Megafloral change in the early and middle Paleocene in the Williston Basin, North Dakota, USA Daniel J. Peppe ⁎ Department of Geology, Baylor University, Waco, TX 76798-7354 USA abstract article info Article history: Received 27 January 2010 Received in revised form 14 September 2010 Accepted 29 September 2010 Available online 8 October 2010 Keywords: Paleocene Megafloral paleobotany Williston Basin Fort Union Formation Biostratigraphy K/Pg boundary Paleoenvironment Climate change Species richness This paper presents a quantitative analysis of megafloral changes in composition and diversity using collections of early and middle Paleocene floras (65.51 to ~ 58 Ma) in the Williston Basin of North Dakota, USA. Based on the floral composition and stratigraphic ranges of taxa, the Williston Basin floral record can be subdivided into three megafloral zones (WBI, WBII, and WBIII), each representing ≥ 1 myr. The floral record of the basin implies that local and regional paleoenvironmental and climatic changes contributed to transitions in the early and middle Paleocene plant communities. The Williston Basin floral record documents a decrease in species richness that mirrors a decrease in mean annual temperatures from the latest Cretaceous to middle Paleocene. These results, combined with previous work from the Hanna and Bighorn Basins, suggest that climate may have played an important role in patterns of floral diversity and plant community composition. Further, these data indicate that it took Paleocene plant communities in the Northern Great Plains millions of years to reach diversity levels common in the Cretaceous. © 2010 Elsevier B.V. All rights reserved. 1. Introduction The North American mid-continental, coal-bearing successions are arguably the best terrestrial Paleocene records in the world (Fig. 1), and are the ideal place to assess the terrestrial ecosystem's response to mass extinction and long term climatic change. The megafloral record of these sediments has been well documented for over 100 years (e.g., Newberry, 1868; Lesquereux, 1878; Knowlton, 1930; Brown, 1962; Hickey, 1977; Nichols and Ott, 1978; Hickey, 1980; Johnson, 1989; Wing et al., 1995; Manchester, 1999; Johnson, 2002; Dunn, 2003). In particular, studies focused on the Cretaceous- Paleogene (K/Pg) boundary intervals have documented major extinctions in the megafloral and pollen records (e.g., Tschudy et al., 1984; Johnson, 1989; Johnson et al., 1989; Johnson and Hickey, 1990; Hotton, 2002; Nichols and Johnson, 2002; Wilf and Johnson, 2004), found demonstrable changes in species composition and diversity across the boundary (e.g., Wolfe and Upchurch, 1986; Johnson, 2002; Wilf and Johnson, 2004), and suggested a correspondence between climatic and floral change (e.g., Wilf et al., 2003; Wilf and Johnson, 2004). However, the response of Paleocene plant communities to the K/Pg boundary extinctions has not yet been characterized fully. Assessing patterns of floral change through the Paleocene is difficult because most studies have been focused on floras from restricted geographic areas or time intervals (e.g., Johnson and Ellis, 2002) or were conducted without stratigraphic control or indepen- dent age constraints (e.g., Brown, 1962). To date, three studies have examined changes in plant assemblages and floral diversity through most of the Paleocene in the Bighorn and Hanna Basins of Wyoming (Hickey, 1980; Wing et al., 1995; Dunn, 2003). These studies suggest that changes in the Paleocene floral record roughly correspond to transitions in the North American Land Mammal Age (NALMA) boundaries. Two of the studies (Hickey, 1980; Dunn, 2003) docu- mented a decrease in species richness from the early to middle Paleocene and a general cooling trend through the sampled interval. These results led Hickey (1980) to hypothesize that the decrease in species richness was linked to cooling. The third study by Wing et al. (1995) showed a gradual increase in plant diversity through the Paleocene. Contrary to Hickey (1980), Wing et al. (1995) suggested that there was no congruence between plant species richness levels and changes in mean annual temperatures. Though the three aforementioned studies characterized much of the Paleocene floral record, none correlated the plant fossil record to either isotopic age determinations or to the geomagnetic polarity time scale. This lack of precise age control makes it difficult to assess if the patterns of plant community change happened at similar times in both basins. Furthermore, none assessed the Cretaceous floral record, preventing comparisons of Cretaceous and Paleocene floral composition and Palaeogeography, Palaeoclimatology, Palaeoecology 298 (2010) 224–234 ⁎ Tel.: + 1 2547102629; fax: + 1 2547102673. E-mail address: daniel.peppe@baylor.edu. 0031-0182/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.palaeo.2010.09.027 Contents lists available at ScienceDirect Palaeogeography, Palaeoclimatology, Palaeoecology journal homepage: www.elsevier.com/locate/palaeo