea-level fluctuations. Palaios 13, 241–262.

84

C.J. Collom et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 277 (2009) 63–85

Briggs, D.E.G., Robison, R.A., 1984. Exceptionally preserved non trilobite arthropods and Anomalocaris from the Middle Cambrian of Utah. University of Kansas Paleontological Contributions, Paper 111, pp. 1–24. Briggs, D.E.G., Erwin, D.H., Collier, F.J., 1994. The Fossils of the Burgess Shale. Smithsonian Institution Press, Washington, D.C. 238 pp. Brown, W.L., 1948. Monarch and Kicking Horse Mines: Structural Geology of Canadian Ore Deposits. Canadian Institute of Mining and Metallurgy, Montreal, pp. 231–237. Buchardt, B., Israelson, C., Seaman, P., Stockmann, G., 2001. Ikaite tufa towers in Ikka Fjord, Southwest Greenland: their formation by mixing of seawater and alkaline spring water. Journal of Sedimentary Research 71 (1), 176–189. Cloetingh, S., 1988. Intraplate stresses: a tectonic cause for third-order cycles in apparent sea level? In: Wilgus, C.K., Posamentier, H., Ross, C.A., Kendall, C.G.St.C. (Eds.), SeaLevel Changes—An Integrated Approach. Society of Economic Paleontologists and Mineralogists, Special Publication, vol. 42, pp. 19–30. Tulsa. Cook, H.E., Mullins, H.T., 1983. Basin margin environment. In: Scholle, P.A., Bebout, D.G., Moore, C.H. (Eds.), Carbonate depositional environments. American Association of Petroleum Geologists Memoir, vol. 33, pp. 540–617. Tulsa, Oklahoma. Collom, C.J., 2000. On the outcrop [summary of year 2000 field work by P.A. Johnston and C.J. Collom]. Marrella — Yoho Burgess Shale Foundation Newsletter 13, 12–15 ⁎Available Upon Request as PDF file from Lead Author. Collom, C.J., Kravec, H., 2000a. The Upper Cretaceous “Milk River—Belly River Problem”. Southern Alberta: Regional Chronostratigraphy and Tectonic Implications: Geocanada 2000 Millennium Geoscience Summit, May 29–June 2, Calgary, Field Trip Guidebook. 192 pp. Collom, C.J., Johnston, P.A., 2000b. Late Cretaceous seep facies in the Western Interior of Canada and the United States: iron ooids, carbonate mounds, and associated tectonism: Geocanada 2000 Millennium Geoscience Summit, May 29–June 2, Calgary, Abstract #1043, 1 figure. Conway Morris, S., Peel, J.S., Higgins, A.K., Soper, N.J., Davis, N.C., 1987. A Burgess Shalelike fauna from the Lower Cambrian of North Greenland. Nature 345, 802–805. Coppold, M., Powell, W.G., 2000. A Geoscience Guide to the Burgess Shale: Geology And Paleontology in Yoho National Park. The Yoho–Burgess Shale Foundation, Canada. 60 pp. Deiss, C., 1940. Lower and Middle Cambrian stratigraphy of southwestern Alberta and southeastern British Columbia: Geological Society of America Bulletin 51, 731–794. Enos, P., 1983. Shelf Environment (Chapter 6). In: Scholle, P.A., Bebout, D.G., Moore, C.H. (Eds.), Carbonate Depositional Environments. American Association of Petroleum Geologists Memoir, vol. 33, pp. 267–296. Tulsa, Oklahoma. Fletcher, T.P., Collins, D.H., 1998. The Middle Cambrian Burgess Shale and its relationship to the Stephen Formation in the southern Canadian Rocky Mountains. Canadian Journal of Earth Sciences 35, 413–436. Fletcher, T.P., Collins, D.H., 2003. The Burgess Shale and associated Cambrian formations west of the Fossil Gully Fault Zone on Mount Stephen, British Columbia. Canadian Journal of Earth Sciences 40, 1823–1838. Fritz, W.H., 1971. Geological setting of the Burgess Shale. Proceedings of the North American Paleontological Convention (Part I), Chicago (1969), pp. 1155–1170. Fritz, W.H., 1991. Cambrian assemblages. In: Gabrielse, H., Yorath, C. (Eds.), Geology of the Cordilleran Orogeny in Canada. Geology of Canada Series, vol. 4. Geological Survey of Canada, pp. 155–184. Fritz, W.H., Mountjoy, E.W.,1975. Lower and early Middle Cambrian formations near Mount Robson, British Columbia and Alberta. Canadian Journal of Earth Sciences 12 (2), 119–131. Fritz, W.H., Simandl, G.J., 1993. New Middle Cambrian fossils and geological data from the Brussilof magnesite mine area, southeastern British Columbia. Current Reserch, Part A; Geological Survey of Canada, Paper 93-1A, pp. 183–190. Fritz, W.H., Cecile, M.P., Norford, B.S., Morrow, D., Geldsetzer, H.J., 1991. Cambrian to Middle Devonian assemblages. In: Gabrielse, H., Yorath, C.J. (Eds.), Geology of the Cordilleran Orogen in Canada. Geological Survey of Canada, Geology of Canada No. 4, Ottawa, pp. 151–218. Glumac, B., Walker, K.R., 2000. Carbonate deposition and sequence stratigraphy of the terminal Cambrian Grand Cycle in the southern Appalachians, U.S.A. Journal of Sedimentary Research 70 (4), 952–963. Gould, S.J., 1989. Wonderful Life: The Burgess Shale and the Nature of History. W.W. Norton and Company, New York. 347 pp. Grammer, G.M., Ginsburg, R.N., Harris, P.M., 1993. Timing of deposition, diagenesis, and failure of steep carbonate slopes in response to a high-amplitude/high-frequency fluctuation in sea level, Tongue of the Ocean, Bahamas. In: Loucks, R.G., Sarg, J.F. (Eds.), Carbonate Sequence Stratigraphy: Recent Developments and Applications. American Association of Petroleum Geologists Memoir, vol. 57, pp. 107–132. Tulsa, Oklahoma. Greenlee, Lehmann, 1993. Stratigraphic framework of productive carbonate buildups. In: Loucks, R.G., Sarg, J.F. (Eds.), Carbonate Sequence Stratigraphy: Recent Developments and Applications. American Association of Petroleum Geologists Memoir, vol. 57, pp. 43–62. Tulsa, Oklahoma. Harland, W.B., Armstrong, R.L., Cox, A.V., Craig, L.E., Smith, A.G., Smith, D.G., 1989. A Geologic Time Scale. Cambridge University Press, Avon. 263 pp. Harris, P.M., 1979. Facies Anatomy and Diagenesis of a Bahamian Ooid Shoal: Sedimenta VII. Division of Marine Geology and Geophysics, University of Miami. 163 pp. Hine, A.C., Locker, S.D., Tedesco, L.P., Mullins, H.T., Hallock, P., Belknap, D.F., Gonzales, J.L., Neumann, A.C., Snyder, S.W., 1992. Megabreccia shedding from modern, low-relief carbonate platforms, Nicaraguan Rise. Geological Society of America Bulletin 104 (8), 928–943. Howe, B., 1987. Tepee Buttes; a Petrological, Paleontological, Paleoenvironmental Study of Cretaceous Submarine Spring Deposits: Unpublished MSc. Thesis, University of Colorado, Boulder, 217 pp.

Ineson, J.R., Peel, J.S., 1997. Cambrian shelf stratigraphy of North Greenland. Geological Survey of Denmark and Greenland, Bulletin 173 120 pp. James, N.P., 1983. Reef environment. In: Scholle, P.A., Bebout, D.G., Moore, C.H. (Eds.), Carbonate Depositional Environments. American Association of Petroleum Geologists Memoir, vol. 33, pp. 345–440. Tulsa. James, N.P., Macintyre, I.G., 1985. Reefs: zonation, depositional facies, and diagenesis. In: Warme, J.E., Shanley, K.W. (Eds.), Carbonate Depositional Environments, Modern and Ancient: Part 1. Colorado School of Mines Quarterly, vol. 80 (3). 70 pp. Jeary, V.A., 2002. The Origin of Middle Cambrian-hosted Mid-Platform Dolomite; Cathedral Formation, Whirlpool Point, Alberta: Unpublished MSc. Thesis, University of Calgary, 207 pp. Jeary, V.A., Spencer, R.J., 2002. The origin of Cambrian mid-platform hydrothermal dolomites, southern Canadian Rocky Mountains. Geological Society of America Annual Meeting, Denver, Oct. 27–30, Abstracts with Programs, 34(6), p. 334. Johnston, P.A., Collom, 2001. Ancient seeps in the Middle Cambrian Burgess Shale: paleoecological and evolutionary implications. Earth System Processes — Joint Meeting of the Geological Society of America and the Geological Society of London, Edinburgh, Program Abstracts, p. 91. Johnston, P.A., Johnston, K.J., Collom, C.J., Powell, W.G., Pollock, R.J., 2009. Palaeontology and depositional environments of ancient brine seeps in the Middle Cambrian Burgess Shale at The Monarch, British Columbia, Canada. Paleogeography. Paleoclimatology, Paleoecology 277, 86–105 (this volume). Kauffman, E.G., 1979. In: Moore, R.C. (Ed.), The Cretaceous: Treatise of Invertebrate Paleontology, vol. A. University of Kansas Press, Lawrence, pp. 418–487. Keiran, M., 2003. Reading the Rocks — a biography of Ancient Alberta (Chapter 3 — the living sea: the Early Paleozoic era). Royal Tyrrell Museum of Palaeontology. Red Deer Press, Calgary. 121 pp. Kelley, D.S., Karson, J.A., Blackman, D.K., Früh-Green, G.L., Butterfield, D.A., Lilley, M.D., Olson, E.J., Schrenk, M.O., Roe, K.K., Lebon, G.T., Rivizzigno, P., The AT3-60 Shipboard Party, 2001. An off-axis hydrothermal vent field near the Mid-Atlantic Ridge at 30° N. Nature 412, 145–149 (July 12). Kubli, T., Simony, P.S., 1994. The Dogtooth Duplex, a model for the structural development of the northern Purcell Mountains. Canadian Journal of Earth Sciences 31, 1672–1686. Laval, B., Cady, S.L., Pollack, J.C., McKay, C.P., Bird, J.S., Grotzinger, J.P., Ford, D.C., Bohm, H.R., 2000. Modern freshwater microbialite analogues for ancient dendritic reef structures. Nature 407, 626–629 (October 5). Lexicon of Canadian Stratigraphy, 1990. In: Glass, D.J. (Ed.), Western Canada, vol. 4. Canadian Society of Petroleum Geologists. 772 pp. Lickorish, W.H., Simony, P.S., 1995. Evidence for late rifting of the Cordilleran margin outlined by stratigraphic division of the Lower Cambrian Gog Group, Rocky Mountain Main Ranges, British Columbia and Alberta. Canadian Journal of Earth Sciences 32, 860–874. Longman, R.W., 1997. Recent and Cretaceous methane “vent” communities as analogs for Dickinson Field's Mississippian Lodgepole mound (Stark County, North Dakota). AAPG Rocky Mountain Section Meeting, Denver, CO, United States, Aug. 24–27. AAPG Bulletin, vol. 81, no. 7, p. 1228. Ludvigsen, R., 1989. The Burgess Shale: not in the shadow of the Cathedral Escarpment. Geoscience Canada 16, 51–59. Masson, D.G., Canals, M., Alonso, B., Urgeles, R., Huhnerbach, V., 1998. The Canary Debris Flow: source area morphology and failure mechanisms. Sedimentology 45, 411–432. McIlreath, I.A., 1977a. Stratigraphy and Sedimentary Relationships at the Western Edge of the Middle Cambrian Carbonate Facies Belt, Field, British Columbia: Ph.D. Dissertation, University of Calgary, Alberta, 259 pp. McIlreath, I.A., 1977b. Accumulation of a Middle Cambrian, deep water debris apron adjacent to a vertical, submarine carbonate escarpment, southern Rocky Mountains, Canada. In: Cook, H.E., Enos, P. (Eds.), Deep-water Carbonate Environments. SEPM Special Publication, vol. 25, pp. 113–124. Moore, S.L.O., 1994. The Origin of Dolomite of the Middle Cambrian Eldon and Pika Formations in the Yoho Glacier Area, Yoho National Park, British Columbia: Unpublished MSc. Thesis, University of Calgary, 268 pp. Montañez, I.P., Osleger, D.A., 1993. Parasequence stacking pattern, third-order accommodation events, and sequence stratigraphy of Middle to Upper Cambrian platform carbonates, Bonanza King Formation, Southern Great Basin (Chapter 12). In: Loucks, R.G., Sarg, J.F. (Eds.), Carbonate Sequence Stratigraphy: Recent Developments and Applications. American Association of Petroleum Geologists Memoir, vol. 57, pp. 305–326. Tulsa, Oklahoma. Motz, K.J., Johnston, P.A., Collom, C.J., Krause, F.F., 2001. New Burgess Shale-like biotas from the Upper Middle Cambrian Duchesnay unit (Chancellor Group) at Haiduk Peak and Miller Pass, British Columbia. Canadian Paleontological Conference, London, Ontario, Program with Abstracts, vol. 11, pp. 33–37. Mounji, D., Bourque, P.A., Savard, M.M., 1998. Hydrothermal origin of Devonian conical mounds (kess–kess) of Hamar Laghdad Ridge, Anti-Atlas, Morocco. Geology 26, 1123–1126. Ney, C.S., 1954. Monarch and Kicking Horse mines, Field, British Columbia. BanffGolden-Radium Guidebook, Alberta Society of Petroleum Geologists Fourth Annual Field Conference, August 26–28, pp. 119–136. North American Commission on Stratigraphic Nomenclature (M.P. Weiss, Chair), 1983. North American Stratigraphic Code. American Association of Petroleum Geologists Bulletin 67 (5), 841–857. Palmer, A.R., 1998. A proposed nomenclature for stages and series for the Cambrian of Laurentia. Canadian Journal of Earth Sciences 35, 323–328. Palmer, A.R. and Geissman, J. (compilers), 1999. 1999 Time Scale: Geological Society of America [Updated Version of the 1982 DNAG Time Scale]. Paull, C.K., Neumann, A.C., 1987. Continental margin brine seeps: their geological consequences. Geology 15, 545–548.

C.J. Collom et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 277 (2009) 63–85 Paull, C.K., Hecker, B., Commeau, R., Freeman-Lynde, R.P., Neumann, C., Corso, W.P., Golubic, S., Hook, J.E., Sikes, E., Curray, J., 1984. Biological communities at the Florida escarpment resemble hydrothermal vent taxa. Science 226, 965–967. Posamentier, H.W., Vail, P.R., 1988. Eustatic controls on clastic deposition II — sequence and system tract models. In: Wilgus, C.K., et al. (Ed.), Sea Level Changes — An Integrated Approach. Society of Economic Paleontologists and Mineralogists Special Publication, vol. 42, pp. 125–154. Tulsa. Post, R.T., 2001. Sedimentology and Tectonic Significance of Cambrian Stratigraphy, Muncho Lake, Northern British Columbia: Evidence for the Initiation of the Kechika Trough: Unpublished Master’s Thesis, Laurentian University, Sudbury, Ontario, 262 pp. Powell, W.G., Johnston, P.A., Collom, C.J., 2003. Geochemical evidence for oxygenated bottom waters during deposition of fossiliferous strata of the Burgess Shale Formation. Palaeogeography, Palaeoclimatology, Palaeoecology 201, 249–268. Powell, W.G., Johnston, P.A., Collom, C.J., 2004. Magnesite, Talc, and Pb–Zn mineralization associated with Middle Cambrian brine seeps along The Kicking Horse Rim, southeastern British Columbia. Geological Society of America Annual Meeting, Abstracts, November 7–10, Denver. Powell, W.G., Johnston, P.A., Collom, C.J., Johnston, K.J., 2006. Middle Cambrian brine seeps on the Kicking Horse Rim and their relationship to talc and magnesite mineralization, and associated dolomitization, British Columbia, Canada. Economic Geology 101, 431–451. Pratt, B.R., 1995. The origin, biota and evolution of deep-water mud-mounds. In: Monty, C.L.V., Bosence, D.W.J., Pratt, B.R. (Eds.), Carbonate Mud Mounds—Their Origin And Evolution. Special Publication of the International Association of Sedimentologists, vol. 23. Blackwell Publishers, Oxford, pp. 49–123. Pratt, B.R., 1998. Molar-tooth structure in Proterozoic carbonate rocks: origin from synsedimentary earthquakes, and implications for the nature and evolution of basins and marine sediments. Geological Society of America Bulletin 110 (8), 1028–1045. Pratt, B.R., 2002. Tepees in peritidal carbonates; origin via earthquake-induced deformation, with example from the Middle Cambrian of Western Canada. Sedimentary Geology 153 (3–4), 57–64. Price, R.A., 1986. The southeastern Canadian Cordillera: thrust faulting, tectonic wedging, and delamination of the lithosphere. Journal of Structural Geology 8, 239–254. Rasetti, F., 1951. Middle Cambrian stratigraphy and faunas of the Canadian Rocky Mountains. Smithsonian Miscellaneous Collections 116 (5) 270 pp. Sarg, J.F., 1988. Carbonate sequence stratigraphy. In: Wilgus, C.K., et al. (Ed.), Sea Level Changes — An Integrated Approach. Society of Economic Paleontologists and Mineralogists Special Publication, vol. 42, pp. 155–181. Tulsa. Schlager, W., 1992. Sedimentology and sequence stratigraphy of reefs and carbonate platforms. American Association of Petroleum Geologists Continuing Education Course Note Series No. 34. Tulsa, 71 pp.

85

Sloss, L.L., 1963. Sequences in the Cratonic interior of North America. Bulletin of the Geological Society of America 74, 93–113. Sloss, L.L., 1988. Tectonic evolution of the craton. In: Sloss, L.L. (Ed.), Phanerozoic time. Sedimentary Cover—North America Craton, U.S.; The Geology of North America, vol. D-2. Geologic Society of America, Boulder, Colorado, pp. 25–51. Somerville, I.D., 2003. Review of Irish Lower Carboniferous (Mississippian) mud mounds; depositional setting, biota, facies, and evolution. Special Publication — Society for Sedimentary Geology 78, 239–252. Spencer, R.J., Demicco, R.V., 2002. Facies Stratigraphy of two Cambrian grand cycles: implications for Cambrian sea level and the origin of grand cycles. Canadian Society of Petroleum Geologists Bulletin 50, 478–491. Steiner, M., Wallis, E., Erdtmann, B.-D., Zhao, Y., Yang, R., 2001. Submarine– hydrothermal exhalative ore layers in black shales from South China and associated fossils — insights into a Lower Cambrian facies and bio-evolution. Palaeogeography, palaeoclimatology, palaeoecology 169, 165–191. Stewart, W.D., 1991. Stratigraphy and Sedimentology of the Chancellor Succession (Middle and Upper Cambrian), Southeastern Canadian Rocky Mountains: Unpublished Ph.D. Dissertation, University of Ottawa, Ontario, 534 pp. Stewart, W.D., Dixon, O.A., Rust, B.R., 1993. Middle Cambrian carbonate-platform collapse, southeastern Canadian Rocky Mountains. Geology 21, 687–690. Struik, L.C., 1987. The ancient western North American margin: an Alpine rift model for the east-central Canadian Cordillera. Geological Survey of Canada Paper 87–15 19 pp. Tremaine, R.J., Johnston, P.A., Collom, C.J., Pratt, B.R., 2001. Paleoecology of the Middle Cambrian Burgess Shale at The Monarch, British Columbia. Canadian Paleontological Conference, London, Ontario, Program with Abstracts 11, 52–55. Van Balen, R.T., Cloetingh, S.A., 1993. Stress-induced fluid flow in rifted basins. In: Horbury, A.D., Robinson, A.G. (Eds.), Diagenesis and Basin Development. American Association of Petroleum Geologists Studies in Geology, vol. 36, pp. 87–98. Tulsa, Oklahoma. Van der Velden, A.J., Cook, F.A., 1996. Structure and tectonic development of the southern Rocky Mountain trench. Tectonics 15 (3), 517–544. Van Wagoner, J.C., Posamentier, H.W., Mitchum, R.M., Vail, P.R., Sarg, J.F., Loutit, T.S., Hardenbol, J., 1988. An overview of the fundamentals of sequence stratigraphy and key definitions. In: Wilgus, C.K., et al. (Ed.), Sea Level Changes — An Integrated Approach Society of Economic Paleontologists and Mineralogists Special Publication, vol. 42, pp. 39–45. Tulsa. Von Damm, K.L., 2001. Lost City found. Nature (News and Views) 412 (July 12), 127–128. Walcott, C.D., 1911. A geologist's paradise. National Geographic 22 (6), 509–520. Whittington, H.B., 1985. The Burgess Shale. Yale University Press, New Haven. Wilson, J.L., 1975. Carbonate Facies in Geologic History. Springer–Verlag, New York. 471 pp. Young, T.P., 1989. Phanerozoic ironstones: an introduction and review. In: Young, T.P., Taylor, W.E.G. (Eds.), Phanerozoic Ironstones Geological Society Special Publication, vol. 46. London, p. 9–25.