In lieu of an abstract, here is a brief excerpt of the content:

495 32 A Centrosaurine Mega-Bonebed from the Upper Cretaceous of Southern Alberta: Implications for Behavior and Death Events DAVID A. EBERTH, DONALD B. BRINKMAN, AND VAIA BARKAS the hilda area of southern Alberta preserves the remains of at least 14 discrete monodominant Centrosaurus apertus bonebeds that occur in a single, organic-rich mudstone bed deposited in a coastal-plain interfluve. The identical and traceable stratigraphic positions of the 14 bonebeds allow us to group and interpret them as a single mega-bonebed, covering an estimated area of 2.3 km2 and containing a cumulative minimum number of individuals (MNI) estimated to be in the very low thousands. The Hilda mega-bonebed is a rare example of a multi-kilometer-scale, macrovertebrate bonebed that is inferred to have formed, in large part, due to intrinsic biogenic means: gregarious behavior in a group of Centrosaurus apertus. We infer that formation of the Hilda mega-bonebed was initiated as more than one thousand gregarious ceratopsians , and other large terrestrial vertebrates, drowned during a major coastal-plain flooding event. As floodwaters receded, individuals and groups of drowned Centrosaurus apertus became pooled across the floodplain. Exposed skeletons experienced large degrees of disarticulation and component elements were subject to breakage, varying amounts of abrasion, and rare tooth scarring as a result of month-to-multiyear exposure times, high rates of soft tissue rotting, and reworking due to scavenging and trampling. Stratigraphic and sedimentologic data, and comparisons with similar bonebeds at Dinosaur Provincial Park, suggest that the mega-bonebed assemblage was eventually buried during a subsequent coastal-plain flooding event. Data from the Hilda mega-bonebed are compatible with a previously proposed hypothesis that some centrosaurines exhibited east-west seasonal migratory behaviors . The existence of the Hilda mega-bonebed suggests that many of the stratigraphically close centrosaurine bonebeds at Dinosaur Park may also be parts of two or more mega-bonebeds in that area. Testing of this hypothesis may be possible using geochemical means. Introduction Alberta’s centrosaurine bonebeds, studied since the 1980s, have provided abundant evidence for gregarious behavior, mass mortalities and patterns of taphonomic modification within ceratopsians (Currie and Dodson 1984; Visser 1986; Ryan et al. 2001; Eberth and Getty 2005). Although numerous areas within the province are known to produce monotaxicto -monodominant centrosaurine bonebeds (e.g., Dinosaur Provincial Park, Grande Prairie, South Saskatchewan River, Drumheller, Scabby Butte), the majority of studies on these kinds of bonebeds has taken place at Dinosaur Provincial Park (DPP; Currie and Dodson 1984; Visser 1986; Ryan et al. 2001; Eberth and Getty 2005). There, at least 20 docu- 496 eberth, brinkman, & barkas mented centrosaurine bonebeds occur in a 44 m thick stratigraphic zone in the upper one-half of the Belly River Group (Eberth and Getty 2005: 501). Among these, 17 bonebeds contain Centrosaurus apertus and occur within the lowest 14 m of the Dinosaur Park Formation (DPFm; Eberth and Getty 2005). The centrosaur-bonebed rich lower 14 m of the DPFm at DPP is exposed over an area of approximately 50 km2. The occurrence of 17 ceratopsian bonebeds in this thin stratigraphic slice, and within such a limited area, has suggested to some bonebed researchers the possibility that some of these bonebeds in the Park formed during the same event(s) (e.g., Eberth and Getty 2005). Testing this hypothesis, however, has been difficult. The three-dimensional geomorphology of the badlands at DPP significantly limits one’s ability to trace individual bonebeds and their host beds over more than a few hundreds of meters. More importantly, bonebeds are frequently truncated by surfaces that mark paleochannel incision and other Cretaceous erosion/nondeposition events (Koster and Currie 1987; Wood 1989; Eberth and Hamblin 1993). Thus, attempts at physically correlating these previously known centrosaurine bonebeds have been largely unsuccessful . To date, it has been confirmed only that the Centrosaurus apertus bonebeds at DPP formed during at least two separate events (i.e., BB91a can be shown to stratigraphically overlie BB91; Eberth and Getty 2005). Because of the limitations on assessing the stratigraphic relationships among the centrosaurine bonebeds at DPP, a 1997 field study was conducted by the Royal Tyrrell Museum at ‘‘Hilda,’’ 80 km to the southeast along the South Saskatchewan River, 25 km west of the village of Hilda. Field notes from Wann Langston, Jr. (1959) and data from a small bonebed collection amassed under the direction of Don Taylor (Provincial Museum of Alberta, now the Royal Alberta Museum, 1964–1966) suggested that at...

Share