Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2021 Apr 19:9:e11013.
doi: 10.7717/peerj.11013. eCollection 2021.

Geology and taphonomy of a unique tyrannosaurid bonebed from the upper Campanian Kaiparowits Formation of southern Utah: implications for tyrannosaurid gregariousness

Alan L Titus  1 , Katja Knoll  1 , Joseph J W Sertich  2 , Daigo Yamamura  3 , Celina A Suarez  3 , Ian J Glasspool  4 , Jonathan E Ginouves  1 , Abigail K Lukacic  1 , Eric M Roberts  5
Affiliations

Geology and taphonomy of a unique tyrannosaurid bonebed from the upper Campanian Kaiparowits Formation of southern Utah: implications for tyrannosaurid gregariousness

Alan L Titus et al. PeerJ. .

Abstract

Tyrannosaurids are hypothesized to be gregarious, possibly parasocial carnivores engaging in cooperative hunting and extended parental care. A tyrannosaurid (cf. Teratophoneus curriei) bonebed in the late Campanian age Kaiparowits Formation of southern Utah, nicknamed the Rainbows and Unicorns Quarry (RUQ), provides the first opportunity to investigate possible tyrannosaurid gregariousness in a taxon unique to southern Laramidia. Analyses of the site's sedimentology, fauna, flora, stable isotopes, rare earth elements (REE), charcoal content and taphonomy suggest a complex history starting with the deaths and transport of tyrannosaurids into a peri-fluvial, low-energy lacustrine setting. Isotopic and REE analyses of the fossil material yields a relatively homogeneous signature indicating the assemblage was derived from the same source and represents a fauna living in a single ecospace. Subsequent drying of the lake and fluctuating water tables simultaneously overprinted the bones with pedogenic carbonate and structurally weakened them through wet-dry cycling. Abundant charcoal recovered from the primary bone layer indicate a low temperature fire played a role in the site history, possibly triggering an avulsion that exhumed and reburied skeletal material on the margin of a new channel with minimal transport. Possible causes of mortality and concentration of the tyrannosaurids include cyanobacterial toxicosis, fire, and flooding, the latter being the preferred hypothesis. Comparisons of the RUQ site with other North American tyrannosaur bonebeds (Dry Island-Alberta; Daspletosaurus horneri-Montana) suggest all formed through similar processes. Combined with ichnological evidence, these tyrannosaur mass-burial sites could be part of an emerging pattern throughout Laramidia reflecting innate tyrannosaurid behavior such as habitual gregariousness.

Keywords: Behavior; Bonebed; Campanian; Cretaceous; Kaiparowits; Laramidia; Taphonomy; Teratophoneus; Tyrannosauridae; Utah.

PubMed Disclaimer

Conflict of interest statement

The authors declare they have no competing interests.

Figures

Figure 1
Figure 1. Overview map of Laramidia.
Reference map for North America showing the ancient Laramidian land mass and locations of the Dry Island, Two Medicine, and Rainbows and Unicorns Quarry (RUQ) tyrannosaurid mass mortality sites. Also shown is the Wapiti Formation tyrannosaurid tracksite in northern British Columbia and a hypothesized ecotone between northern and southern Laramidia.
Figure 2
Figure 2. Photograph of the Rainbows and Unicorns Quarry.
Photograph of the Rainbows and Unicorns Quarry tyrannosaurid mass-mortality site (14UTKA-8) on the day of discovery. Tyrannosaurid bones were first found eroding out the small rill indicated by the white line. Dr. Michael Knell holding a one-meter grid for scale.
Figure 3
Figure 3. Reference geologic map of the Kaiparowits Plateau.
Reference geologic map of the Kaiparowits Plateau region showing Cretaceous bedrock units and location of the Rainbows and Unicorns Quarry (RUQ). TKcg = Canaan Peak and Grand Castle Formations; Kk = Kaiparowits Formation; Kw = Wahweap Formation; Kdts = Dakota (now referred to the Naturita Formation), Tropic Shale and Straight Cliffs formations. N.M. = National Monument. Monument boundary is as of 2016 and does not reflect the current boundary which was altered in 2017.
Figure 4
Figure 4. Graphic measured sections and photographs of major rock unit types for the RUQ.
Left is of the lower portion of the Kaiparowits Formation from the RUQ area showing the stratigraphic position of the quarry and extrapolated absolute age. (A), (B) and (C) are detailed sections from the quarry whose locations are shown in Fig. 7 and whose unit numbers correspond with samples in the photographs. (D) Hand sample of unit 3 showing gleyed color and white pedogenic carbonate masses. (E) Hand sample of unit 4 showing every major lithic component of the unit except fossils. (F) Unit 5 exposed in fossil jacket during preparation. Marks in upper right are from an airscribe tool. A large piece of turtle carapace is visible in the lower right. (G) Medium grained sandstone of unit 6. Abbreviations are as follows: W = Wahweap; CS = capping sandstone; x-bed SS = cross-bedded sandstone; Calc. = calcareous; Ped. = pedogenic. Grain size abbreviations are; m = mudstone; s = siltstone; f = fine sandstone; m = medium sandstone; c = coarse sandstone; p = pebble; c = cobble.
Figure 5
Figure 5. Photos of RUQ geologic features.
(A) Concentration of gar scales and bones (referable to Lepisosteus sp.) at the base of unit 4. (B) Cross-sectional view showing architectural relationships of RUQ units 3, 4 and 5 and erosional relief imposed on unit 3. Taken at east edge of main tyrannosaurid bone area, just north of where section (B) in Fig. 4 was measured. Scale bar divided into cm. (C) Charcoal specimens in unit 5 from near section Fig. 4B. (D) 6 cm diameter carbonized root and associated mold in unit 3 below fossil jacket capping units 4 and 5. From near section (B). Brush is 50 mm wide for scale.
Figure 6
Figure 6. Details of juvenile tyrannosaur vertebra (#384) from unit 5.
Both pedogenic carbonate (A) and green mudstone (B) infillings are visible inside pneumatic pore spaces. Mudstone infill is only observed inside small pore spaces of intact robust elements like vertebrae.
Figure 7
Figure 7. Quarry map and rose diagram for area of main tyrannosaurid accumulation at the RUQ.
Also shown are sub-unit 4 surface features, source of charcoal samples, and location of detailed sections (Figs. 4A–4C). Dashed lines represent outlines of field jackets.
Figure 8
Figure 8. Photograph showing RUQ tyrannosaur elements of Voorhies groups I–III.
(A–K) Group I. (A) Juvenile dorsal rib. (B) Juvenile cervical rib. (C–E) Juvenile gastralia. (F and G) Juvenile caudal vertebrae. (H) Juvenile dorsal vertebra. (I–K) Juvenile pedal phalanges. (L) Group II, juvenile tibia. (M and N) Group III. (M) Juvenile dentary. (N) Juvenile lacrimal. Scale bar is in centimeters.
Figure 9
Figure 9. Faunal elements from units 4 and 5 at the RUQ site.
(A) Viviparus sp. and an unidentified juvenile unionid (unit 4). (B) Articulated vertebrae of an unidentified teleost fish (unit 5). (C) Overview of area south of main tyrannosaurid accumulation with dense occurrence of turtle carapaces and skeletal material (unit 4). Two intact large Neurankylus utahensis Lively shells, one exposed, one jacketed are visible. (D) Mandible of unidentified giant panchelonioid turtle (unit 5). (E) Osteoderms of juvenile Deinosuchus hatcheri Holland, 1909 from northern portion of RUQ (unit 5). (F) Pedal phalanges of cf. Teratophoneus and a non-paravian coelurosaur grade theropod (likely a juvenile Teratophoneus) exhibiting three, possibly four different growth stages (units 4 and 5). Left to right: possible small juvenile III-3 (#728), juvenile right II-2 (#30), subadult right II-2 (#313), somatic adult III-1 (#21). Bar scales: A = 2 cm; E = 5 cm.
Figure 10
Figure 10. Adult cf. Teratophoneus partial skull complex (element #360) from contact between units 4 and 5 preserving the jugal, postorbital, and maxilla.
Right lacrimal (element #375) is displaced slightly to the upper right. Bar scale = 40 cm.
Figure 11
Figure 11. Plots of stable isotope data.
(A) C–O isotope cross-plot of pedogenic carbonate nodules. Unit 3 in-situ nodules contain both primary micrite yielding tightly clustered results and sparry calcite veins with more widely scattered results that bias the bulk (unit 3 bulk) results towards a heavier average. Unit 4 conglomerate nodules are indistinguishable from those in unit 3 in δ18OCO3 space but are heavier in δ13CCO3 space. (B) Box plot of turtle δ18Op. Solid line represents median δ18Op, box boundary represents third and fourth quantile, whiskers represent maximum and minimum and points represent likely outliers. Abbreviation: G. = Gilmoremys; N. = Neurankylus.
Figure 12
Figure 12. Pre-erosional soil profile showing hypothetical δ18O and δ13C isotopic patterns for unit 3 and overlying strata (now removed).
δ18O values stay relatively constant because they are generated by similar water regimes while δ13C values increase towards the surface as atmospheric input into soil increases. Isotopic values are from Table 1.
Figure 13
Figure 13. NASC-normalized REE patterns and ratios for RUQ fossils and carbonate.
(A) REE patterns that show a HREE-depleted pattern for all bones, dentine and carbonate nodules. (B) Ternary diagram of representative normalized light (NdN), middle (GdN) and heavy (YbN) REEs. The closely similar REE patterns and tight clustering demonstrate similar early fossilization histories for all sampled specimens.
Figure 14
Figure 14. Standard light and SEM images of charcoal specimens from units 4 and 5.
(A) 17C-1B. (B, D, F and L) 17C-1A. (C) 17C-3B. (E, I and K) 15C-1A. (G) 17C-3 Strew 7. (H and N) 15C-1B. (J and M) X-A. (O) 774A-A. (P) 17C-3 Strew 1. Image mode: (A) incident light; (B, I–M and O) scanning electron microscopy; (C–H, N and P) reflected light. Abbreviations: bf = brittle fracture, dpw = degraded parenchyma wall, dr = degraded ray cells, hcw = homogenized cell wall, if = infilled cell, mc = microchecking, ml = middle lamella, nw = nodular parenchyma wall, p = pitting, pe = pitting elongation, rc = resin canal, tec = thick celled epithelial cells, wl = warty layer. Description: A: a cubic (though compressed in the z-dimension) woody fragment of charcoal and showing characteristic black coloration and silky luster. B: cell wall homogenization and a possible warty layer. C: conifer resin canal bordered by thick walled epithelial cells. D: ray parenchyma exhibiting nodular transverse walls. E: cell walls superficially appeared homogenized when viewed by SEM (see: fig. I) though preserving localized evidence of a middle lamella. Here, the lignin-rich middle lamella is clearly visible as a less reflective (darker grey) layer (arrows) surrounded by higher reflecting cellulosic layers. F: ray parenchyma exhibiting nodular end walls. G: unidentified chaotic tissue with differential reflectance, cell infillings and variably thickened and distorted cell walls. H: degraded parenchymal tissues (bottom left) and a tracheid in which microchecks are discernible in cross-section penetrating into the tracheid wall. I: tracheids in which microchecks are visible both through the cellulose wall and as elongation of the pitting. The specimen also preserves a bordered pit within degraded ray parenchyma. The tracheid end walls exhibit brittle fracture and appear homogenized except at the arrow where evidence of the middle lamella is discernible. J: cell wall homogenization and brittle fracture. K: tracheids in which microchecks are visible both through the cellulose wall and as elongations of the pitting. L: tracheids with homogenized cell walls and a bordered pit within a parenchyma ray, the transverse walls of the ray cells are nodular. M: degraded ray cells in longitudinal section. N: degraded, distorted parenchymal tissues in which the cell walls show evidence of separation (dr) and a tracheid in which microchecks are discernible penetrating into the tracheid wall. O: brittle fracture and homogenized cell walls in comparably well preserved tracheids and highly degraded ray cells. P: woody tissues in which the tracheids exhibit brittle fracture and are well-preserved, with clean, entire cell margins (hcw) conversely ray cells have collapsed and show evidence of cell wall degradation (dpw).
Figure 15
Figure 15. Hypothetical growth curve for cf. Teratophoneus with calculated sizes of the four confirmed and fifth possible specimen plotted to show relative developmental stages.
Curve is slightly modified from that of Daspletosaurus given in Erickson et al. (2004). Age-based growth stages based on Carr’s (2020) estimates for Tyrannosaurus rex.
Figure 16
Figure 16. Frequency diagram of all identified elements of cf. Teratophoneus from the RUQ site (A), compared to one for a complete skeleton of Tyrannosaurus rex (B).
The correlation coefficient between the two curves is 0.91, indicating strong positive correlation between the two data sets and suggesting the RUQ tyrannosaurid sample is derived from still largely complete specimens.
Figure 17
Figure 17. Frequency diagram showing total number and weighted averages (dashed vertical line) of taphonomic grade assignments for both the entire graded assemblage (A) and by major taxonomic groups (B–F). Data from Table S6.
Figure 18
Figure 18. Photos of fractured tyrannosaur elements.
Juvenile tyrannosaurid tibia (A) and adult tooth crown (B) from unit 5 showing typical rectangular (orthogonal) fracture patterns observed on nearly all skeletal material from the RUQ.
Figure 19
Figure 19. Four key pre-diagenetic stages interpreted for RUQ bonebed development.
(A) Deposition of vertebrate and other fossils in the context of a low energy lake setting. (B) Lowering of water table and colonization of the former lake area by plants, accompanied by formation of pedogenic carbonate nodules at and below the bone layer. (C) Lower temperature fire event creating hydraulic instability in the region. (D) Avulsion of nearby river channel leading to exhumation and reburial of the former lake and floodplain fossils into a channel setting.
Figure 20
Figure 20. Photograph of accumulated drowned cattle carcasses following massive floodplain flooding in 2019, Queensland, Australia.
Note concentration of carcasses around margins of the lower elevation channel, analogous with the oxbow setting hypothesis we propose for the RUQ. Photo credit: The Fence Post Magazine online version.
See this image and copyright information in PMC

References

    1. Azeem S, Bengis R, Van Aarde R, Bastos ADS. Mass die-off of African elephants in Botswana: pathogen, poison or a perfect storm? African Journal of Wildlife Research. 2020;50:149–156. doi: 10.3957/056.050.0149. - DOI
    1. Barrick RE, Fischer AG, Showers WJ. Oxygen isotopes from turtle bone: applications for terrestrial paleoclimates? PALAIOS. 1999. pp. 186–191.
    1. Beasley CR, Túry E, Vale WG, Tagliaro CH. The reproductive cycle and conservation and management of P. syrmatophorus (Meuschen, 1781) in the Tocantins River. Brazil. Journal of Molluscan Studies. 2000;66:393–402.
    1. Behrensmeyer AK. Taphonomic and ecologic information from bone weathering. Paleobiology. 1978;4:150–162.
    1. Behrensmeyer AK. Terrestrial vertebrate accumulations. In: Allison PA, Briggs DEG, editors. Taphonomy: Releasing the Data in the Fossil Record. New York: Plenum; 1991. pp. 291–335.

LinkOut - more resources