Megaherbivorous dinosaur turnover in the Dinosaur Park Formation (upper Campanian) of Alberta, Canada
Highlights
► The Dinosaur Park Formation is composed of two broad megaherbivore assemblage zones. ► These zones span ~ 600 Ka. ► Each zone is divisible into two ~ 300 Ka sub-zones. ► Megaherbivorous dinosaur turnover does not track known palaeoenvironmental proxies.
Introduction
Distinct but essentially contemporaneous northern and southern dinosaur faunal provinces in the Campanian and Maastrichtian (Late Cretaceous) of western North America are believed to reflect adaptation to latitudinal climatic gradients and associated differences in vegetation (Lehman, 1987, Lehman, 1997, Lehman, 2001, Gates et al., 2010, Sampson et al., 2010b). Purported differences in faunal composition between inland and more coastal palaeoenvironments in the northern province imply that proximity to coastline influenced the distribution of dinosaurs and other taxa (Brinkman, 1990, Brinkman et al., 1998, Lehman, 2001). This suggests that, unlike mammals of comparable size, megaherbivorous dinosaurs (> 1 t) had relatively small geographic ranges that were governed by strict habitat preferences, particularly during the late Campanian when the Western Interior Seaway was at its transgressive maximum (Lehman, 2001), and that populations of these animals were sensitive to palaeoenvironmental and palaeoclimatic change.
The palaeontologically well-documented Belly River Group (ca. 79.1–74.8 Ma ago) of Alberta is of particular importance for understanding these patterns of Late Cretaceous dinosaur palaeoecology and biogeography, and testing the sensitivity of megaherbivores to environmental change. The uppermost unit of this group, the upper Campanian Dinosaur Park Formation (DPF), is particularly rich in megaherbivorous dinosaur fossils that represent Ankylosauria, Ceratopsidae, and Hadrosauridae. Recent high-resolution biostratigraphic and detailed taxonomic work has confirmed a non-random distribution of taxa within the DPF. Species are restricted in range to different parts of the formation (Sternberg, 1950, Godfrey and Holmes, 1995, Holmes et al., 2001, Currie and Russell, 2005, Eberth and Getty, 2005, Evans and Reisz, 2007, Evans et al., 2009), prompting the recognition of discrete assemblage (faunal) zones within the DPF where it is best exposed in the area of what is now Dinosaur Provincial Park (DPP) near Brooks (Béland and Russell, 1978, Ryan and Evans, 2005). The DPF was deposited during a major transgressive phase of the Western Interior Seaway (Eberth and Hamblin, 1993), and geological, vertebrate microfossil, and invertebrate fossil data indicate that the local environment became increasingly more coastally influenced up-section at DPP until its final inundation by the sea (Eberth and Hamblin, 1993, Brinkman et al., 1998, Eberth, 2005, Johnston and Hendy, 2005). Species turnover within the formation is often linked to palaeoenvironmental change associated with this marine transgression (Ryan and Evans, 2005, Evans, 2007, Evans and Reisz, 2007), and the possibility of turnover pulses has been suggested (Sampson, 2009, Sampson and Loewen, 2010), but these hypotheses have not been rigorously tested. In this paper, previous zonation schemes of the DPF megaherbivorous dinosaurs, and their potential association with palaeoenvironmental factors, are quantitatively tested using multivariate clustering and ordination methods for the first time. The suggestion that the pattern of megaherbivorous dinosaur distribution is related to turnover pulses is also tested.
The DPF, previously subsumed under the Oldman and Judith River formations, is the uppermost unit of the Belly River Group, which is a transgressive sedimentary succession found throughout much of southern Alberta and southeastern Saskatchewan (Eberth, 2005). The DPF is best exposed in the area of DPP, where it is ~ 70 m thick and disconformably overlies the Oldman Formation. The sediments of the DPF were deposited during the late Campanian, ca. 76.5–74.8 Ma ago, and reflect the third order transgressive event of the Western Interior Seaway (Eberth, 2005). The sandy lower exposures of the DPF comprise alluvial palaeochannel deposits that have been interpreted as having been at least partly influenced by rhythmic tidal backwater conditions, due to the presence of inclined heterolithic strata in the palaeochannel deposits (Koster et al., 1987, Wood et al., 1988, Wood, 1989). However, Thomas et al. (1987) provided evidence that these strata can develop in alluvial (non-coastal) settings as well, and the palaeoenvironment of the lower DPF is currently interpreted as an alluvial plain dominated by meandering river systems (Eberth, 2005). The muddier upper exposures of the DPF comprise overbank facies that culminate in the Lethbridge Coal Zone (LCZ). These exposures are interpreted as poorly drained, organic-rich (swampy) palaeoenvironments that formed in coastal settings. Marine flooding surfaces are common in the LCZ, recognized on the basis of marine microfossils and trace fossils, and lateral continuity with mud-filled incised valleys (Eberth, 1996). The LCZ is estuarine in origin, comprising a coaly to carbonaceous-dominated mixed freshwater, brackish, and marine succession, and was likely deposited along a wave-dominated shoreline (Eberth, 1996). The LCZ interfingers with the overlying marine Bearpaw Formation, which is composed of structureless and laminated shales.
The DPF is best known for its abundant dinosaur fossils, which often occur as articulated skeletons or in bonebeds, although isolated bones and vertebrate microfossil sites are also common (Dodson, 1971, Brinkman, 1990, Brinkman et al., 2005, Eberth and Currie, 2005, Eberth and Getty, 2005). The locations of many of these fossils within the bounds of present day DPP were initially mapped by Sternberg, 1936, Sternberg, 1950, who also used an aneroid barometer to provide hyposmetric elevation data (metres above sea level = masl) for many of the quarries collected until then. With these data, Sternberg (1950) demonstrated that the stratigraphic distributions of the megaherbivorous dinosaur genera of the DPF are not homogeneous. For example, he observed that, among ceratopsids, Centrosaurus occurs low in section, followed successively by Chasmosaurus and Styracosaurus. The hadrosaurids were said to exhibit a similar pattern, whereby the hadrosaurines Gryposaurus and Prosaurolophus appear low and high in section, respectively, and, among the lambeosaurines, Corythosaurus occurs low in the formation, and Lambeosaurus occurs nearer the top. Sternberg (1950) further resolved the stratigraphic distribution of the lambeosaurine genera at the species level. Within Corythosaurus, Co. casuarius appears low in section and Co. intermedius appears higher up. Within Lambeosaurus, L. clavinitialis and L. lambei occur relatively low in section, and L. magnicristatus occurs at the top of the DPF.
Béland and Russell (1978) also commented on the biostratigraphy of the DPF. They arbitrarily divided the DPF into three vertical “levels” and observed the distribution of dinosaur genera within each, using the data of Sternberg (1950) and subsequently collected specimens. They recognized the existence of two assemblage zones within the DPF: a lower one characterized by the presence of the ankylosaur Euoplocephalus, the hadrosaurid Kritosaurus (= Gryposaurus) and the large theropod Albertosaurus (= Gorgosaurus), and an upper one characterized by the presence of the ankylosaur Panoplosaurus, and the hadrosaurids Lambeosaurus and Prosaurolophus.
Unfortunately, these early attempts at describing the biostratigraphy of the DPF were hampered by their reliance on raw altitudes. The beds of the entire Belly River Group, including the DPF, exhibit a gradual dip < 1° to the west (Eberth and Hamblin, 1993, Eberth, 2005), thereby rendering the elevation data of Sternberg (1950) and Béland and Russell (1978) of limited use. More recent work on the biostratigraphy of the DPF has relied on the use of the more sophisticated survey grade Ground Positioning System (MacDonald et al., 2005) and the Oldman/DPF contact as an isochronous datum to more accurately reflect the stratigraphic positions of the fossils within the area of DPP. The contact is not isochronous over a wider geographic area because the DPF thins in an eastward direction so that it is just 31 m thick where it is exposed near Manyberries, Alberta (Eberth and Hamblin, 1993). The most recent studies to comment on the biostratigraphy of the DPF have used the Oldman/DPF contact isochron to refine biostratigraphic patterns within DPP with great success. Different patterns have been noted for the distribution of various fossil palynomorphs (Braman and Koppelhus, 2005), mollusks (Johnston and Hendy, 2005), fish, frogs, turtles (Brinkman, 1990), mammals (Sankey et al., 2005), and dinosaurs (Brinkman, 1990, Brinkman et al., 1998, Currie and Russell, 2005, Eberth and Getty, 2005, Ryan and Evans, 2005). Currie and Russell (2005) noted that, among centrosaurine ceratopsids, Ce. apertus occurs low in section, followed sequentially by S. albertensis and an as-yet-unnamed ‘pachyrhinosaur’ (Ryan et al., 2010a). The chasmosaurine ceratopsids are represented by Ch. russelli low in section, followed sequentially by Ch. belli and “Ch.” (= Vagaceratops) irvinensis. Among hadrosaurine hadrosaurids, G. notabilis occurs low in section and Pr. maximus occurs higher up. Among lambeosaurine hadrosaurids, Corythosaurus occurs low in section, and is replaced by Lambeosaurus in the upper half of the DPF, with the only specimen of L. magnicristatus occurring near the base of the LCZ (Ryan and Evans, 2005, Evans and Reisz, 2007). Evans (2007) further demonstrated that the hypothesized sexual dimorphs of Co. casuarius and L. lambei (sensu Dodson, 1975) are separated stratigraphically, and therefore likely represent either distinct species or perhaps phyletic chronospecies (Evans et al., 2006). Parasaurolophus sp. is restricted to the lower half of the DPF (Evans et al., 2009).
These distributional patterns (Fig. 1) led Ryan and Evans (2005) to propose three informal assemblage zones within the DPF: a lower zone (0–30 m) comprising Centrosaurus, Ch. russelli and Corythosaurus, a middle zone (30–50 m) comprising Styracosaurus, Ch. belli and Lambeosaurus, and a possible third upper zone (~ 50–55 m) comprising rare occurrences of poorly-known taxa including “Ch.” (= Vagaceratops) irvinensis, a ‘pachyrhinosaur’, and L. magnicristatus.
ACM, Beneski Museum of Natural History, Amherst College, Amherst, Massachusetts; AMNH, American Museum of Natural History, New York; CM, Carnegie Museum of Natural History, Pittsburgh; CMN, Canadian Museum of Nature, Ottawa; FMNH, Field Museum of Natural History, Chicago; MCSNM, Museo Civico di Storia Naturale di Milano, Milan; MLP, Museo de la Plata, La Plata, Argentina; NHMUK, Natural History Museum, London; ROM, Royal Ontario Museum, Toronto; SDNHM, San Diego Natural History Museum, San Diego; TMM, Texas Memorial Museum, Austin; TMP, Royal Tyrrell Museum of Palaeontology, Drumheller, Alberta; USNM, Unites States National Museum, Washington; UALVP, University of Alberta Laboratory of Vertebrate Palaeontology, Edmonton; YPM, Yale Peabody Museum, New Haven.
Section snippets
Data collection
Megaherbivorous dinosaur distributions have proved useful in the establishment of assemblage zones within the DPF for the following reasons: (1) ankylosaurs, ceratopsids, and hadrosaurids are collectively among the most abundant vertebrate fossils in the DPF (Dodson, 1983, Brinkman, 1990); (2) they are found throughout the DPF (Sternberg, 1950, Béland and Russell, 1978, Currie and Russell, 2005, Ryan and Evans, 2005); (3) individual species exhibit limited stratigraphic ranges within the DPF (
UPGMA
The first series of Q-mode cluster analyses was performed on the raw data matrix using the UPGMA algorithm. Clustering was generally poor, regardless of the similarity fossil record and insufficient sampling. In order to improve clustering, a range-through assumption was applied to the data matrix. This involves coding all intervals between the first and last appearance data for a taxon as ‘1’, effectively assuming that taxon to be continuously present throughout its entire range in the
Megaherbivore assemblage zones of the DPF
The establishment of assemblage zones in biostratigraphy is useful because it not only facilitates the characterization of the structure underlying the data, but it also allows the study of evolutionary palaeoecology to be framed so that the confounding effects of time-averaging are limited in a biologically meaningful way (as opposed to binning stratigraphic intervals using some arbitrary criterion). The generally symmetric biostratigraphic clustering schemes of the DPF presented above are
Conclusions
The DPF is among the most fossiliferous dinosaur-bearing deposits in the world and records significant palaeoenvironmental changes associated with marine transgression and tectonism (Eberth and Hamblin, 1993). As such, it provides an excellent dataset to test previous hypotheses of megaherbivorous dinosaur distribution, niche segregation, and turnover pulses. This statistical study supports the existence of two broad assemblage zones in the DPF, each lasting ~ 600 Ka: a lower one characterized by
Acknowledgments
Thanks to P. Currie, E. Koppelhus, M. MacDonald, W. Spencer, D. Tanke, and all others who collected and made available the data used here. D. Braman, D. Brinkman, D. Eberth, and D. Tanke offered valuable help and insight. Ø. Hammer provided technical assistance. This manuscript was greatly improved by the editorial efforts of F. Surlyk and two anonymous reviewers. This work was funded by scholarships from the Natural Sciences and Engineering Research Council, Alberta Innovates Technology Futures
References (122)
Patterns of elephant damage to vegetation in northern Botswana
Biological Conservation
(1993)Multivariate analysis of stratigraphic data in geology: a review
Chemometrics and Intelligent Laboratory Systems
(1987)Paleoecology of the Judith River Formation (Campanian) of Dinosaur Provincial Park, Alberta, Canada: evidence from vertebrate microfossil localities
Palaeogeography, Palaeoclimatology, Palaeoecology
(1990)Sedimentology and taphonomy of the Oldman Formation (Campanian), Dinosaur Provincial Park, Alberta (Canada)
Palaeogeography, Palaeoclimatology, Palaeoecology
(1971)- et al.
Biogeography of terrestrial and freshwater vertebrates from the late Cretaceous (Campanian) Western Interior of North America
Palaeogeography, Palaeoclimatology, Palaeoecology
(2010) Coniss: a Fortran 77 program for stratigraphically constrained cluster analysis by the method of incremental sum of squares
Computational Geosciences
(1987)Comparisons of multivariate analytical techniques for use in pre- quaternary plant paleoecology
Review of Palaeobotany and Palynology
(1989)Late Maastrichtian paleoenvironments and dinosaur biogeography in the western interior of North America
Palaeogeography, Palaeoclimatology, Palaeoecology
(1987)- et al.
40Ar/39Ar age of the Kaiparowits Formation, southern Utah, and correlation of contemporaneous Campanian strata and vertebrate faunas along the margin of the Western Interior Basin
Cretaceous Research
(2005) Multivariate data analysis in palaeoecology and palaeobiogeography–a review
Palaeogeography, Palaeoclimatology, Palaeoecology
(1993)
The Belly River suite: final chronofauna of Mesozoic deep time
Late Pleistocene faunal turnover in the Turkana Basin, Kenya and Ethiopia
Science
Paleoecology of Dinosaur Provincial Park (Cretaceous), Alberta, interpreted from the distribution of articulated vertebrate remains
Canadian Journal of Earth Sciences
Determination of the number of zones in a biostratigraphical sequence
The New Phytologist
Numerical Methods in Quaternary Pollen Analysis
The range-through method and first–last appearance data in paleontological surveys
Journal of Vertebrate Paleontology
Punctuated equilibrium versus community evolution
Campanian palynomorphs
A review of nonmarine turtles from the Late Cretaceous of Alberta
Canadian Journal of Earth Sciences
Turtles: diversity, paleoecology, and distribution
The paleogeographic and stratigraphic distribution of ceratopsids (Ornithischia) in the Upper Judith River Group of Western Canada
Palaios
Vertebrate microfossil sites and their contribution to studies of paleoecology
Multivariate analysis of assemblage zones
Palaeoenvironmental controls on the distribution of Cretaceous herbivorous dinosaurs
Naturwissenschaften
Hadrosaurs as ungulate parallels: lost lifestyles and deficient data
Acta Palaeontologica Polonica
A numerical index for biostratigraphic zonation in the mid-Tertiary of the eastern Gulf
Transactions - Gulf Coast Association of Geological Societies
Non-parametric multivariate analyses of changes in community structure
Australian Journal of Ecology
Statistical design and analysis for a ‘biological effects’ study
Marine Ecology Progress Series
The families of the ornithischian dinosaur order Ankylosauria
Palaeontology
Theropods, including birds
Mass death of a herd of ceratopsian dinosaurs
The geographic and stratigraphic distribution of articulated and associated dinosaur remains
Phenetic clustering in biology: a critique
The Quarterly Review of Biology
Measures of the amount of ecological association between species
Ecology
Taxonomic implications of relative growth in lambeosaurine hadrosaurs
Systematic Zoology
A faunal review of the Judith River (Oldman) Formation, Dinosaur Provincial Park, Alberta
The Mosasaur
Origin and significance of mud-filled incised valleys (Upper Cretaceous) in southern Alberta, Canada
Sedimentology
The geology
Vertebrate taphonomy and taphonomic modes
Ceratopsian bonebeds: occurrence, origins, and significance
Tectonic, stratigraphic, and sedimentologic significance of a regional discontinuity in the upper Judith River Group (Belly River wedge) of southern Alberta, Saskatchewan, and northern Montana
Canadian Journal of Earth Sciences
A centrosaurine mega-bonebed from the Upper Cretaceous of southern Alberta: implications for behaviour and death events
Anatomy and relationships of Lambeosaurus magnicristatus, a crested hadrosaurid dinosaur (Ornithischia) from the Dinosaur Park Formation, Alberta
Journal of Vertebrate Paleontology
High resolution lambeosaurine dinosaur biostratigraphy, Dinosaur Park Formation, Alberta: sexual dimorphism reconsidered
Journal of Vertebrate Paleontology
An unusual hadrosaurid braincase from the Dinosaur Park Formation and the biostratigraphy of Parasaurolophus (Ornithischia: Lambeosaurinae) from southern Alberta
Canadian Journal of Earth Sciences
A new centrosaurine from the Late Cretaceous of Alberta, Canada, and the evolution of parietal ornamentation in horned dinosaurs
Acta Palaeontologica Polonica
A consideration of the trophic dynamics of a Late Cretaceous large-dinosaur community (Oldman Formation)
Ecology
Cited by (75)
Filling the Bearpaw gap: Evidence for palaeoenvironment-driven taxon distribution in a diverse, non-marine ecosystem from the late Campanian of west-Central Alberta, Canada
2022, Palaeogeography, Palaeoclimatology, PalaeoecologyRefined geochronology and revised stratigraphic nomenclature of the Upper Cretaceous Wahweap Formation, Utah, U.S.A. and the age of early Campanian vertebrates from southern Laramidia
2022, Palaeogeography, Palaeoclimatology, PalaeoecologyCitation Excerpt :Many similar stratigraphic intervals elsewhere in the Western Interior preserve multiple, biostratigraphically-defined local faunas, including the Fruitland and Kirtland formations of northwestern New Mexico (e.g., Sullivan and Lucas, 2003; Lucas et al., 2006), the Two Medicine and Judith River formations of Montana (e.g., Horner et al., 2001; Trexler, 2001; Mallon et al., 2016), and the Aguja Formation of West Texas (e.g., Lehman et al., 2017, 2019). Investigations of well-preserved non-marine faunas from the Campanian infer turnover rates as rapid as 600 kyr (e.g., Dinosaur Park Formation; Mallon et al., 2012), although these estimates may be influenced by preservation potential and paleoenvironmental preference (Cullen et al., 2016). Based on these examples and a temporal range of ~3.2 million years for the fossiliferous portion of the Wahweap Formation, interpretation of a single, sympatric Wahweap fauna is almost certainly over-simplistic.
Sierraceratops turneri, a new chasmosaurine ceratopsid from the Hall Lake Formation (Upper Cretaceous) of south-central New Mexico
2022, Cretaceous ResearchCitation Excerpt :Additionally, the neoceratopsian Zuniceratops christopheri is known from the middle Turonian Moreno Hill Formation in west-central New Mexico (e.g., Wolfe and Kirkland, 1998; Wolfe et al., 2010). This study adds to a growing body of work on the evolution of chasmosaurine ceratopsids in North America (e.g., Loewen et al., 2010; Sampson et al., 2010; Sullivan and Lucas, 2010; Longrich, 2011; Mallon et al., 2012, 2014, 2016; Wick and Lehman, 2013; Ryan et al., 2014; Brown and Henderson, 2015; Lehman et al., 2017; Fowler and Freedman Fowler, 2020) showing high horned dinosaur diversity, and complex biogeography. The diversity of chasmosaurines in southern Laramidia is poorly understood compared to better-sampled assemblages in the north.
Evolution of the Western Interior Seaway in west-central Alberta (late Campanian, Canada): Implications for hydrocarbon exploration
2021, Marine and Petroleum Geology