Palaeogeography, Palaeoclimatology, Palaeoecology 286 (2010) 88–96 Contents lists available at ScienceDirect Palaeogeography, Palaeoclimatology, Palaeoecology j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / p a l a e o Regional differences in bone collagen δ13C and δ15N of Pleistocene mammoths: Implications for paleoecology of the mammoth steppe Paul Szpak a,⁎, Darren R. Gröcke b, Regis Debruyne a, Ross D.E. MacPhee c, R. Dale Guthrie d, Duane Froese e, Grant D. Zazula f, William P. Patterson g, Hendrik N. Poinar a a McMaster Ancient DNA Centre, Departments of Anthropology and Biology, McMaster University, Hamilton, Ontario, Canada L8S 4K1 Department of Earth Sciences, Durham University, Durham, United Kingdom DH1 3LE, UK Division of Zoology, American Museum of Natural History, New York, New York 10024, USA d Institute of Arctic Biology, University of Alaska Fairbanks, Fairbanks, Alaska, 99775-5940, USA e Department of Earth and Atmospheric Sciences, University of Alberta, Edmonton, Alberta, Canada T6G 2E3 f Yukon Palaeontology Program, Department of Tourism and Culture, Yukon Government, P.O. Box 2703, Whitehorse, Yukon Territory, Canada Y1A 2C6 g Department of Geological Sciences, University of Saskatchewan, Saskatoon, Saskatchewan, Canada S7N 5E2 b c a r t i c l e i n f o Article history: Received 24 July 2009 Received in revised form 9 December 2009 Accepted 23 December 2009 Available online 4 January 2010 Keywords: Stable isotopes Bone collagen Woolly mammoth Paleoecology Beringia Mammoth steppe a b s t r a c t In this study, we present bone collagen δ13C and δ15N values from a large set of Pleistocene woolly mammoths (Mammuthus primigenius) from Siberia, Alaska and Yukon. Overall, results for mammoth specimens from eastern Beringia (Alaska and Yukon) significantly differ, for both δ13C and δ15N values, from those from western Beringia (northeastern Siberia). In agreement with palynological, entomological, and physiographic data from the same regions, these isotopic differences strongly imply that the ‘mammoth steppe,’ the extensive ice-free region spanning northern Eurasia and northwestern North America, was ecologically variable along its east–west axis to a significant degree. Prior to the Last Glacial Maximum (LGM), the high-latitude portions of Siberia and the Russian Far East appear to have been colder and more arid than central Alaska and Yukon, which were ecologically more diverse. During the LGM itself, however, isotopic signatures of mammoths from eastern Beringia support the argument that this region also experienced an extremely cold and arid climate. In terms of overall temporal trend, Beringia thus went from a condition prior to the LGM of greater ecological variability in the east to one of uniformly cold and dry conditions during the LGM. © 2009 Elsevier B.V. All rights reserved. 1. Introduction During the last glaciation, sea levels were considerably lower than at present, and Alaska and Siberia were connected by the exposed Bering Land Bridge (BLB). Beringia, the more extensive region of which the BLB was a part, extended from the Northwest Territories in northern Canada to the Kolyma River in northeastern Siberia (Hoffecker and Elias, 2007). Beringia was in turn a major component of the mammoth steppe, an even larger megacontinental biome initially described by Guthrie (1968) as stretching continuously from western Europe to North America. The fauna of the Late Pleistocene mammoth steppe was markedly different than that of modern high-latitude ecosystems, with numerous large herbivores such as woolly mammoth (Mammuthus primigenius), bison (Bison sp.) and horse (Equus sp.) thriving across the ⁎ Corresponding author. Present address: Department of Anthropology, The University of Western Ontario, London, Ontario, Canada N6A 5C2. Tel.: + 1 289 396 6641; fax: + 1 519 661 2157. E-mail address: pszpak@uwo.ca (P. Szpak). 0031-0182/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.palaeo.2009.12.009 landscape (Guthrie, 1982; Guthrie, 1990). Faunal continuities suggest, but do not demonstrate, that the mammoth steppe was functionally similar across this vast area. Indeed, although low-growing herbaceous plants (grasses, sedges, forbs and sages) dominated the landscape of the mammoth steppe (Guthrie, 2001b), this biome was characterized by considerable regional variability in floral elements (Guthrie, 1990; Elias et al., 1997; Guthrie, 2001b; Zazula et al., 2007). More formally, Guthrie (1982) described the mammoth steppe as a mosaic of locally unique elements contributing to a series of related habitats, reflecting both the continuity of faunal elements and the variability in ecological conditions characteristic of this biome. The existence of marked environmental differences between eastern and western Beringia during the Late Pleistocene is of considerable faunistic and ecological interest. Guthrie (2001b) posited that central Beringia acted as an ecological ‘buckle’ of more mesic climatic conditions that served as a strong filter or even a complete barrier to some steppe-adapted species. Arguably, this mesic buckle prevented the xeric-adapted Eurasian woolly rhinoceros (Coelodonta antiquitatus), whose range extended at least as far east as Chukotka, from migrating eastward across the Bering Isthmus toward the Americas. Similarly, the range of the Pleistocene North American P. Szpak et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 286 (2010) 88–96 camel (Camelops hesternus) continued into the valley of the Yukon River, but apparently no further. In general, the area to the west of this proposed mesic zone has been reconstructed as cold and very dry during the interval leading up to the LGM, with the landscape dominated by herbaceous steppetundra. The Middle Valdai glacial period (50 ka–24 ka [thousand calendar years] BP) was characterized by relatively few major climatic oscillations, with a trend toward increasing aridity and decreasing temperature that apparently peaked during the LGM ca. 21.5 ka (Iacumin et al., 2000). Numerous lines of evidence, including pollen records (Lozhkin et al., 1993), macrobotanical remains (Schirrmeister et al., 2002), and invertebrate taxa (Kuzmina and Sher, 2006), support the argument that Western Beringia (northeastern Siberia) was extremely cold and arid during the LGM. Paleoenvironmental reconstructions from the Berelekh Site in the Allahovsk Region of the Sakha Republic (Yakutia) indicate a dry, steppe-tundra environment with herbaceous species (∼90%) dominating pollen spectra after the LGM (Müller-Beck, 1982). Further to the west, the Taimyr Peninsula was also characterized by cold and dry conditions before and during the LGM. Plant macrofossils from the northwest shore of Lake Taimyr dating to 34 ka, 31 ka and 20 ka BP indicate very dry conditions, with herbsteppe-tundra elements being most prevalent (Kienast et al., 2001). Pollen assemblages dating to the end of the Middle Valdai (30 ka– 22 ka BP) from the Taimyr Peninsula also indicate cold steppe-tundra (Kienel et al., 1999). Although during the Middle Valdai the Taimyr Peninsula exhibited somewhat heterogeneous environmental conditions, with three warm and two cold stages, this area can still be characterized as cold and arid, with consistently low mean annual precipitation (400–450 mm annually) throughout the region (Drozdov and Chekha, 2006). Paleoecological data for the area to the east of the mid-isthmian mesic buckle (Alaska and Yukon) also imply the predominance of cryoxerophilous herb-tundra or steppe-tundra, though with a greater proportion of mesic habitats than western Beringia. Pollen data from western Alaska indicate the dominance of graminoid herb-tundra flourishing under cold, dry conditions prior to and during the LGM (Ager, 2003). Plant macrofossils from an in situ vegetated surface that was buried by volcanic ash on the Seward Peninsula confirms the presence of a xerophilous sedge dominated herb-tundra in western Alaska during the LGM (Goetcheus and Birks, 2001). Plant and insect macrofossils from a variety of contexts in central and northern Yukon and dating from 30 ka to 16 ka BP are indicative of cold, dry herbaceous steppe conditions at the regional scale (Zazula et al., 2003). LGM pollen data from northeastern Yukon suggest extreme conditions with sparse fell-field tundra (Kozhevnikov and Ukraintseva, 1997). Loess deposits in western and interior Alaska also signal the predominance of cold, arid conditions through the Late Pleistocene (Muhs et al., 2003). The distribution of fossil insect faunas indicates diverse ecological conditions throughout eastern Beringia in the Late Pleistocene. Aquatic beetles were found to be abundant in eastern Beringia, but almost entirely absent from its western end (Elias et al., 2000). Furthermore, it has been suggested that southwestern Alaska was a refugium for mesic and hygrophilous beetle species before, during and after the LGM, while northern and interior Alaska contained a much higher number of xeric and steppe-adapted species (Elias, 1992). In general, paleoecological reconstructions indicate a higher level of environmental variability to the east of the Bering Isthmus, with evidence of both mesic and xeric conditions; by contrast, mesic indicator species are lacking for all of the Late Pleistocene in Siberia. 2. Stable isotopes in bone collagen Stable isotopes of bone collagen offer a direct way to analyze the diet and ecology of extinct and extant fauna. Because both carbon and 89 nitrogen in bone collagen are derived directly from the diet (Ambrose and Norr, 1993), the isotopic signatures recorded in these molecules are ideal for paleodietary and paleoecological studies. A number of factors can influence the δ15N values of bone collagen within a trophic level or a single taxon. It has been noted that species that excrete very concentrated urea in their urine tend to recycle nitrogen, which leads to elevated δ15N values (Ambrose and DeNiro, 1986). This is likely related to drought tolerance, since in some mammalian species (including African elephants) there is a strong negative correlation between δ15N values and local precipitation (Sealy et al., 1987; van der Merwe et al., 1990; Vogel et al., 1990; Gröcke et al., 1997; Schwarcz et al., 1999; Pate and Anson, 2008). An enrichment in plant δ15N values at more arid locales has also been observed, although this relationship is not as strong as it is in animals (Heaton, 1987). In addition to drought stress, nutritional or protein stress may lead to elevated δ15N values in animal tissues (Hobson et al., 1993; Fuller et al., 2005). Bone collagen δ13C values are informative with respect to the contribution of C3 and C4 forage to the diet (DeNiro and Epstein, 1978). Grazers and browsers can be differentiated on the basis of δ13C values, with the former having tissues that are less depleted in 13C than the latter (Ambrose and DeNiro, 1986). Moreover, because of the depletion in 13C that is characteristic of plants in heavily forested areas (van der Merwe and Medina, 1991), animals living in forested and open environments can also be identified using δ13C values in bones, teeth and other tissues (Cerling et al., 2004). With respect to plants, a negative correlation is thought to exist between annual rainfall and δ13C values in some C3 species (Stewart et al., 1995), but other studies have found only weak or no correlation (e.g. Swap et al., 2004). Low temperatures are also thought to produce more negative δ13C values in plants due to reduced CO2 uptake (Tieszen, 1991). Previous isotopic data from woolly mammoth bone collagen suggest regional differences in carbon and nitrogen isotopic signatures (Koch, 1991; Bocherens et al., 1994; Bocherens et al., 1996; Iacumin et al., 2000; Fox-Dobbs et al., 2008). Specifically, mammoths from Siberia tend to be more depleted in 13C and more enriched in 15N than those from Alaska. These patterns are suggestive of climatic or environmental differences that may mirror the data obtained from other paleoecological indicators. Our aim in this study was to establish, from isotopic evidence, whether any ecological or dietary differences could be distinguished between mammoths living in eastern and western Beringia. 3. Methods Bone collagen was extracted using a modified Longin method (Longin, 1971), as has been described previously (Szpak et al., 2009). Briefly, bone fragments (50–450 mg) were sampled using a hammer and chisel. Any visible foreign material was cleaned with a small brush and/or a brief treatment in an ultrasonic bath. Samples were soaked in chloroform–methanol (2:1 v:v, 4 ml) to remove lipids. Following cleaning and lipid extraction, samples were dried and reduced to fragments 1.0 mm to 2.0 mm in diameter. Bone fragments were soaked in 0.50 M HCl at room temperature until the bone was completely demineralized. They were rinsed with MQ water and treated with 0.1 M NaOH for 30 min to remove humic acids. Samples were then neutralized with MQ water, rinsed with 0.25 M HCl and finally with MQ water, leaving the insoluble collagen residue in a slightly acidic solution. The solution containing the insoluble residue was heated in sealed glass tubes at 80 ± 2 °C and then dried. The extracted collagen was then transferred into tin cups for isotopic analysis. Stable isotope values of bone collagen (δ13C and δ15N) and relative percentages of carbon and nitrogen were determined using a Thermo Finnigan DeltaPLUS XP continuous flow mass spectrometer coupled to a Costech elemental analyzer at McMaster University. Stable carbon and nitrogen isotope ratios are reported 90 P. Szpak et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 286 (2010) 88–96 versus VPDB and atmospheric nitrogen respectively. All samples were analyzed in at least duplicate and standard deviation was better than ±0.1 for δ13C and ±0.2 for δ15N measurements. 4. Results Mammoth samples were obtained primarily from three geographic regions (Fig. 1): eastern Beringia (central Alaska and Yukon), northcentral Siberia (Taimyr Peninsula) and northeastern Siberia (Yakutia or Sakha Republic). Fifty-eight samples of woolly mammoth bone collagen were analyzed for stable carbon and nitrogen isotope ratios, forty-seven of which have been radiocarbon dated (Guthrie, 2006; Debruyne et al., 2008). The specimens were derived from a large temporal span, ranging in age from N51 ka–5 ka BP. All isotopic measurements, relative percentages of carbon and nitrogen and collagen yields are presented in Table 1. It is generally accepted that the relative abundances of carbon and nitrogen (C:N ratio) of bone collagen can be used to determine whether postmortem alteration of the collagen has occurred. All samples analyzed had C:N ratios between 2.9 and 3.6 (see Table 1), which are typically assumed to be indicative of well-preserved collagen (DeNiro, 1985). Mammoth δ13Ccoll values ranged from − 23.2 to −20.3‰, and 15 δ Ncoll ranged from 4.4 to 11.4‰. Siberian mammoths recorded δ13Ccoll values that are more 13C depleted (−21.9 ± 0.5‰, N = 32) than mammoths from Alaska and Yukon (−20.7 ± 0.3‰, N = 26). Additionally, Siberian mammoths recorded more elevated δ15Ncoll values (9.3 ± 1.0) than Alaskan/Yukon mammoths (7.2 ± 1.3‰). These differences are graphically illustrated in Fig. 2. 5. Discussion 5.1. Stable nitrogen isotopes in Mammuthus primigenius The average δ15Ncoll values obtained for Mammuthus primigenius in this study (8.4±1.6‰) are relatively high for herbivores. Although no coeval species were analyzed in this study, data from previous analyses indicate that mammoths in general exhibit higher δ15Ncoll values than other herbivores (Bocherens et al., 1996; Iacumin et al., 2000). While a physiological explanation for the high δ15Ncoll values of Pleistocene mammoths has been suggested (Matheus et al., 2003), the nature of this mechanism has yet to be satisfactorily explained. Preserved stomach contents from frozen mammoth carcasses are dominated by high-fibre, low-protein forage (mainly graminoids), but also include the leaves of some other forbs, shrubs and a minor amount of moss (Ukraintseva et al., 1996). On the basis of digestive physiology, Guthrie and Stoker (1990) noted that mammoths were capable of tolerating diets composed of extremely low quantities of protein as long as sufficient quantities of forage were available. Because δ15Ncoll values are affected by the quality of dietary intake, this general trend toward elevated δ15Ncoll values in mammoths might reflect a relatively nutrient-poor and/or protein-deficient diet. It has been noted that herbaceous vegetation tends to produce relatively high δ15N values relative to woody vegetation (Delwiche et al., 1979). Modern Alaskan grasses in particular exhibit relatively high δ15N values, possibly due to rooting depth and organic matter utilization (Schulze et al., 1994). As such, a diet composed mainly of low quality herbaceous and graminoid forage is likely an important factor contributing to the elevated δ15Ncoll values observed in mammoths. Aside from generally high δ15Ncoll values in mammoths, a clear separation between Siberian and eastern Beringian mammoths was also evident and is statistically significant (Student's t-test; α = 7.17 for υ = 56, p b 0.0001). By contrast, there were no significant differences between Alaskan and Yukon samples (α = 1.34 for υ = 24, p = 0.19) or between samples from eastern Siberia and the Taimyr Peninsula (α = 0.28 for υ = 29, p = 0.78). This is suggestive of different ecological conditions on either side of the Bering Isthmus, with (as explained below) western Beringia experiencing more arid conditions throughout the Late Pleistocene than eastern Beringia. Water availability has been demonstrated to affect both plant and soil δ15N values (Delwiche et al., 1979; Heaton, 1987). Soils and vegetation at drier sites lose relatively less 15N and are characterized by more open cycling of nitrogen (Swap et al., 2004). Because the protein and nitrogen content of plants in arid environments tend to be lower than those in wetter regions, herbivores consuming these plants are more likely to recycle urea in order to conserve nitrogen, which would cause the tissues of these herbivores to be more enriched in 15N (Sealy et al., 1987). Recently, a strong negative correlation between δ15N values and modern annual precipitation was found in woolly mammoth hair Fig. 1. Map showing sample locations. (1) Taimyr Peninsula, (2) Yakutia, (3) Gydan Peninsula, (4) Wrangel Island, and (5) Central Alaska and Yukon. 91 P. Szpak et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 286 (2010) 88–96 Table 1 Isotopic and relevant contextual data for all analyzed woolly mammoths. Sample # Locality Site Latitude Longitude Age (years BP) % collagen %C %N C:N δ13C (‰) δ15N (‰) AM 1114 AM 1208 AM 2446 AM 523 AM 8052 NMC 42135 NMC 6746 NMC 42292 NMC 49562 NMC 49927 NMC 49928 YU 130.2 YU 133.18 YU 136.9 YU 137.3 YU 3.133 YU 3.135 YU 3.136 YU 3.19 YU 3.229 YU 3.256 YU 5.130 YU 5.46 YU 5.69 YU 52.36 YU 57.1 Eastern Beringia (Alaska) Eastern Beringia (Alaska) Eastern Beringia (Alaska) Eastern Beringia (Alaska) Eastern Beringia (Alaska) Eastern Beringia (Alaska) Eastern Beringia (Alaska) Eastern Beringia (Yukon) Eastern Beringia (Yukon) Eastern Beringia (Yukon) Eastern Beringia (Yukon) Eastern Beringia (Yukon) Eastern Beringia (Yukon) Eastern Beringia (Yukon) Eastern Beringia (Yukon) Eastern Beringia (Yukon) Eastern Beringia (Yukon) Eastern Beringia (Yukon) Eastern Beringia (Yukon) Eastern Beringia (Yukon) Eastern Beringia (Yukon) Eastern Beringia (Yukon) Eastern Beringia (Yukon) Eastern Beringia (Yukon) Eastern Beringia (Yukon) Eastern Beringia (Yukon) Eastern Beringia (N=26) Western Beringia (Siberia) Western Beringia (Siberia) Western Beringia (Siberia) Western Beringia (Siberia) Western Beringia (Siberia) Western Beringia (Siberia) Western Beringia (Siberia) Western Beringia (Siberia) Western Beringia (Siberia) Western Beringia (Siberia) Western Beringia (Siberia) Western Beringia (Siberia) Western Beringia (Siberia) Western Beringia (Siberia) Western Beringia (Siberia) Western Beringia (Siberia) Western Beringia (Siberia) Western Beringia (Siberia) Western Beringia (Siberia) Western Beringia (Siberia) Western Beringia (Siberia) Western Beringia (Siberia) Western Beringia (Siberia) Western Beringia (Siberia) Western Beringia (Siberia) Western Beringia (Siberia) Western Beringia (Siberia) Western Beringia (Siberia) Western Beringia (Siberia) Western Beringia (Siberia) Western Beringia (Siberia) Western Beringia (Siberia) Western Beringia (N = 32) Ester Creek Sullivan Creek Cripple Creek Cleary Creek Cleary Creek Eldorado Creek Tanana Dawson Paron's Lake Dawson Dawson Quartz Creek Whitman Gulch Sulphur Creek Whitman Gulch Finning Finning Finning Finning Finning Finning Hunker Creek Hunker Creek Hunker Creek Indian River VGFN Foot 64° 65° 64° 65° 65° 64° 65° 64° – 64° 64° 63° 63° 63° 63° 63° 63° 63° 63° 63° 63° 63° 63° 63° 63° 68° 148° 00′W 151°W 148°W 147° 30′W 147° 30′W 147° 40′W 151° 54′W 139° 25′W – 139° 25′W 139° 25′W 139° 02′W 138° 38′W 138° 50′W 138° 38′W 138° 15′W 138° 15′W 138° 15′W 138° 15′W 138° 15′W 138° 15′W 139° 02′W 139° 02′W 139° 02′W 139° 19′W 140° 32′W 59,583 15,047 30,889 47,030 22,019 34,276 27,745 41,782 N50,867 50,002 N49,485 41,653 36,537 49,985 43,724 33,495 49,709 33,598 48,491 13.2 11.4 3.0 2.9 19.8 13,4 14.5 13.7 23.3 19.9 19.7 13.3 17.8 12.9 13.0 22.8 16.5 18.6 19.5 22.9 10.4 14.5 11.4 13.2 12.0 n.d. 40.5 34.7 38.5 25.8 36.9 37.9 35.8 39.4 40.2 37.8 37.3 35.0 39.5 41.1 31.0 38.5 39.7 30.3 38.3 37.4 39.7 33.5 37.7 36.7 36.8 36.7 14.9 12.2 13.0 8.7 13.6 14.0 13.4 13.5 14.0 13.3 13.0 12.4 14.8 14.5 11.0 13.6 13.8 10.8 13.5 13.3 14.0 12.5 13.3 13.0 13.3 13.9 3.2 3.3 3.5 3.5 3.2 3.2 3.1 3.4 3.4 3.3 3.3 3.3 3.1 3.3 3.3 3.3 3.4 3.3 3.3 3.3 3.3 3.1 3.3 3.3 3.2 3.1 Arilakh Arilakh Lake Taimyr Cape Sablera Ari Mas Arilakh Arilakh Arilakh Baikura-Turku Lake Taimyr Lake Taimyr Taimyr Peninsula Baikura-Turku Baikura-Turku Baikura-Turku Soposhnaya Cape Sablera Popigay Popigay Oymyakon Berelekh Berelekh Berelekh Berelekh Berelekh Berelekh Berelekh Berelekh Berelekh Gydan Peninsula Sanga-Yuriakh Wrangel Island 74° 25′N 74° 25′N 74° 30′N 74° 35′N 72°N 74° 25′N 74° 25′N 74° 25′N 74° 15′N – – – 73° 45′N 73° 45′N 73° 45′N 72° 30′N 73° 24′N 72° 38′N 72° 38′N 63° 30′N 70° 33′N 70° 33′N 70° 33′N 70° 33′N 70° 33′N 70° 33′N 70° 33′N 70° 33′N 70° 33′N 71°N 79°E 71° 20′N 71°N 10.7 14.7 12.2 16.3 7.6 22.4 24.0 15.2 20.1 17.2 20.0 22.3 26.2 17.9 14.7 20.9 25.2 23.8 23.4 23.9 20.4 18.2 26.4 23.4 23.0 11.9 24.4 16.2 19.0 18.9 18.6 15.5 38.2 39.1 40.5 37.1 41.1 38.2 38.8 39.8 38.4 36.6 39.9 35.0 36.7 38.2 35.4 36.4 37.3 36.8 32.0 36.7 39.8 37.8 38.1 37.8 36.4 43.7 40.5 38.5 33.9 44.1 35.9 37.9 14.2 14.5 15.0 13.7 15.2 13.4 14.5 12.8 13.6 12.9 14.2 12.2 13.6 13.5 12.5 11.9 13.1 13.2 11.1 13.6 14.6 13.5 13.6 13.2 12.9 15.0 14.8 13.4 12.0 15.4 12.6 13.6 3.1 3.1 3.2 3.2 3.2 3.3 3.1 3.6 3.3 3.3 3.3 3.3 3.1 3.3 3.3 3.6 3.3 3.3 3.4 3.1 3.2 3.3 3.3 3.3 3.3 3.4 3.2 3.4 3.3 3.3 3.3 3.3 − 20.5 − 20.8 − 20.8 − 20.8 − 21.0 − 20.7 − 21.0 − 20.6 − 21.3 − 20.4 − 20.3 − 20.5 − 20.7 − 20.8 − 20.5 − 20.5 − 20.6 − 20.9 − 20.6 − 20.5 − 20.6 − 20.6 − 21.0 − 20.7 − 20.9 − 21.2 − 20.7 ± 0.2 − 22.5 − 21.7 − 21.4 − 21.0 − 21.8 − 22.0 − 22.0 − 23.2 − 21.7 − 21.9 − 21.0 − 22.1 − 21.9 − 21.7 − 21.4 − 22.4 − 20.8 − 21.6 − 22.0 − 21.7 − 21.9 − 22.2 − 22.0 − 22.0 − 22.1 − 21.5 − 21.9 − 22.4 − 22.2 − 21.1 − 22.6 − 22.2 −21.9 ± 0.5 5.0 5.9 6.4 4.4 7.0 7.8 9.7 7.9 5.8 8.7 6.9 6.6 8.1 7.5 7.3 7.2 6.4 7.1 5.7 7.1 6.8 6.9 9.4 9.0 6.4 9.3 7.2±1.3 9.1 10.7 9.4 10.6 10.5 10.0 11.4 7.8 8.9 9.7 10.7 8.1 10.0 8.9 8.7 8.0 9.1 8.4 7.5 8.1 9.5 9.4 9.9 9.4 9.1 9.2 8.9 8.9 9.0 9.8 10.9 8.6 9.3 ± 1.0 2000/173 2000/174 2000/183 2000/198 2001/412 2002/472 2002/473 2002/594 2005/897 2005/900 2005/901 2005/907 2005/915 2005/916 2005/917 2005/928 2005/945 2005/988 2005/999 2006/001 BER 11 BER 12 BER 13 BER 16 BER 20 BER 28 BER 5 BER 7 BER 9 GDY1 SYU 3 WR2 50′N 10′N 60′N 10′N 10′N 50′N 18′N 03′N 03′N 03′N 49′N 43′N 44′N 43′N 50′N 50′N 50′N 50′N 50′N 50′N 59′N 59′N 59′N 46′N 11′N 107° 45′E 107° 45′E 100° 30′E 100° 30′E 101°E 107° 45′E 107° 45′E 107° 45′E 101° 20′E – – – 102° 00′E 102° 00′E 102° 00′E 108° 00′E 101° 39′E 106° 40′E 106° 40′E 142° 45′E 149° 03′E 149° 03′E 149° 03′E 149° 03′E 149° 03′E 149° 03′E 149° 03′E 149° 03′E 149° 03′E 18,136 151°E 179°W from northern Siberia (Iacumin et al., 2006). We suggest that the findings of this study are indicative of a similar pattern in Pleistocene mammoths on a larger scale, with the high δ15Ncoll values observed west of the Bering Isthmus being characteristic of drier steppe-tundra environments. Conversely, lower δ15Ncoll values of mammoths inhabiting eastern Beringia suggest that this area was more ecologically variable, containing more mesic habitats. On the basis of the dimensions of glaciers and ice sheets, as well as their equilibrium line altitudes, western Beringia was likely more arid or colder (or both) than eastern Beringia in the Late Pleistocene (Elias 33,436 36,990 27,098 N51,282 34,944 N51,652 13,795 32,640 32,672 18,466 N47,755 N52,803 51,013 43,920 33,169 20,763 44,649 32,312 40,246 24,008 N54,260 44,800 14,545 14,416 14,146 18,136 N51,652 5007 et al., 2000). Zazula et al. (2007) examined paleoecological proxies dating to ∼28 ka BP from central Yukon and found that it was characterized by a high degree of local level ecological heterogeneity within a regional environment dominated by xeric steppe-tundra. Steppe-indicator beetle species were ubiquitous in Late Pleistocene assemblages from Siberia, even during interglacials, while this is not the case for eastern Beringia (Elias and Kuzmina, 2008). This suggests that northern Siberia was drier than eastern Beringia throughout much of the Late Pleistocene, which fits well with the elevated δ15Ncoll values observed for Siberian mammoths in this study. This does not 92 P. Szpak et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 286 (2010) 88–96 Fig. 2. Plot of δ13C and δ15N for mammoth bone collagen. Closed circles represent samples from Siberia, open circles represent samples from eastern Beringia (Alaska and Yukon). Additional data from Alaska (open diamond — Koch, 1991; open triangle — Fox-Dobbs et al., 2008; open square — Bocherens et al., 1994), Yakutia (closed diamond — Iacumin et al., 2000; closed square — Bocherens et al., 1996) and the Taimyr Peninsula (closed triangle — Iacumin et al., 2000) are shown ±one standard deviation. suggest that dry conditions and steppe-tundra were absent from eastern Beringia, but that moister habitats, characterized by more mesic vegetation, were more common than they were throughout Siberia. Thus, mammoths in eastern Beringia would have been more likely to encounter and feed in these mesic zones, which is reflected in the isotopic composition of their tissues. 5.2. Stable carbon isotopes in Mammuthus primigenius The δ13Ccoll values for mammoths in this study are indicative of a diet based entirely on C3 vegetation, which is consistent with paleobotanical reconstructions based on pollen and plant macrofossils from the region (Wooller et al., 2007). While the specimens analyzed in this study showed a very limited range in δ13Ccoll values, a clear distinction between mammoths on either side of the Bering Isthmus was observed, with Siberian mammoths being more depleted in 13C than eastern Beringian mammoths. A Student's t-test confirms that this difference is statistically significant (α = 10.68 for υ = 56, p b 0.0001). Again, there was no significant difference between Alaskan and Yukon samples (α = 0.98 for υ = 24, p = 0.34) or between samples from eastern Siberia and the Taimyr Peninsula (α = 1.49 for υ = 29, p = 0.15). The fact that the Siberian samples recorded more 13C-depleted values was surprising since it has generally been noted that plant δ13C values are positively correlated with aridity (Stewart et al., 1995). In addition to this, Wooller et al. (2007) examined δ13C values of plants from eastern Beringia and found that plants inhabiting more arid environments exhibited higher δ13C values than plants from wetter environments. Therefore, given the observed difference and interpretation in δ15Ncoll values, we would expect to see higher δ13Ccoll values in Siberia mammoths (i.e., those with the highest δ15Ncoll values). In this case, however, the exact opposite was observed. There are likely several factors that combine to produce these seemingly conflicting values obtained for mammoth δ15Ncoll and δ13Ccoll. While the data and interpretation of Wooller et al. (2007) suggest a positive correlation between δ13C and aridity, they could only distinguish between wet and dry locations when the plant δ13C value was b−28‰. Based on typical carbon isotope enrichment factors for mammalian herbivores (following Bocherens and Drucker, 2003), mammoths would have been consuming plants with δ13C values between − 25‰ and −20‰. Thus, based on the model of Wooller et al. (2007) the δ13Ccoll values observed in this study are not likely to be indicative of either dry or wet conditions. A large number of studies from a diverse geographical area have examined δ13C values in tree rings and found positive correlations between δ13C and temperature, although the basis for this relationship is not known (Heaton, 1999). Nikolaev et al. (2004) analyzed modern woody and herbaceous plants from Yakutia (Siberia) and found weak or no correlation between plant δ13C and precipitation/ relative humidity. Instead, they suggested that temperature was the controlling factor with respect to plant δ13C values rather than moisture availability. Because of atmospheric circulation and topography, temperatures in Siberia are generally much colder than are locations in Alaska and Yukon at similar latitudes. In the northern hemisphere, weather is strongly influence by the two centers of semi-permanent atmospheric circulation: the Siberian High and Aleutian Low. The Siberian High produces extremely low temperatures in northeast Asia, including Siberia (Elias et al., 2000), while the Aleutian Low has a moderating effect on North American temperatures, bringing warm Pacific air masses into Alaska (Hoffecker and Elias, 2007). Furthermore, the Brooks and Richardson mountain ranges in the far north of North America act as barriers against cold Arctic air masses in the interior of Alaska and Yukon. Continental northeast Asia is directly exposed to these Arctic air masses and as a result, temperatures are significantly lower in Siberia (Elias et al., 2000). The atmospheric circulation affecting Late Pleistocene Beringia was likely similar to that of the present day (Alfimov and Berman, 2001). If the assertion that temperature affects δ13C values in high-latitude plants, the δ13C values observed for mammoths in this study may be a reflection of regional differences in mean annual temperature. The factors affecting δ13C values in Arctic plants are, however, not well understood. Additional data from North America, Europe and Asia need to be collected to document potential environmental influences on highlatitude plant δ13C values. In addition, isotopic analysis of plant macrofossils preserved in frozen carcasses of megafauna would greatly aid in the understanding of carbon cycling in Arctic plants throughout the Late Pleistocene. An alternative, but related, explanation for this distinction in δ13Ccoll values in mammoths concerns variable reliance on seasonal fat P. Szpak et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 286 (2010) 88–96 Fig. 3. Mammoth δ13Ccoll plotted against time for Beringian mammoths (A) and western Beringian mammoths (C). Mammoth δ15Ncoll plotted against time for eastern Beringian mammoths (B) and western Beringian mammoths (D). 93 94 P. Szpak et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 286 (2010) 88–96 stores. Mammoths generally tend to exhibit slightly more 13Cdepleted δ13Ccoll values than do coeval herbivores. Because fat is more depleted in 13C than bone collagen (Ambrose and Norr, 1993), the more negative δ13Ccoll values in mammoths relative to other herbivores could be the result of the metabolism of stored fat during the winter (Bocherens et al., 1996; Bocherens, 2003). Seasonal reliance on fat reserves in mammoths has been widely discussed (Guthrie, 1990; Guthrie, 2001a; Guthrie, 2001b). Forage available during winter months would have been far below even maintenance levels, which suggests that winter forage would act only as a buffer to stave off rapid burning of fat reserves (Guthrie, 2001b). A mummified mammoth recently recovered from Yakutia possessed a very large fat storage area extending from the base of the skull into the region between the scapulae and onto either side of the body (Boeskorov et al., 2007). Long seasonal migrations by mammoths has been suggested to account for the lack of forage available during the winter months (e.g. Churcher, 1980; Colvinaux and West, 1984), but evidence for this is lacking (but see Sharp et al., 2003). On the basis of dental evidence for winter deaths and the extraordinarily high metabolic costs for migrations in mammoths due to biomechanical limitations, Guthrie (1990) suggests that mammoths did not undertake seasonal migrations. Given these conflicting interpretations, it is difficult to determine whether and to what extent mammoths may have migrated, seasonally or otherwise. Regardless, fat storage would have been of particular importance for mammoths during the winter months when forage was in short supply. Even during periods of winter hibernation, the bone of polar bears and grizzly bears undergoes active remodeling (Lennox and Goodship, 2008; McGee et al., 2008). As such, it is reasonable to suggest that the bone of mammoths was also remodeled during winter months; this process would have required the routing of carbon from stored fat into bone. This would in turn lead to more 13C-depleted values in animals that relied more heavily on stored fat during the winter. Because the turnover rate for bone collagen is ten years (Stenhouse and Baxter, 1979) or more (Hedges et al., 2007), we would expect to see several seasons worth of metabolized fats represented in the isotopic composition of this tissue. In the Late Pleistocene, Siberian winters were colder and summers were warmer than they are today (Kienast et al., 2005). Therefore, the fact that Siberian mammoths exhibited significantly more negative δ13Ccoll values suggests that this may be the result of a higher reliance on stored fat during winter relative to eastern Beringian mammoths. Iacumin et al. (2005) performed incremental δ13C and δ15N analyses of mammoth hairs and observed regular negative peaks in δ13C, which they attribute to the consumption of different types of plants. Interestingly, these negative peaks are also narrower than the positive peaks, suggesting that they represent periods of restricted hair growth. In modern large arctic grazers, such as the musk ox (Ovibos moschatus), hair growth is significantly restricted during the winter months (Flood et al., 1989). The pattern observed in the mammoth hair δ13C and δ15N presented by Iacumin et al. (2005) can also be explained as a result of an increased reliance on 13C-depleted fat during the winter. 5.3. Temporal variability in Mammuthus primigenius δ13Ccoll and δ15Ncoll When interpreting isotopic data, it is important to keep in mind that temporal trends are regionally specific (Fox-Dobbs et al., 2008). Furthermore, different taxa likely have variable physiological responses to environmental stresses (e.g. Levin et al., 2006), and trends should only be examined within a species and a region. According to the large dataset presented by Richards and Hedges (2003), the most significant changes in stable carbon isotope values of mammalian bone collagen appear to have occurred following the LGM. The δ13Ccoll values exhibited little variability prior to 15 ka BP, following atmospheric CO2 concentrations, which also show relative stability. With respect to mammoths in particular, Iacumin et al. (2000) examined stable carbon and nitrogen isotopes in mammoth bone collagen (Siberia and the Russian plain) and found that δ15Ncoll values shifted around 14 ka BP, probably due to a marked increase in precipitation and associated vegetation changes. The dataset contains large temporal gaps for any given region (e.g. Yakutia, Taimyr Peninsula, Russian Plain), making interpretation difficult. Although data for mammoth bone collagen are omitted by Richards and Hedges (2003), they suggest that their data are in general agreement with those of Iacumin et al. (2000). When data are examined with respect to time, it is apparent that the separation in both δ13Ccoll and δ15Ncoll between Siberian and eastern Beringian mammoths is temporally consistent (see Fig. 3). On this basis, we suggest that Siberia was on average colder and more arid than central Alaska and Yukon prior to the LGM. It is difficult to assess the scale of this difference in Siberia during the LGM because relevant samples are lacking. By contrast, in eastern Beringia, for which we have better temporal coverage, the most negative δ13Ccoll and most elevated δ15Ncoll values are associated with the LGM (see Fig. 3). Indeed, it is only for the period of the LGM that the pattern of isotopic results for eastern Beringia strongly converges with that of western Beringia. Although limited, our data are in agreement with the argument, noted previously, that interior Alaska and Yukon had to have been extremely cold and dry at this time (Goetcheus and Birks, 2001; Hoffecker and Elias, 2007). Although only one of our analyzed samples postdates the LGM, δ13Ccoll and δ15Ncoll values closely correspond to those observed for samples from eastern Beringia prior to the LGM. While not statistically significant, this result does suggest that mammoths responded relatively well to the extreme conditions of the LGM. This is in agreement with phylogeographic data from the same mammoth samples published by Debruyne et al. (2008), which show that the LGM had only a very limited impact on the overall genetic diversity of mammoths living at high latitudes in both Eurasia and North America. 6. Conclusion This study demonstrates the utility of isotopic analysis of mammalian bone collagen for the study of regional and temporal environmental variability. Many lines of evidence establish that ecologically variable conditions prevailed across the mammoth steppe during the Pleistocene, with substantial regional differences in precipitation and temperature. According to the data presented here, Siberia was colder and more arid than central Alaska and Yukon prior to the LGM; however, following the onset of the LGM, eastern Beringia likewise became much cooler and drier. Environmental change at this time was not fatal to the mammoths of either region; they persisted, evidently unscathed at least in terms of their genetic diversity, until the time of the Pleistocene/Holocene transition. Acknowledgements We thank Martin Knyf, Carsten Schwarz and Krysta L. Bedient for providing technical assistance. This manuscript was improved due to insightful discussions with Alison M. Devault, Martin Knyf, Henry P. Schwarcz and Christine D. White. Bernard Buiges and Alexei Tikhonov provided samples. The quality of this manuscript was improved by the comments of two anonymous reviewers. This project was supported by an NSERC Discovery Grant (# 288321) and an SSHRC Research Grant (# 646-2006-1097) awarded to DRG, the Talisman Energy Inc. Jim Buckee Research Support Program (WPP), a Natural Sciences and Engineering Research Council of Canada (grant #299103-2004 HNP), the Social Sciences and Humanities Research Council of Canada (grant #410-2004-0579 HNP), the Government of Ontario Early Researcher Award program (HNP), the Canadian Research Chairs program (HNP) and McMaster University. Author P. Szpak et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 286 (2010) 88–96 Contributions: DRG, HNP and RDM designed the project. PS performed the research. HNP and RDM collected and contributed samples. PS and DRG analyzed the data. PS, DRG, HNP, RD, WPP and RDM wrote the paper. GDZ, RDG and DF provided critical assessment of the palaeoecological interpretation. References Ager, T.A., 2003. Late Quaternary vegetation and climate history of the central Bering land bridge from St. Michael Island, western Alaska. Quat. Res. 60, 19–32. Alfimov, A.V., Berman, D.I., 2001. Beringian climate during the Late Pleistocene and Holocene. Quat. Sci. Rev. 20, 127–134. Ambrose, S.H., DeNiro, M.J., 1986. The isotopic ecology of east African mammals. Oecologia 69, 395–406. Ambrose, S.H., Norr, L., 1993. Experimental evidence for the relationship of the carbon isotope ratios of whole diet and dietary protein to those of bone collagen and carbonate. In: Lambert, J.B., Grupe, G. (Eds.), Prehistoric Human Bone: Archaeology at the Molecular Level. Springer-Verlag, New York, pp. 1–37. Bocherens, H., Fizet, M., Mariotti, A., Gangloff, R.A., Burns, J.A., 1994. Contribution of isotopic biogeochemistry (13C, 15N, 18O) to the paleoecology of mammoths (Mammuthus primigenius). Hist. Biol. 7, 187–202. Bocherens, H., Pacaud, G., Lazarev, P.A., Mariotti, A., 1996. Stable isotope abundances (13C, 15N) in collagen and soft tissues from Pleistocene mammals from Yakutia: implications for the palaeobiology of the Mammoth Steppe. Palaeogeogr. Palaeoclimatol. Palaeoecol. 126, 31–44. Bocherens, H., 2003. Isotopic biogeochemistry and the paleoecology of the mammoth steppe fauna. In: Reumer, W.F., Braber, F., Mol, D., de Vos, J. (Eds.), Advances in Mammoth Research, Proceedings of the Second International Mammoth Conference, May 16–20 1999. Deinsea, Rotterdam, pp. 57–76. Bocherens, H., Drucker, D., 2003. Trophic level isotopic enrichment of carbon and nitrogen in bone collagen: case studies from recent and ancient terrestrial ecosystems. Int. J. Osteoarchaeol. 13, 46–53. Boeskorov, G., Tikhonov, A., Lazarev, P., 2007. A new find of a mammoth calf. Dokl. Biol. Sci. 417, 480–483. Cerling, T.E., Hart, J.A., Hart, T.B., 2004. Stable isotope ecology in the Ituri Forest. Oecologia 138, 5–12. Churcher, C.S., 1980. Did the North America mammoths migrate? Can. J. Anthropol. 1, 103–106. Colvinaux, P.A., West, F.H., 1984. The Beringian ecosystem. Q. Rev. Archaeol. 5, 10–16. Debruyne, R., Chu, G., King, C.E., Bos, K., Kuch, M., Schwarz, C., Szpak, P., Gröcke, D.R., Matheus, P., Zazula, G., Guthrie, D., Froese, D., Buigues, B., de Marliave, C., Flemming, C., Poinar, D., Fisher, D., Southon, J., Tikhonov, A.N., MacPhee, R.D.E., Poinar, H.N., 2008. Out of America: ancient DNA evidence for a New World origin of Late Quaternary woolly mammoths. Curr. Biol. 18, 1320–1326. Delwiche, C.C., Zinke, P.J., Johnson, C.M., Virginia, R.A., 1979. Nitrogen isotope distribution as a presumptive indicator of nitrogen fixation. Bot. Gaz. 140, S65–S69. DeNiro, M.J., Epstein, S., 1978. Influence of diet on the distribution of carbon isotopes in animals. Geochim. Cosmochim. Acta 42, 495–506. DeNiro, M.J., 1985. Postmortem preservation and alteration of in vivo bone collagen isotope ratios in relation to palaeodietary reconstruction. Nature 317, 806–809. Drozdov, N.I., Chekha, V.P., 2006. The mammoth fauna and environmental changes on the Taimyr Peninsula from the Middle Pleistocene through the Holocene. Anthropol. Archeol. Eurasia 2, 23–34. Elias, S.A., 1992. Late Quaternary beetle faunas of southwestern Alaska: evidence of a refugium for mesic and hygrophilous species. Arct. Alp. Res. 24, 133–144. Elias, S.A., Short, S.K., Birks, H.H., 1997. Late Wisconsin environments of the Bering Land Bridge. Palaeogeogr. Palaeoclimatol. Palaeoecol. 136, 293–308. Elias, S.A., Berman, D., Alfimov, A., 2000. Late Pleistocene beetle faunas of Beringia: where east met west. J. Biogeogr. 27, 1349–1363. Elias, S.A., Kuzmina, S., 2008. Palaeontology. In: Jolivet, P., Santiago-Blay, J., Schmitt, M. (Eds.), Research on Chrysomelidae. Brill Academic Publishers, Leiden, pp. 174–193. Flood, P.F., Stalker, M.J., Rowell, J.E., 1989. The hair follicle density and seasonal shedding cycle of the muskox (Ovibos moschatus). Can. J. Zool. 67, 1143–1147. Fox-Dobbs, K., Leonard, J.A., Koch, P.L., 2008. Pleistocene megafauna from eastern Beringia: paleoecological and paleoenvironmental interpretations of stable carbon and nitrogen isotope and radiocarbon records. Palaeogeogr. Palaeoclimatol. Palaeoecol. 261, 30–46. Fuller, B.T., Fuller, J.L., Sage, N.E., Harris, D.A., O'Connell, T.C., Hedges, R.E.M., 2005. Nitrogen balance and δ15N: why you're not what you eat during nutritional stress. Rapid Commun. Mass Spectrom. 19, 2497–2506. Goetcheus, V.G., Birks, H.H., 2001. Full-glacial upland tundra vegetation preserved under tephra in the Beringia National Park, Seward Peninsula. Alaska. Quat. Sci. Rev. 20, 135–147. Gröcke, D.R., Bocherens, H., Mariotti, A., 1997. Annual rainfall and nitrogen-isotope correlation in macropod collagen: application as a palaeoprecipitation indicator. Earth Planet. Sci. Lett. 153, 279–285. Guthrie, R.D., 1968. Paleoecology of the large-mammal community in interior Alaska during the Late Pleistocene. Am. Midl. Nat. 79, 346–363. Guthrie, R.D., 1982. Mammals of the mammoth steppe as paleoenvironmental indicators. In: Hopkins, D.M., MatthewsJr. Jr., J.V., Schweger, C.E., Young, S.B. (Eds.), Paleoecology of Beringia. Academic Press, New York, pp. 179–191. Guthrie, R.D., 1990. Frozen Fauna of the Mammoth Steppe: The Story of Blue Babe. University of Chicago Press, Chicago. 95 Guthrie, R.D., Stoker, S., 1990. Paleoecological significance of mummified remains of Pleistocene horses from the North Slope of the Brooks Range, Alaska. Arctic 43, 267–274. Guthrie, R.D., 2001a. Reconstruction of woolly mammoth life history. In: Cavarretta, G., Gioia, P., Mussi, M., Palombo, M.R. (Eds.), The World of Elephants. Proceedings of the First International Congress, Consiglio Nazionale delle Ricerche, Rome, pp. 276–279. Guthrie, R.D., 2001b. Origin and causes of the mammoth steppe: a story of cloud cover, woolly mammal tooth pits, buckles, and inside-out Beringia. Quat. Sci. Rev. 20, 549–574. Guthrie, R.D., 2006. New carbon dates link climatic change with human colonization and Pleistocene extinctions. Nature 441, 207–209. Heaton, T.H.E., 1987. The 15N/14N ratios of plants in South Africa and Namibia: relationship to climate and coastal/saline environments. Oecologia 74, 236–246. Heaton, T.H.E., 1999. Spatial, species, and temporal variations in the 13C/12C ratios of C3 plants: implications for palaeodiet studies. J. Archaeol. Sci. 26, 637–649. Hedges, R.E.M., Clement, J.G., Thomas, D.L., O'Connell, T.C., 2007. Collagen turnover in the adult femoral mid-shaft: Modeled from anthropogenic radiocarbon tracer measurements. Am. J. Phys. Anthropol. 133, 808–816. Hobson, K.A., Alisauskas, R.T., Clark, R.G., 1993. Stable-nitrogen isotope enrichment in avian tissues due to fasting and nutritional stress: implications for isotopic analyses of diet. Condor 95, 388–394. Hoffecker, J.F., Elias, S.A., 2007. Human Ecology of Beringia. Columbia University Press, New York. Iacumin, P., Nikolaev, V., Ramigni, M., 2000. C and N stable isotope measurements on Eurasian fossil mammals, 40 000 to 10 000 years BP: herbivore physiologies and palaeoenvironmental reconstruction. Palaeogeogr. Palaeoclimatol. Palaeoecol. 163, 33–47. Iacumin, P., Davanzo, S., Nikolaev, V., 2005. Short-term climatic changes recorded by mammoth hair in the Arctic environment. Palaeogeogr. Palaeoclimatol. Palaeoecol. 218, 317–324. Iacumin, P., Davanzo, S., Nikolaev, V., 2006. Spatial and temporal variations in the 13C/ 12 C and 15N/14N ratios of mammoth hairs: palaeodiet and palaeoclimatic implications. Chem. Geol. 231, 16–25. Kienast, F., Siegert, C., Dereviagin, A., Mai, D.H., 2001. Climatic implications of Late Quaternary plant macrofossil assemblages from the Taymyr Peninsula. Siberia. Glob. Planet. Change 31, 265–281. Kienast, F., Schirrmeister, L., Siegert, C., Tarasov, P., 2005. Palaeobotanical evidence for warm summers in the East Siberian Arctic during the last cold stage. Quat. Res. 63, 283–300. Kienel, U., Siegert, C., Hahne, J., 1999. Late Quaternary palaeoenvironmental reconstructions from a permafrost sequence (North Siberian Lowland, SE Taymyr Peninsula)—a multidisciplinary case study. Boreas 28, 181–193. Koch, P.L., 1991. The isotopic ecology of Pleistocene proboscideans. J. Vertebr. Paleontol. 11, 40A. Kozhevnikov, Y.P., Ukraintseva, V.V., 1997. Arguments for and against a Pleistocene tundra-steppe. Polar Geogr. 21, 51–69. Kuzmina, S., Sher, A., 2006. Some features of the Holocene insect faunas of northeastern Siberia. Quat. Sci. Rev. 25, 1790–1820. Lennox, A.R., Goodship, A.E., 2008. Polar bears (Ursus maritimus), the most evolutionary advanced hibernators, avoid significant bone loss during hibernation. Comp. Biochem. Physiol. A Comp. Physiol. 149, 203–208. Levin, N.E., Cerling, T.E., Passey, B.H., Harris, J.M., Ehleringer, J.R., 2006. A stable isotope aridity index for terrestrial environments. Proc. Natl. Acad. Sci. 103, 11,201–11,205. Longin, R., 1971. New method of collagen extraction for radiocarbon dating. Nature 230, 241–242. Lozhkin, A.V., Anderson, P.M., Eisner, W.R., Ravako, L.G., Hopkins, D.M., Brubaker, L.B., Colinvaux, P.A., Miller, M.C., 1993. Late Quaternary lacustrine pollen records from southwestern Beringia. Quat. Res. 39, 314–324. Matheus, P., Guthrie, R.D., Kunz, M.L., 2003. Predator–Prey Links in Pleistocene East Beringia: Evidence from Stable Isotopes. Third International Mammoth Conference, Program and Abstracts, Occasional Papers in Earth Sciences No. 5, Palaeontology Program, Government of the Yukon, Whitehorse, p. 80. McGee, M.E., Maki, A.J., Johnson, S.E., Nelson, O.L., Robbins, C.T., Donahue, S.W., 2008. Decreased bone turnover with balanced resorption and formation prevent cortical bone loss during disuse (hibernation) in grizzly bears (Ursus arctos horribilis). Bone 42, 396–404. Muhs, D.R., Ager, T.A., Been, J., Bradbury, J.P., Dean, W.E., 2003. A Late Quaternary record of eolian silt deposition in a maar lake, St. Michael Island, western Alaska. Quat. Res. 60, 110–122. Müller-Beck, H., 1982. Late Pleistocene man in northern Alaska and the mammothsteppe biome. In: Hopkins, D.M., MatthewsJr. Jr., J.V., Schweger, C.E., Young, S.B. (Eds.), Paleoecology of Beringia. Academic Press, New York, pp. 329–352. Nikolaev, V., Davanzo, S., Kuznetsova, L., Iacumin, P., 2004. Carbon isotope studies of mammoth's food and fossil remains. In: Nikolaev, V. (Ed.), Isotope and Palaeogeographical Studies in Northern Russia. Monography of the Russian Academy of Sciences, Moscow, pp. 21–40. Pate, F.D., Anson, T.J., 2008. Stable nitrogen isotope values in arid-land kangaroos correlated with mean annual rainfall: potential as a palaeoclimatic indicator. Int. J. Osteoarchaeol. 18, 317–326. Richards, M.P., Hedges, R.E.M., 2003. Variations in bone collagen δ13C and δ15N values of fauna from northwest Europe over the last 40 000 years. Palaeogeogr. Palaeoclimatol. Palaeoecol. 193, 261–267. Schirrmeister, L., Siegert, C., Kuznetsova, T., Kuzmina, S., Andreev, A., Kienast, F., Meyer, H., Bobrov, A., 2002. Paleoenvironmental and paleoclimatic records from permafrost deposits in the Arctic region of northern Siberia. Quat. Int. 89, 97–118. Schulze, E.D., Chapin, F.S., Gebauer, G., 1994. Nitrogen nutrition and isotope differences among life forms at the northern treeline of Alaska. Oecologia 100, 406–412. 96 P. Szpak et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 286 (2010) 88–96 Schwarcz, H.P., Dupras, T.L., Fairgrieve, S.I., 1999. 15N enrichment in the Sahara: in search of a global relationship. J. Archaeol. Sci. 26, 629–636. Sealy, J.C., van der Merwe, N.J., Lee-Thorp, J.A., Lanham, J.L., 1987. Nitrogen isotopic ecology in southern Africa: implications for environmental and dietary tracing. Geochim. Cosmochim. Acta 51, 2707–2717. Sharp, Z.D., Atudorei, V., Panarello, H.O., Fern·ndez, J., Douthitt, C., 2003. Hydrogen isotope systematics of hair: archeological and forensic applications. J. Archaeol. Sci. 30, 1709–1716. Stenhouse, M.J., Baxter, M.S., 1979. The uptake of bomb 14C in humans. In: Berger, R., Suess, H.E. (Eds.), Radiocarbon Dating. University of California Press, Berkeley, pp. 324–341. Stewart, G.R., Turnbull, M.H., Schmidt, S., Erskine, P.D., 1995. 13C natural abundance in plant communities along a rainfall gradient: a biological integrator of water availability. Aust. J. Plant Physiol. 22, 51–55. Swap, R.J., Aranibar, J.N., Dowty, P.R., Gilhooly III, W.P., Macko, S.A., 2004. Natural abundance of 13C and 15N in C3 and C4 vegetation of southern Africa: patterns and implications. Glob. Change Biol. 10, 350–358. Szpak, P., Orchard, T.J., Gröcke, D.R., 2009. A Late Holocene vertebrate food web from southern Haida Gwaii (Queen Charlotte Islands, British Columbia). J. Archaeol. Sci. 36, 2734–2741. Tieszen, L.L., 1991. Natural variations in the carbon isotope values of plants: implications for archaeology, ecology, and paleoecology. J. Archaeol. Sci. 18, 227–248. Ukraintseva, V.V., Agenbroad, L.D., Mead, J.I., 1996. A palaeoenvironmental reconstruction of the “Mammoth Epoch” of Siberia. In: West, F.H., West, C.F. (Eds.), American Beginnings: The Prehistory and Palaeoecology of Beringia. University of Chicago Press, Chicago, pp. 129–136. van der Merwe, N.J., Lee-Thorp, J.A., Thackeray, J.F., Hall-Martin, A., Kruger, F.J., Coetzee, H., Bell, R.H.V., Lindeque, M., 1990. Source-area determination of elephant ivory by isotopic analysis. Nature 346, 744–746. van der Merwe, N.J., Medina, E., 1991. The canopy effect, carbon isotope ratios and foodwebs in amazonia. J. Archaeol. Sci. 18, 249–259. Vogel, J.C., Eglington, B., Auret, J.M., 1990. Isotope fingerprints in elephant bone and ivory. Nature 346, 747–749. Wooller, M.J., Zazula, G.D., Edwards, M., Froese, D.G., Boone, R.D., Parker, C., Bennett, B., 2007. Stable carbon isotope compositions of eastern Beringian grasses and sedges: investigating their potential as paleoenvironmental indicators. Arct. Antarct. Alp. Res. 39, 318–331. Zazula, G.D., Froese, D.G., Schweger, C.E., Mathewes, R.W., Beaudoin, A.B., Telka, A.M., Harington, C.R., Westgate, J.A., 2003. Palaeobotany: ice-age steppe vegetation in east Beringia. Nature 423, 603-603. Zazula, G.D., Froese, D.G., Elias, S.A., Kuzmina, S., Mathewes, R.W., 2007. Arctic ground squirrels of the mammoth-steppe: paleoecology of Late Pleistocene middens (∼24 000–29 450 14C yr BP), Yukon Territory, Canada. Quat. Sci. Rev. 26, 979–1003.