The lower Cambrian Cranbrook Lagerstätte of British Columbia

Konservat-Lagerstätten represent sites with exceptional preservation, including articulated skeletons with or without soft tissues and sometimes entire soft-bodied organisms (Seilacher 1970). The more famous Konservat-Lagerstätten offer unique insights into the anatomy of fossilized animals and the composition of ancient communities, and therefore bolster understanding of important chapters in the history of life (Seilacher et al. 1985). Sampling effort and a combination of ecological, taphonomic and palaeoenvironmental constraints determine the abundance and diversity of soft-bodied organisms. However, relatively few Cambrian Burgess-Shale-type Konservat-Lagerstätten yield very rich assemblages or capture a wide range of soft-tissue types and thus soft-bodied anatomy (e.g. Collins et al. 1983; Hou et al. 2004; Caron et al. 2014; Gaines 2014; Fu et al. 2019); inevitably, examples such as the Chengjiang or the Burgess Shale tend to be the most celebrated. Cambrian sites that yield less diversity and abundance of soft-bodied fossils, and typically preserve rare, heavily sclerotized arthropod elements, have received less attention despite sharing the same mode of preservation as carbonaceous compressions (Gaines et al. 2008). These lower tier Burgess-Shale-type deposits are more numerous, cover greater geographical and stratigraphic ranges (e.g. Conway Morris 1989), and are not limited to the Cambrian period (e.g. Saleh et al. 2022). Like the Cranbrook fossil Lagerstätte discussed here, they have the potential to illuminate important aspects of palaeoecosystems and palaeoenvironments.

The Cranbrook Lagerstätte is one of the oldest to yield Tuzoia and Anomalocaris, and the oldest to yield A. canadensis (Briggs 1979). Tuzoia valves and isolated raptorial appendages of Anomalocaris were initially reported by Resser (1929). Additional material of both taxa was subsequently described (Copeland 1993), but since then no other soft-bodied taxa from this site have been discussed in the literature. In this paper, after reviewing previous research, we discuss initial results of a 2015 Royal Ontario Museum (ROM)-led expedition, the first systematic field-based geological and palaeontological investigation of the Cranbrook Lagerstätte. This, and the recent museum accession of important fossil specimens hitherto residing in private collections, allow us to present a fuller summary of the biota of the Lagerstätte (including new soft-bodied taxa, potentially new species of lower Cambrian Tuzoia diagnosed on the basis of a morphometric analysis, and the trilobites), taphonomic biases, the depositional setting and the ichnofauna.

Cambrian fossils from NE of Cranbrook, British Columbia, were first reported over a century ago by the Geological Survey of Canada (Schofield 1921, 1922) (Fig. 1a–c). The abundance of large and complete trilobites, as well as accessible outcrops, made this area attractive to many fossil enthusiasts, in particular Lieutenant-Colonel C. H. Pollen, who discovered some of the first fossils (Fig. 2a). Schofield (1922) mapped three fossil sites (localities A, B and C; Fig. 1c, Supplementary material Fig. 1) and established the Eager Formation along the old Cranbrook–Fort Steele Road, assigning it to the upper part of the lower Cambrian (now Series 2, Stage 4) based on identifications of the trilobites by C. Walcott. Following Schofield's reports, this area became well known to fossil collectors and the local community (Fig. 2b and c). Field trips were even advertised in local newspapers, in particular visits to the easternmost fossil site along the Cranbrook–Fort Steele Road, a locality that requires restudy (A in Fig. 1b and c). Despite the long history of collecting, there has been little scientific research on these fossil deposits. Trilobites have been investigated as part of broader doctoral studies (Best 1959; Bohach 1997) but, notwithstanding their abundance and sometimes exquisite preservation, limited results have been published (Best 1952; Hu 1985).

During the early Cambrian, the Cranbrook area was located in the Eager Trough, a structurally active, fault-controlled basin on the outer shelf along the northern margin of Laurentia (Larson and Price 2006). The thickness of the Eager Formation varies regionally, but in the Cranbrook area it is estimated to be up to 2000 m (Rice 1937). Only a portion of this unit, which is mainly represented by shale and siltstone with subordinate carbonate beds (Schofield 1922), is exposed around Cranbrook. Unfortunately, the exact position of the Cranbrook Lagerstätte within the Eager Formation is poorly constrained, as the overlying units have been completely eroded and the contact with the underlying Cranbrook Formation is not exposed (Fig. 1j). The Eager Formation lies within what was formerly called the Bonnia–Olenellus zone (Briggs and Mount 1982; Conway Morris 1989) of the Dyeran Stage of Laurentia (Palmer 1998a), but the Bonnia–Olenellus zone has since been abandoned (Webster 2011). A taxonomic revision of the trilobite assemblage indicates a mid-Dyeran Stage 4 age (c. 514–509 Ma, Fig. 1j; Webster and Caron, in press) in an as-yet undefined species-level trilobite zone.

The 2.65 m interval at locality B, studied during the 2015 excavation (Supplementary Material and methods), appears uniform and generally featureless in outcrop, but polished blocks and thin sections reveal structures ranging from laminae to thin beds from 2 to 12 mm in thickness (Fig. 1i). The interval is dominated by pure claystone but ∼35% of the section by thickness is composed of graded mudrock beds with medium silt to very fine sand deposited in thin quartz laminae at bed bases. These coarser-grained beds commonly include winnowed ‘hash layers’ of disarticulated skeletal debris, including Tuzoia carapaces (Fig. 3a), hyoliths (Fig. 4h), brachiopods (Fig. 4i), and trilobites (Fig. 5d). No structures owing to oscillatory flows (e.g. wave ripple cross-lamination) were observed, indicating deposition below the influence of storm waves. In this outer shelf setting, which probably lay within the photic zone (as evidenced by the presence of algae, cyanobacteria and microbial mat textures; see below), deposition occurred episodically from turbid flows initiated by storm wave disturbance on the shallower shelf. Sometimes these flows had sufficient energy to transport quartz grains and scour the seabed, reworking sediments and skeletal material and generating the ‘hash layers’ that contain many disarticulated trilobite elements (mostly cephala) and more rarely fragments of Tuzoia carapaces (Fig. 5d). Coarser-grained sediments were sometimes trapped within dwelling burrows (Diplocraterion isp. assemblage: see below). More frequently, however, turbid flows reaching the site carried only a clay-sized fraction, which was deposited across the seafloor without sediment reworking. This observation is reflected in a high proportion of articulated trilobite material, complete Tuzoia valves, and a low-density ichnofauna (Helminthoidichnites tenuis assemblage: see below), indicating colonization between sediment influx events. Pulses of oxygenated conditions allowed brief intermittent colonization by a benthos dominated by trilobites, with rare epifaunal organisms such as brachiopods, hyoliths, sponges and pterobranchs. The abundance of articulated trilobites suggests an autochthonous or parautochthonous fossil assemblage, with limited transport (see Trilobites section below). However, the rarity of epifaunal organisms may reflect a high frequency of burial events that prevented larval settlement. As in the Burgess Shale (Caron and Jackson 2006), isolated sclerotized elements of nektonic arthropods (i.e. the raptorial appendages of Anomalocaris and valves of Tuzoia) are probably moults and might have settled from the water column to the seafloor or been transported into the local environment. This is evidenced by disarticulation, and also fragmentation in the case of Tuzoia, of specimens of both taxa.

Decay was a strong taphonomic bias, as evidenced by the rare preservation of more decay-prone elements such as pterobranch stolons (Fig. 4g). Soft-bodied fossils, which are preserved as carbonaceous compressions (Figs 2 and 3), are rare and limited almost exclusively to heavily sclerotized tissues. This echoes a broad pattern of Tier 3 Burgess Shale-type deposits, characterized by rarity of soft-bodied forms and very low soft-bodied diversity (Gaines 2014), where conditions for soft-bodied preservation were far more limiting compared with those of richer Tier 1 deposits such as the Burgess Shale.

The depositional setting of the Eager Formation is similar to those of the Latham Shale (Gaines and Droser 2002) and Ruin Wash Lagerstätten (Webster et al. 2008) of the Great Basin (Laurentia), as well as the Mural Formation Lagerstätte of Alberta (Sperling et al. 2018), all of which are also of Cambrian Series 2, Stage 4 age. A high proportion of articulated trilobites, together with the preservation of sclerotized organic cuticle (in the case of the Eager Formation and Latham Shale), was facilitated to a first order by rapid entombment in fine-grained sediments, preferentially in beds without winnowing or scour. These beds yield very low-diversity assemblages (except for the trilobites) and rare soft-bodied organisms, and are enclosed by (Ruin Wash) or intercalated with (Eager Formation, Latham Shale) beds resulting from higher energy deposition, suggesting that they lay close to a taphonomic threshold controlled by physical reworking and oxygenated conditions. Although anoxic conditions alone are not sufficient to account for exceptional preservation, they are widely recognized as a contributory factor (e.g. Allison 1988; Butterfield 1995). In this respect, the setting of the Cranbrook site shares similarities with the Mural Formation Lagerstätte of the Canadian Rockies. In that unit, the low diversity and rarity of soft-bodied fossils, and the selective preservation of sclerotized tissues, were linked directly to the prevalence of oxygenated conditions during deposition, which limited the potential for soft-bodied preservation (Sperling et al. 2018).

The fossil assemblages are dominated by trilobites through almost the entire 2.65 m section sampled, suggesting a recurrence of the same type of community and/or environmental or taphonomic conditions (Fig. 1k). Olenellus and Wanneria are the most abundant trilobites in all intervals, reflecting earlier tallies based on museum collections (Best 1952; Bohach 1997) and confirming the dominance of these genera. Hyoliths, brachiopods, sponges and soft-bodied fossils represent a minority of specimens. The most abundant soft-bodied organism, Tuzoia (see next section), represents only 0.6% of specimens, occurring in just 10 of 59 intervals sampled (Fig. 1k). Only one partial and disarticulated Anomalocaris anterior limb was recovered from an interval between −100 and −60 cm (Supplementary material Fig. 2a). Overall, systematic fossil collecting yielded a low-diversity fossil assemblage (Fig. 1k) and confirmed the rarity of soft-bodied organisms compared with shelly fossils.

Tuzoia is a large Cambrian arthropod with a global distribution (Wen et al. 2019), known mostly from isolated carapace valves, with a potential affinity to the hymenocarine bivalved arthropods (Izquierdo-López and Caron 2022). The morphology of the Tuzoia valve is highly variable (Box 1) and this, together with its wide geographical occurrence, resulted in taxonomic oversplitting in the past. Lieberman (2003) proposed several characters to distinguish species of Tuzoia: peripheral spines (number and position); reticulation (size of cells); lateral ridge (relative position and relief); valve shape (ventral and dorsal margins, notches below cardinal processes); angles between hinge line and cardinal processes.

Resser (1929) described three new species of Tuzoia from the Eager Formation, T. polleni, T. nodosa and T. spinosa (Fig. 2d–i). Lieberman (2003) recognized only T. polleni as represented in the Cranbrook Lagerstätte. He noted that the holotype of T. spinosa (Fig. 2g–i) is poorly preserved but shares the spinous margin and other features with T. polleni. He considered T. nodosa (Fig. 2f) to be ‘essentially morphologically identical’ to T. polleni. Lieberman (2003) attributed variation in the number and position of spines among the Cranbrook Lagerstätte specimens to preservational differences. In their review of Tuzoia, Vannier et al. (2007) followed Lieberman (2003) in recognizing only T. polleni among the Cranbrook Lagerstätte specimens. Both of these assessments were based on the illustrations in the paper by Resser (1929).

New material collected by the ROM expedition of 2015, together with specimens donated to museum repositories and new observations and photography of the type material, permitted a morphometric analysis (Box 1, Supplementary material). The results, based on 22 specimens in total, support the decision of Lieberman (2003) to synonymize T. nodosa with T. polleni. Both species share four posterior spines, and no other spines are present along the ventral margin. The holotype of T. nodosa (Fig. 2e and f) is notably smaller than the other T. polleni specimens studied, and minor variations of spine numbers on the dorsal margin may be ontogenetic (e.g. Fig. 3a–e). The fragmentary nature of the two specimens of T. spinosa illustrated by Resser (Resser 1929), which appear to belong to two different valves from separate individuals on the same shale surface, is more problematic (Fig. 2g–i). One of these specimens (Fig. 2i) has four shorter posterior spines similar to those in T. polleni, whereas there are additional and longer spines in the other, referred to here as morphotype A (Fig. 2h), which is similar to other spinous specimens recovered from the site (Fig. 3g–m). A single specimen with six short postero-ventral spines might represent a separate species (morphotype B, Fig. 3n–p). Tuzoia polleni is the most abundant morphotype, accounting for more than three-quarters of all specimens across all repositories surveyed (Supplementary material Table 1).

Anomalocaris canadensis was the first described representative of the clade now known as Radiodonta (Collins 1996), based on specimens from the Trilobite Beds on Mount Stephen (Whiteaves 1892). Resser (1929) erected a new species, A. cranbrookensis, based on a raptorial limb from Cranbrook (Fig. 2d). The Anomalocaris limb was interpreted as the trunk of an arthropod for nearly 90 years, and Resser (1929, p. 6) considered Tuzoia as the likely carapace of the same animal based on the association of these genera at several localities. Briggs (1979, p. 633) recognized that the specimens of Anomalocaris were limbs rather than bodies, and synonymized A. cranbrookensis with A. canadensis, a view tentatively accepted by Copeland (1993) and endorsed by Potin and Daley (2023), by which time 37 species of radiodont had been described. The anterior limbs of Anomalocaris are rare in the Cranbrook Lagerstätte. All previously known specimens are preserved articulated (e.g. Fig. 2c; see also Copeland (1993) and Supplementary material Fig. 2b), but the single specimen recovered during the 2015 expedition (Supplementary material Fig. 2a) is disarticulated, suggesting a greater level of transport or decay.

Sponges are reported for the first time from the Cranbrook Lagerstätte based on two tubular specimens (Fig. 4a–c) with a relatively simple skeleton showing large and long oxeas on the outside layer and an inner layer of horizontal short monaxial spicules (Fig. 4b). These specimens are provisionally identified as leptomitids, a group with a wide stratigraphic and geographical distribution but greatest generic diversity in the Cambrian (Wang et al. 2019). The affinities of leptomitids are disputed: they have been considered as demosponges (Rigby 1987; Rigby and Collins 2004), hexactinellids (Reitner and Mehl 1995), or a problematic group (Botting and Muir 2018) potentially stem to Porifera or Silicea (Botting 2021).

Local clusters of small cylindrical structures (Fig. 4c and d), reminiscent of Fuxianospira (LoDuca et al. 2017), and putative Morania-like cyanobacteria (Fig. 4e) are also present in the assemblage. Several specimens of an undescribed benthic colonial pterobranch (Fig. 4f and g) are reminiscent of Yuknessia from the Burgess Shale and Utah (LoDuca et al. 2015). Each tubarium consists of elongate tubes that branch near their base where they emerge from a holdfast. A thin median darker area along the length of some tubes also preserved in Cambrian rhabdopleurids from Siberia (Sennikov 2016) may represent remnants of a stolon (Fig. 4g), a feature that is apparently common in some dendroid graptolites (Maletz 2020), but is absent in Yuknessia (LoDuca et al. 2015).

Trilobites (Figs 5 and 6a, h) are by far the most abundant fossils in the Cranbrook Lagerstätte (Fig. 1k), but have received little research attention. The study by Best (1952) remains the primary published source on the Eager Formation trilobites; he examined 1400 specimens and identified five species: the olenelloids Olenellus gilberti Meek in White 1874, Olenellus eagerensis Best 1952 (subsequently assigned to the genus Mesonacis), Olenellus schofieldi Best 1952, Wanneria walcottana (Wanner 1901), and the dorypygid Bonnia columbensis Resser 1936. Hu (1985) briefly described the ontogenetic development of Olenellus gilberti based on a handful of specimens originally collected from Cranbrook by Best. Subsequent unpublished work by Best (1959) and Bohach (1997) suggested the presence of seven trilobite species (with a tentative eighth) and questioned the identification of some of the earlier reported species. Palmer (1998b) and Webster (2015) also noted that the material historically identified as Olenellus gilberti from Cranbrook differs from the type material of that species and represents a distinct, undescribed species.

The number and preservational quality of the Cranbrook trilobites (Fig. 5) yield valuable insights into the palaeobiology of these ancient arthropods (see Box 2). Full documentation of the trilobites, including descriptions of the species and their respective ontogenies, is the subject of another publication (Webster and Caron in press).

Hyoliths are reported for the first time from the Cranbrook Lagerstätte. They are rare in the assemblage (Fig. 1k), but include specimens showing the characteristic conch, operculum and curved lateral spines or ‘helens’ of one of the two sub-groups of hyoliths, the hyolithids (Fig. 4h). Hyoliths are generally regarded as early lophophorates (Moysiuk et al. 2017) or a basal lophotrochozoan clade (Liu et al. 2019), although they have traditionally been considered closer to molluscs based on shell microstructure (Li et al. 2019).

Small, c. 1.5 mm acrotretid brachiopods (Fig. 4i) are rare in the sampled interval (Fig. 1k). The shells show ornamentations (Fig. 1l), and features such as a pointed ventral valve with a divided pseudointerarea, and a dorsal valve with a median division suggest affinity with Prototreta or Homotreta (sensu Bell 1941). In contrast, brachiopod specimens from near locality C (Fig. 4m–q) are relatively large, c. 8–11 mm. Most of these specimens belong to an elongate ovoid linguloid form with pustulose ornamentation characteristic of the family Eoobolidae (Fig. 4m and n), whereas transversely ovoid linguloid specimens may represent the family Botsfordiidae (Fig. 4p). Two other specimens (Fig. 4q) possess radial wrinkles and a sulcus, and are reminiscent of calcareous-shelled obolellids such as Brevipelta (Geyer 1994). In addition, a pair of elongate linguliform specimens appear distinct from the above morphotypes but are poorly preserved (Fig. 4o).

The Cranbrook Lagerstätte ichnofauna represents a new record of trace fossil assemblages in a Burgess Shale-type deposit, offering valuable clues to benthic life and environmental conditions during deposition of the Eager Formation. The trace fossils record the activities of a surficial epifauna and a shallow-tier infauna dominated by worm producers, although arthropod trackways (Diplichnites, Fig. 6d) have been recovered from at least two stratigraphic levels (−49 and −90 cm). Both levels also display microbially induced sedimentary structures (MISS), which are commonly associated with trackways and surficial trails in Cambrian assemblages elsewhere (e.g. Buatois et al. 2012, 2014; MacNaughton et al. 2019). The relative scarcity of true trackways and undertracks in this succession is probably due to the absence of well-developed, non-erosive, bedding plane surfaces and lithological interfaces that typically enhance their visibility. Alpha ichnodiversity is low and bioturbation is typically sparse, as evidenced in cross-section (Fig. 6c). The Bioturbation Index (BI of Taylor and Goldring 1993, is 0–1; i.e. less than 5% biogenic disruption), and reaches 2 (6–30% biogenic disruption) only very locally. The Bedding Plane Bioturbation Index (BPBI of Miller and Smail 1997) is difficult to assess owing to limited exposure of well-defined bedding surfaces. However, a BPBI value of 2 (less than 10% of bioturbated area) has been assessed in a few larger slabs. Regardless of the low ichnodiversity and density overall, ichnotaxa are persistent through the quarry section (Fig. 1k). Helmithoidichnites tenuis (Fig. 6e), Palaeophycus tubularis (Fig. 6c and g), Diplocraterion isp. (Fig. 6a–c), ?Arenicolites isp. (Fig. 6a) and a diversity of large finger-like structures containing small (2–3 mm) disarticulated or broken trilobite sclerites (mainly cephala) and, less commonly, fine sand, collectively represent over 90% of recorded specimens. Other ichnotaxa such as Cochlichnus isp., Helminthopsis isp., Teichichnus rectus (Fig. 6f) and Treptichnus isp. are rare. Bergaueria isp. shows only very local occurrence of relatively high density.

Overall, the Cranbrook trace fossil assemblages do not show well-developed tiering. Two broad assemblages recur through the studied interval: (1) the Helminthoidichnites tenuis assemblage is dominated by trails and shallow burrows, and is composed of Helminthoidichnites tenuis, Helminthopsis isp., Palaeophycus tubularis and Diplichnites isp.; (2) the Diplocraterion isp. assemblage is typically monospecific to very low diversity, commonly associated with finger-like structures and/or local presence of P. tubularis and ?Arenicolites isp. Large finger-like structures are somewhat reminiscent of ribbon-like, elongate aggregates interpreted as coprolites or gut contents of endobenthic predators, such as priapulid worms, in other Burgess Shale-type deposits (e.g. Vannier and Chen 2005). However, in the Cranbrook Lagerstätte, these structures are most probably passively infilled burrows. Dwelling burrows containing trilobite fragments and/or coarser grains coarser than the host rock suggest significant sediment bypass and trapping of transported grains within the burrows (see also the section Geological setting, depositional environment and taphonomy). Other ichnotaxa are rare and cannot be assigned to a particular assemblage.

Several Tuzoia specimens are superimposed by Helminthoidichnites tenuis of various sizes (e.g. Fig. 6e) but the trace fossils are not confined to the carapaces and these associations may be incidental. No pellet-infilled (e.g. Alcyonidiopsis) or branching structures, such as those associated with non-biomineralized carapaces in other Burgess Shale-type deposits (Mángano et al. 2012, 2019), have been found. Tuzoia specimens are occasionally modified by overprinting burrows (e.g. Fig. 6a–e).

The Cranbrook Lagerstätte is typical of other Tier 3 Burgess Shale-type deposits (Gaines 2014), which are characterized by a low diversity of soft-bodied taxa and preserve only the most robust types of soft tissues. Other Laurentian Series 2, Stage 4, Lagerstätten yielding radiodonts and Tuzoia typically fall into this category with the possible exception of the Parker Quarry Lagerstätte of Vermont (Pari et al. 2021, 2022) (Table 1). In the Cranbrook Lagerstätte, rare and limited soft-tissue preservation probably reflects physical reworking of the seabed during sedimentation, but also suggests that exceptional preservation was constrained by benthic oxygenation (Gaines et al. 2012) through at least portions of the Eager Formation. The new collections considered here reveal a number of previously unrecorded forms, including an exceptionally preserved colonial pterobranch, a leptomitid sponge, brachiopods and a hyolithid.

Future research on the Cranbrook Lagerstätte (Box 3) will benefit from further investigations of existing collections and, especially, from additional fieldwork, although the potential for new species discoveries appears somewhat limited by the stratigraphic extent of the productive layers and the fidelity of soft-tissue preservation as mentioned above. Both investigations, however, will add to our understanding of early Cambrian marine communities that lived below storm wave-base in a photic zone setting in a distal outer shelf environment, and the depositional setting in which they are preserved.

We thank N. Butterfield and F. Saleh for constructive remarks, and the Editor of this series, P. Donoghue, for his encouragement and feedback. A permit to collect fossils by the ROM-led research team in 2015 was obtained under a Land Act licence of occupation (#405166) from the then Ministry of Forests, Lands and Natural Resource Operations of the Province of British Columbia. We thank E. Deom from the Fossil Management Office for facilitating the permit application. The fossil material collected during the 2015 expedition is reposited at the Royal British Columbia Museum and special thanks are due to V. Arbour and D. Larson for their assistance with research loans. We thank fieldwork volunteers M. Akrami, C. Bell, D. George, J. Moysiuk and D. Nordby for their assistance. F. Katay, Regional Geologist, and B. Robison from TerraLogic Exploration provided logistical support. We thank P. Fenton (ROM) for logistical support and assistance, and M. Akrami (ROM) for assistance with collections. We thank C. Jenkins and D. Askew for donation of brachiopod specimens used in this research now reposited at the Cranbrook History Centre, D. Humphrey for locating historical images and H. Neve for assistance with collections. Donation by G.P. to the ROM of additional material from the Eager Formation collected by R. Drachuck prior to the establishment of the British Columbia Mineral Definition Modification Regulation in 2005 was approved by the Fossil Management Office. J.B.C. thanks C. Luo for discussions on Cambrian sponges.

J-BC: conceptualization (lead), data curation (equal), funding acquisition (lead), investigation (equal), methodology (lead), project administration (lead), resources (lead), supervision (lead), validation (equal), visualization (lead), writing – original draft (lead), writing – review & editing (equal); MW: conceptualization (equal), data curation (supporting), formal analysis (equal), funding acquisition (supporting), investigation (equal), methodology (equal), project administration (supporting), resources (supporting), validation (equal), visualization (supporting), writing – review & editing (supporting); DEGB: conceptualization (supporting), formal analysis (supporting), investigation (equal), methodology (supporting), validation (equal), writing – original draft (supporting), writing – review & editing (equal); GP: investigation (supporting), resources (equal), writing – review & editing (supporting); GS: resources (supporting), writing – review & editing (supporting); MGM: conceptualization (supporting), data curation (supporting), formal analysis (equal), funding acquisition (supporting), investigation (equal), methodology (supporting), validation (supporting), writing – original draft (supporting), writing – review & editing (equal); AI-L: formal analysis (equal), investigation (supporting), methodology (supporting), software (lead), validation (supporting), writing – review & editing (supporting); MS: formal analysis (supporting), investigation (supporting), writing – original draft (supporting), writing – review & editing (supporting); RRG: conceptualization (equal), formal analysis (supporting), funding acquisition (supporting), investigation (equal), methodology (equal), resources (supporting), validation (supporting), visualization (supporting), writing – original draft (supporting), writing – review & editing (equal)

Fieldwork expenses were covered by ROM internal sources (Peer Review Research Grant, Fieldwork Grant, Casual Salary Grant, Restricted Curatorial Funds) and Pomona College. This research was also supported by a Natural Sciences and Engineering Research Council Discovery Grant to J.B.C. (#341944) and M.G.M. (#311727), and by a National Science Foundation EAR Integrated Earth Systems grant (#1410503) to M.W. M.G.M. acknowledges additional funding by the George J. McLeod Enhancement Chair in Geology.

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

All data generated or analysed during this study are included in this published article and its supplementary information files.