Refugial persistence and postglacial recolonization of North America by the cold-tolerant herbaceous plant Orthilia secunda

Previous phylogeographical and palaeontological studies on the biota of northern North America have revealed a complex scenario of glacial survival in multiple refugia and differing patterns of postglacial recolonization. Many putative refugial regions have been proposed both north and south of the ice sheets for species during the Last Glacial Maximum, but the locations of many of these refugia remain a topic of great debate. In this study, we used a phylogeographical approach to elucidate the refugial and recolonization history of the herbaceous plant species Orthilia secunda in North America, which is found in disjunct areas in the west and east of the continent, most of which were either glaciated or lay close to the limits of the ice sheets. Analysis of 596-bp of the chloroplast trnS-trnG intergenic spacer and five microsatellite loci in 84 populations spanning the species' range in North America suggests that O.secunda persisted through the Last Glacial Maximum (LGM) in western refugia, even though palaeodistribution modelling indicated a suitable climate envelope across the entire south of the continent. The present distribution of the species has resulted from recolonization from refugia north and south of the ice sheets, most likely in Beringia or coastal regions of Alaska and British Columbia, the Washington/Oregon region in the northwest USA, and possibly from the region associated with the putative 'ice-free corridor' between the Laurentide and Cordilleran ice sheets. Our findings also highlight the importance of the Pacific Northwest as an important centre of intraspecific genetic diversity, owing to a combination of refugial persistence in the area and recolonization from other refugia.

The Klondike goldfields and Pleistocene environments of Beringia

The Klondike goldfields of Yukon, Canada, contain a key record of Pleistocene Beringia, the region of Alaska, Siberia, and Yukon that remained largely unglaciated during the late Cenozoic. A concentration of mining exposures, with relict permafrost that is locally more than 700,000 years old, provides exceptional preservation of paleoenvironmental archives and a new perspective on the nature of paleoenvironments during the Pleistocene. A critical feature is the stratigraphic association of distal tephra beds with these paleoenvironmental archives, which facilitates their regional correlation and, in many cases, provides independent ages for the paleoenvironmental assemblages. Paleoenvironmental analyses of fossil arctic ground-squirrel middens and buried vegetation indicate the presence of cryoxerophilous ("steppe-tundra") vegetation growing on well-drained substrates with deep active layers (seasonal thaw depths) during cold intervals of the Pleistocene. Studies of full-glacial paleosols and cryostratigraphic relations of associated ground ice indicate the importance of active loess deposition and surface vegetation cover in maintaining the functionally distinct mammoth-steppe biome, which supported grazing mega-fauna populations, including mammoth, horse, and bison.

Northern peatland initiation lagged abrupt increases in deglacial atmospheric CH4

Peatlands are a key component of the global carbon cycle. Chronologies of peatland initiation are typically based on compiled basal peat radiocarbon (14C) dates and frequency histograms of binned calibrated age ranges. However, such compilations are problematic because poor quality 14C dates are commonly included and because frequency histograms of binned age ranges introduce chronological artefacts that bias the record of peatland initiation. Using a published compilation of 274 basal 14C dates from Alaska as a case study, we show that nearly half the 14C dates are inappropriate for reconstructing peatland initiation, and that the temporal structure of peatland initiation is sensitive to sampling biases and treatment of calibrated14C dates. We present revised chronologies of peatland initiation for Alaska and the circumpolar Arctic based on summed probability distributions of calibrated 14C dates. These revised chronologies reveal that northern peatland initiation lagged abrupt increases in atmospheric CH4 concentration at the start of the Bølling–Allerød interstadial (Termination 1A) and the end of the Younger Dryas chronozone (Termination 1B), suggesting that northern peatlands were not the primary drivers of the rapid increases in atmospheric CH4. Our results demonstrate that subtle methodological changes in the synthesis of basal 14C ages lead to substantially different interpretations of temporal trends in peatland initiation, with direct implications for the role of peatlands in the global carbon cycle.