Cervids in a dynamic northern landscape: Holocene changes in the relative abundance of moose and red deer at the limits of their distributions

Abstract

It is predicted that future climate change will have a significant impact on the distribution of large ungulates on a continental scale. At the same time, changes in human land use on a more local scale may affect their distribution and dispersal abilities, possibly confounding the effects of climate. We analyze changes in the Holocene distribution and relative abundance of Alces alces (moose) and Cervus elaphus (red deer) skeletal remains along an overlapping range boundary of these species in western Norway. As moose and red deer are adapted to different climatic conditions we would expect the distribution of finds to reflect large-scale changes in climate. In accordance with this prediction our results indicate that red deer became the predominant ungulate in this area during the mid-Holocene warm period, c. 8000–4000 cal. BP. Contrary to this, remains of moose became even less abundant in the subsequent colder period to the present. This decrease seems tied to the spread of agriculture and deforestation, indicating the importance of considering changes in land use when predicting future changes in ungulate distribution.

Keywords

agriculture AMS radiocarbon dating bone climate change Skipshelleren western Norway

Introduction

Climate is one of the main factors influencing the latitudinal and altitudinal distribution of species (Graham, 1986; Hofreiter and Stewart, 2009; Webb and Bartlein, 1992). Predicted changes in climate are likely to result in profound shifts in the ranges of species in the future (e.g. Root et al., 2003; Walther et al., 2002). At the edge of its range a species is by definition living in a marginal habitat and thus environmental variations are expected to have the most pronounced effects on populations at the range edges (Haldane, 1956; Williams et al., 2003). Studying ecological responses to environmental change at range edges has been identified as a key for understanding the future of populations (Sexton et al., 2009).

Southern Norway is currently a border zone between the distribution of moose (Alces alces) and red deer (Cervus elaphus). The historical distribution of the red deer in Norway has been along the milder west coast while moose have been mainly associated with the more continental eastern inland, with the species being separated by the Scandes mountain range (Collett, 1912). Although there is some range overlap, moose and red deer can be considered as allopatric species on a global scale (Ahlén, 1965; Mitchell-Jones et al., 1999), indicating that they are adapted to different climatic conditions. Moose have a body construction well suited for movement in deep snow (Formozov, 1946; Telfer and Kelsall, 1979) and are able to withstand periods of extreme cold, but are easily stressed by high temperatures (Lenarz et al., 2009; Markgren, 1966; Renecker and Hudson, 1986; van Beest et al., 2012). Red deer, on the other hand, although highly adaptable, are not as tolerant of cold (Semiadi et al., 1996; Simpson et al., 1978) or snowy conditions (Ahlén, 1965; Formozov, 1946). They are also more dependent on grazing on ground vegetation, and have problems feeding if the ground is frozen or covered by deep snow (Ahlén, 1965; Mysterud et al., 2001). Thus, changes in the relative abundance of these two species could be expected to directly or indirectly reflect climatic conditions.

As a consequence of changes in human land use and wildlife management practices many wild European ungulates, including moose and red deer, have been increasing in numbers and distribution following dramatic population declines during recent centuries (Linnell and Zachos, 2011). Through high hunting pressure, game management, translocation of animals and habitat alterations human activities have grown to become a dominant factor in the environment of wild ungulates (Linnell and Zachos, 2011; Mysterud, 2010), and may have become more important than climate in structuring their distribution and abundance. The Holocene record of moose (Schmölcke and Zachos, 2005) and red deer (Sommer et al., 2008) in central Europe show that their prehistoric distributions were more sympatric than today, highlighting both climate and humans as potential factors in structuring their ranges.

Here we expand upon this by investigating changes in the distribution of moose and red deer along a range boundary for these species in a long-term perspective. This is done by reviewing the distribution of subfossil bone finds from the Holocene (the last 11,500 years) in western Norway, and a more in-depth study of changes in the relative abundance of moose to red deer at the archaeological rock shelter site Skipshelleren for which the bone record is particularly rich. During the Holocene there have been considerable climatic fluctuations (Nesje, 2009), and our hypothesis is that these would have affected the relative distributions of moose and red deer. If so, we predict that moose bones would be relatively less abundant in archaeological sites dated to warmer periods of the Holocene and more abundant during the cooler periods. Human land use may, however, have influenced these changes, and the results are discussed in light of current knowledge on the development of late-Holocene agricultural activity and deforestation.

Study area

The Scandes mountain range makes a natural barrier of alpine tundra between eastern and western Norway, in many parts reaching out to the coast creating long and narrow fjords with steep hillsides (Figure 1). Seaside areas and fjords are mild and wet for most of the year, with temperatures rarely falling below 0°C, and low elevations usually receive very little snow. Higher areas have more pronounced climatic fluctuations with longer and more snow-rich winters. Presently, overall precipitation decreases eastwards, varying between more than 3500 to less than 400 mm/yr (Moen, 1999), but snow cover increases and may reach considerable depth (> 2 m) in some areas during winter (October–April). Average annual temperatures range from about 8°C on the western coast to −6°C in mountainous areas (Moen, 1999). Broadleaf forests cover parts of the coast, but the outer coast is characterized by treeless coastal heathlands (Moen, 1999). Further inland mixed coniferous and deciduous forests change to more boreal vegetation.

Figure 1.

The Holocene distribution of moose (triangles) and red deer (circles) skeletal remains in western Norway and adjacent mountain areas by time periods. Sites with remains of both species are marked by stars. Site numbers correspond with Table 1. Location of the study area is indicated in panel (d) along with topographic information.

The rock shelter Skipshelleren (site 19 in Figure 1) is a prehistoric human dwelling site containing one of the largest and best stratified subfossil faunal collections in Norway. The site is situated at the ecotone between boreal (coniferous) forest and temperate (broadleaved) woodland. Skipshelleren is situated in a fjord/river valley leading about 35 km inland to Voss, which presently has a small but stable moose population (with an average of approximately five moose shot per year for the last 10 years compared with approximately 300 shot red deer). Only a few straying moose are at times observed near the site, but the area supports a large population of red deer (Overvoll and Wiers, 2004).

Material and methods

Distribution analysis

Files from the collection archive of the Osteology collections at the University Museum of Bergen, published papers and publicly available archaeological reports and theses that report subfossil remains of moose and red deer in western Norway and adjacent mountain regions were reviewed. Those sites found to be securely dated, either by ¹⁴C or typology, were used in the analysis (Figure 1, Table 1). Bone or antler tools were excluded as these may be more prone to long distance transport. Additional ¹⁴C dating by accelerator mass spectrometry was performed on two sites,Skipshelleren and Vistehulen, to clarify the stratigraphy of the sites, as well as material from the previously unpublished site Tjuvanotten (Table 2). Based on ¹⁴C results and stratigraphic information the sites were separated into time periods given in calibrated calendar years before present (cal. BP): early Holocene (> 8000 cal. BP), mid Holocene (c. 8000–4000 cal. BP) and late Holocene (c. 4000–1000 cal. BP). The three time periods were tentatively chosen on the basis of three defined main climatic periods during the Holocene (e.g. Bjune et al., 2005). The ¹⁴C dates were calibrated using CALIB 6.1.1 (Stuiver and Reimer, 1993), based on the data set IntCal09 (Reimer et al., 2009), with 2σ ranges.

Table 1.

List of sites used in the study with references to their dating. Site numbers correspond to sites listed in Figure 1.

	Site	County	Age^a	Type of date^b	References to dates
1	Austbø	Rogaland	B	DS	Juhl (2001)
2	Geitalemen	Hordaland	B–C	DS, TY	Bommen (2009); Hougen (1922)
3	Grimstadneset	Hordaland	C	DB	Rosvold et al. (2012)
4	Grønehelleren	Sogn og Fjordane	B–C	DS	Jansen (1998); Rosvold et al. (2012)
5	Halnefjorden	Buskerud	B	DB	Indrelid (1994)
6	Hestneshulen	Sør-Trøndelag	C	DB, DS	Lie (1991); Petersen (1910); Rosvold et al. (2010)
7	Hjartøy	Hordaland	C	DS	Johannesen (1998)
8	Kobbehelleren	Hordaland	C	DS	Bommen (2009)
9	Kotedalen	Hordaland	A–B	DS	Hufthammer (1992a)
10	Kuhidlaren	Hordaland	C	DS, TY	Bakka (1972); Bommen (2009)
11	Lyngaland	Rogaland	C	TY	Petersen (1936)
12	Olsteinhelleren	Hordaland	B	DS	Bergsvik and Hufthammer (2009)
13	Osterbakken	Hordaland	C	DS	Bommen (2009)
14	Pålsbufjorden	Buskerud	A	DB	Groseth (2004)
15	Rundøyno	Hordaland	C	DB	Rosvold et al. (2012)
16	Ruskeneset	Hordaland	C	TY	Brinkmann and Shetelig (1920)
17	Sauehelleren	Møre og Romsdal	C	DS	Nummedal (1913); Rosvold et al. (2010)
18	Setrehelleren	Hordaland	C	DS, TY	Bommen (2009); Olsen and Shetelig (1934)
19	Skipshelleren	Hordaland	B–C	DB, DS	Hjelle et al. (2006); Olsen (1976); Table 2
20	Skjonghelleren	Møre og Romsdal	C	TY	Brøgger (1910)
21	Skrivarhelleren	Sogn og Fjordane	C	DS	Prescott (1991)
22	Slettabø	Rogaland	B–C	DS	Skjølsvold (1977)
23	Smiehelleren	Møre og Romsdal	C	DS	Haug (2012)
24	Stangelandshidleren	Rogaland	B	DB, TY	Brøgger (1911); Høgestøl and Prøsch-Danielsen (2006)
25	Sævarhelleren	Hordaland	A	DS	Bergsvik and Hufthammer (2009)
26	Tjuvanotten	Hordaland	C	DB	Table 2
27	Vistehulen	Rogaland	A–B	DB	Indrelid (1978); Table 2

Notes: ^aA (>8000 cal. BP), B (8000–4000 cal. BP), C (4000–1000 cal. BP).

DB (¹⁴C dated bone), DS (¹⁴C dated site or layer), TY (typologically dated site).

Table 2.

Overview of new and published radiocarbon dates from specific layers at Skipshelleren and the new dates from Vistehulen and Tjuvanotten. Calibrated calendar ages are given with 2⊠ ranges.

Site	Context	Age in ¹⁴C yr BP	Calibrated age (cal. BP)	Analyzed species	Lab code/reference
Skipshelleren	Layer 1	1875 ± 30	1803 ± 77	Sus scrofa	Poz-25447
Skipshelleren	Layer 1	2070 ± 35	2040 ± 93	Cervus elaphus	TUa-5861
Skipshelleren	Layer 1	2125 ± 40	2148 ± 155	Alces alces	TUa-5862
Skipshelleren	Layer 1	2315 ± 30	2272 ± 89	Alces alces	TUa-5860
Skipshelleren	Layer 1	2790 ± 30	2877 ± 83	Sus scrofa	Poz-25448
Skipshelleren	Layer 2	2105 ± 40	2125 ± 174	Cervus elaphus	TUa-5856
Skipshelleren	Layer 2	2485 ± 35	2544 ± 178	Cervus elaphus	TUa-5857
Skipshelleren	Layer 3	2500 ± 40	2553 ± 187	Cervus elaphus	TUa-5854
Skipshelleren	Layer 3	3460 ± 90	3723 ± 243	Unspecified	1
Skipshelleren	Layer 4	3635 ± 65	3939 ± 210	Bos taurus	2
Skipshelleren	Layer 4	4000 ± 90	4492 ± 327	Unspecified	1
Skipshelleren	Layer 4	4255 ± 40	4792 ± 163	Cervus elaphus	TUa-6755
Skipshelleren	Layer 4	4400 ± 80	5066 ± 221	Unspecified	1
Skipshelleren	Layer 5	4655 ± 45	5439 ± 137	Alces alces	TUa-5855
Skipshelleren	Layer 5	5160 ± 90	5924 ± 259	Unspecified	1
Skipshelleren	Layer 6	6000 ± 100	6898 ± 260	Unspecified	1
Skipshelleren	Layer 6	6120 ± 90	7016 ± 230	Unspecified	1
Skipshelleren	Layer 7	6275 ± 50	7166 ± 145	Cervus elaphus	TUa-5858
Skipshelleren	Layer 7	6490 ± 50	7390 ± 101	Cervus elaphus	TUa-5859
Vistehulen	III F:9	7820 ± 50	8609 ± 155	Alces alces	TUa-5944
Vistehulen	IV H:2	7315 ± 60	8153 ± 153	Alces alces	TUa-5945
Vistehulen	X K:3	6405 ± 50	7342 ± 84	Alces alces	TUa-5946
Vistehulen	XIII K:3	7250 ± 45	8074 ± 94	Alces alces	TUa-5947
Vistehulen	IX K:3	6630 ± 45	7507 ± 69	Alces alces	TUa-5948
Vistehulen	IX K:6	6710 ± 70	7569 ± 109	Cervus elaphus	TUa-5949
Vistehulen	XII I:4	7830 ± 60	8711 ± 261	Alces alces	TUa-5950
Vistehulen	XII K:3	8050 ± 60	8893 ± 233	Alces alces	TUa-5951
Vistehulen	XIV K:2	6175 ± 50	7093 ± 149	Cervus elaphus	TUa-5952
Tjuvanotten	Surface	1515 ± 40	1423 ± 96	Alces alces	TUa-7758
Tjuvanotten	Surface	1535 ± 40	1435 ± 89	Alces alces	TUa-7759

Notes: 1: Olsen (1976); 2: Hjelle et al. (2006).

Relative abundance in Skipshelleren

Skipshelleren was excavated in 1930–1931 and separated into seven distinctive stratigraphic layers (Bøe, 1934). Seven ¹⁴C dates were already available prior to this study (Hjelle et al., 2006; Olsen, 1976), and our new ¹⁴C dates (see above) ensured at least two dates from each layer. These dates show that layer 1 and 2 cannot be differentiated in time and we consider them as a single unit for this analysis.

The skeletal material from Skipshelleren are highly fragmented, but well preserved, and 23 species of mammals as well as numerous fish and bird species have been identified (Olsen, 1976). The bone material was later revised and the analysis is based on this revision. The site contains 5962 skeletal fragments (including bones, teeth and antler) of moose and red deer that can be traced to specific layers based on excavation reports. The total number of identified specimens (NISP) and the relative abundance of moose to red deer NISP in each layer were calculated (Table 3). The change in relative abundance through time was tested for significant difference using chi-square analysis for linear trend (χ_t²), as advocated by Cannon (2001). We also performed chi-square tests (χ²) of the relative abundance between adjacent layers.

Table 3.

Numbers of identified specimens (NISP) of moose and red deer by stratigraphic layer at Skipshelleren, with approximate layer ages.

Layer	Moose NISP	Red deer NISP	Age(cal. BP)
1&2	39	2443	2500–1800
3	2	302	3900–2500
4	40	972	5300–3900
5	34	568	6200–5300
6	63	1452	7100–6500
7	3	44	7500–7100
Sum	181	5781

The general morphological similarity between moose and red deer leads us to expect no significant differences in fragmentation or preservation rates. Different butchering and transportation practices may, however, alter the composition of body parts found at the site (e.g. Grayson, 1989; Lyman, 1985), which can affect a NISP-based analysis (Grayson and Delpech, 1998). Hunters may, for example, have discarded more of the larger moose bones at the kill site rather than bringing them home, thereby making moose under-represented at the site, e.g. the Schlepp effect (Perkins and Daly, 1968). This was investigated by grouping specimens into eight different body part groups; phalanges, metapodials, carpals/tarsals, radius/ulna, tibia, proximal limb bones (humerus, femur and scapula), trunk (ribs, sternum and vertebra) and head (antlers, skull and teeth); and performing a chi-square test for significant differences in the body part representation of the species. We also performed a temporal analysis of the relative occurrence of different body parts by pooling the red deer material into two broad time periods, late Holocene (layers 1–3) and mid Holocene (layers 4–7). These pooled layers also include skeletal material from more unspecific layers that can only be traced to a wider time period, thereby increasing the sample size to 7984 fragments.

Results

Spatio-temporal pattern of moose and red deer finds

While red deer is clearly the predominant ungulate species when considering subfossil skeletal remains in western Norway, there seem to have been changes in the relative abundance of moose and red deer through time (Figure 1).

Only three sites have so far been found west of the Scandes mountain range with identified remains of moose and red deer older than 8000 cal. BP (Figure 1a): Sævarhelleren (site 25), Vistehulen (site 27) and Kotedalen (site 9). Sævarhelleren, dated c. 9000–7800 cal. BP, is a newly excavated site and is not yet fully analyzed. However, preliminary results show that moose remains are more abundant than red deer at this site (Bergsvik and Hufthammer, 2009). Moose remains are also much more abundant than red deer in Vistehulen, dated c. 9000–5000 cal. BP, with NISP = 85 and 17, respectively (Degerbøl, 1951). The new ¹⁴C dates (Table 2) show some uncertainties in the stratigraphy at Vistehulen, possibly indicating some degree of mixing of layers, and thus we do not know with certainty how the relative abundance changes with time. Moose bones were, however, mostly found in the deeper layers and those of red deer in the upper, and the presumably oldest bone of red deer from this site has been dated to 7569 ± 109 cal. BP. The skeletal material from Kotedalen, dated c. 8500–4700 cal. BP, is heavily fragmented and mostly burned (Hufthammer, 1992a). Red deer are more abundant than moose at this site, but only 11 fragments have been identified to either of the species from the early-Holocene layers.

Several sites are available from the mid Holocene (Figure 1b), showing a predominance of red deer remains both among and within sites. Olsteinhelleren (site 12), dated c. 7600–6800 cal. BP, lies just a few metres from the older site Sævarhelleren (see above). However, the skeletal material indicates that red deer had now become more frequent than moose in the area (Bergsvik and Hufthammer, 2009). Red deer remains become even more predominant among late-Holocene sites (Figure 1c), where Skipshelleren (site 19) and Tjuvanotten (site 26) are the only sites with reliable evidence of the presence of moose in southwestern Norway in this period. Tjuvanotten has not yet been excavated but bones of both moose and red deer have been found on the cave floor, dating to 1500–1400 cal. BP (Table 2). The site is situated close to the Scandes Mountains and lies in a valley leading northwest to Skipshelleren, indicating that the source populations for these animals might have been the same. Bone finds from the early and mid Holocene in northwestern Norway are generally very scarce, making the distribution in this area uncertain before the late Holocene. These sites show a continuous distribution of red deer along the coast and the presence of moose closer to central Norway, which presently represents one its core areas.

Changes in the relative abundance of moose to red deer in Skipshelleren

The general picture from the Skipshelleren bone assemblage, dated to c. 7500–1800 cal. BP (Figure 2), corresponds well with the spatiotemporal distribution of finds (Figure 1). Moose remains are present in all layers but in small numbers compared with red deer, with a maximum of about 6.5% of cervid bones in the oldest layer. There is a significant decline in the relative abundance of moose to red deer over time (χ_t² = 36.21, P<0.001). Only two adjacent layers, 3 and 4 (χ²=7.18, P<0.01), and the combined layers 1–3 and 4–7 (χ²=57.02, P<0.01) differ significantly from each other, showing that moose bones became rarer at the site in the late Holocene.

Figure 2.

The changing relative abundance of moose to red deer skeletal material in Skipshelleren. The graph shows precentage of moose specimens, according to stratigraphic layer and climate regime. Approximate age of layer boundaries are shown in calendar years before present (cal. BP).

Both species are represented by bones from all parts of the body and in similar frequencies for most parts (Figure 3a). Moose remains are, however, more dominated by phalanges than red deer, causing a significant difference in the body part representation between the species (χ²=126.08, P<0.001). This could indicate an under-representation of moose in the material, suggesting that more bones of moose were discarded before reaching the site. There is also a significant change in body part representation of red deer between the combined layers 1–3 and 4–7 (χ²=142.50, P<0.001), caused by a change in the number of phalanges, but the overall distribution is very similar (Figure 3b). The feet of ungulates are often brought back to the dwelling as they contain strong tendons that are valuable raw material and phalanges often remain attached to the skin of butchered animals (Klein, 1989; Perkins and Daly, 1968). Alternatively the site could have been used as a butchering site, where bones from the meatier parts of the body have been transported away. Repeating the analysis using only phalanges produces the same pattern of change between layers as in Figure 2. However, the relative abundance of moose in the mid-Holocene layers increase to an average of 8%, creating a larger difference between the mid and late Holocene. Thus the drop in relative abundance of moose in the late Holocene does not seem to be caused by different body part representation.

Figure 3.

Body part representation at Skipshelleren based on the full data set. (a) Moose (triangles) and red deer (circles) skeletal material. (b) Red deer skeletal material at different time periods: mid Holocene (filled circles) and late Holocene (open circles).

Discussion

Climatic reconstructions from western Norway show that the early Holocene was relatively cool and dry followed by a warm and wet mid Holocene, about 8000–4000 cal. BP, with summer temperatures up to 2°C higher and winter precipitation up to 225% of today (Bjune et al., 2005). Winter temperatures were also higher and the climate seems to have become more oceanic in the mid Holocene (Giesecke et al., 2008). By the end of the mid Holocene the climate of western Norway started getting colder again and the period from about 4000 cal. BP is characterized as cooler and drier than the previous period (Bjune et al., 2005).

The oldest postglacial remains of moose on the Scandinavian Peninsula have been dated to about 13,200 cal. BP (Aaris-Sørensen, 2009) while red deer were present from at least 10,600 cal. BP (Liljegren and Ekström, 1996), both from Scania in southern Sweden. Not surprisingly the first traces of these species in Norway are in southeastern Norway, not far from the Swedish border (Hufthammer, 2006). The oldest directly dated remains of moose and red deer are bog finds of shed antlers. These confirm the presence of moose in Norway from around 10,300 cal. BP (Grøndahl et al., 2010) whilst a red deer antler, from Viul in Ringerike, has been dated to c. 8500 cal. BP (Ø Wiig, UiO Natural History Museum, personal communication, 2009).

Our finds from western Norway (see results and Figure 1) indicate that moose were more abundant than red deer in the early Holocene and a more rapid postglacial colonizer. However, the higher abundance of moose bones does not necessarily indicate that moose was a common animal in western Norway, as all of these sites are highly dominated by wild boar (Sus scrofa) (Rosvold et al., 2010). The high abundance of a temperate species such as the wild boar indicates that conditions should have been good for red deer as well. The reason for this apparently slower colonization by red deer is uncertain and the material is too scarce to provide a good explanation. It could be argued that this might in part be an effect of different dispersal behaviour, as female moose are more prone to natal dispersal than female red deer (Liberg and Wahlström, 1995; Loe et al., 2009). However, the capacity for rapid dispersal of red deer has been shown by its expansion to large parts of the country during the last century (Langvatn, 1998).

The absence of moose bones among the many sites containing red deer indicates a low density of moose in western Norway from the mid Holocene onwards. Moose were still found in western Norway but, as illustrated by layers 4–7 in Skipshelleren(Figure 2), in low numbers relative to red deer. The takeover of red deer as the dominating ungulate along the west coast in the warmer mid Holocene is therefore in agreement with the hypothesis based on climatic adaptations. This warm climate also led to the broadleaved forests reaching their maximum density in western Norway (Moe et al., 1996), indicating that this type of vegetation may have been important for the establishment of red deer. One would, however, expect moose to have become relatively more abundant in western Norway during the colder late Holocene, but this is apparently not the case (Figure 1c, Figure 2).

Farming and domestic animal husbandry were introduced to Norway about 6000 cal. BP, but was, at least in western Norway, not firmly established before the Late Neolithic c. 4500–4000 cal. BP (Hjelle et al., 2006; Høgestøl and Prøsch-Danielsen, 2006). This coincides with the reduction in the relative abundance of moose to red deer at Skipshelleren (Figure 2) and the otherwise lack of moose bones in western Norway (Figure 1c). As these bones are the remains of human hunting activities, one cannot dismiss the idea that this decrease is a result of changes in hunting habits caused by a more sedentary way of life. However, hunting has continued to be an important part of the Norwegian economy up to modern times and the high species richness in the deposits at Skipshelleren shows a continued hunting economy in addition to animal husbandry (Olsen, 1976).

Pollen diagrams show deforestation and opening of the landscape following agriculture and the colder climate of the late Holocene (Bjune, 2005; Hjelle et al., 2006, 2010; Høgestøl and Prøsch-Danielsen, 2006; Kaland, 1986). This started earlier and was most pronounced at the outer coast, with the development of the coastal heathlands, but was also obvious further inland. Western Norway remained largely forested but mostly on the steeper hillsides less favorable to agriculture (Hjelle et al., 2006) and the most productive soils were the first to be cultivated (Hjelle et al., 2010; Overland and Hjelle, 2009). Red deer easily make use of steep hillsides and readily adapt to more open terrain (e.g. Clutton-Brock et al., 1982), while the more heavily built moose may have been less able to make use of the remaining forest cover. Early animal husbandry could also have been positive for red deer as areas were improved for grazing and a certain degree of cattle grazing has been found to facilitate feeding opportunities for red deer (Gordon, 1988; Kuiters et al., 2005). An analysis of ancient DNA from western Norwegian red deer support this, indicating a large population size around 2000 cal. BP with a high genetic diversity (Rosvold et al., 2012). Continuous livestock grazing does, however, prevent the growth of browse (Austrheim et al., 2011; Speed et al., 2010), and has been found to limit food availability for moose (Wolfe, 1974).

During the early and mid Holocene the treeline altitudes were higher than today and forest covered much of the present-day alpine tundra regions such as Hardangervidda (Aas and Faarlund, 1988; Bjune, 2005; Moe, 1979). This would have linked forested habitats on either side of the Scandes Mountains, providing an east–west migration route, and there are indeed traces of moose at Hardangervidda (i.e. sites 5 and 14 in Figure 1). Owing to the isostatic uplift, colder climate and increased human impact through domestic livestock grazing, the treeline started decreasing around 4000–5000 years ago, leaving previously forested mountain areas open (Bjune, 2005; Eide et al., 2006; Gunnarsdóttir,1996). These areas were, until the beginning of the ad 20th century, heavily used by humans, both as pastures for livestock and as hunting grounds for wild reindeer (Austrheim et al., 2011; Blehr, 1973; Indrelid, 1994; Olsson et al., 2000). Narrower dispersal corridors restrict dispersal rates (Travis and Dytham, 1999), and immigration of moose from the east could thus have been limited. Isolated from the eastern core area, the dwindling western populations would have experienced increased demographic stochasticity and greater risks of local extinction (Lande et al., 2003; Stacey and Taper, 1992).

The transformation from a forested landscape to a more open cultural landscape also coincided with the loss of wild boar in Norway (Rosvold and Andersen, 2008). Red deer seem to have tackled these changes better. Forest clearance accelerated over the last two millennia as the human population expanded and society advanced (Myhre, 2004; Rolstad et al., 2001). At this time genetic analyses indicate that red deer also suffered a decrease in numbers (Rosvold et al., 2012), leading to restricted and fragmented populations, a trend persisting until the end of the 19th century (Collett, 1909). As a consequence of modern forestry practices and falling numbers of livestock, woodlands are expanding and the treelines are rising again (Aas and Faarlund, 1995; Hofgaard, 1997; Speed et al., 2010). Strict management of hunting practices have allowed the populations of both moose and red deer to increase considerably during the last decades (Austrheim et al., 2011). Following this, red deer have spread into eastern Norway and moose are again observed in the inner parts of the western Norwegian fjords. It is also worth noting that roe deer (Capreolus capreolus), which seem to have had a very limited occurrence in Norway during prehistory (Hufthammer, 1992b; Hufthammer and Aaris-Sørensen, 1998), have spread to most of Scandinavia during this last century (Andersen et al., 2004).

Historic records and subfossil finds from other parts of Europe show that the distribution of both moose and red deer has changed a lot during the Holocene. While red deer have relatively rapidly expanded their range across most of Europe after the last glacial maximum (Sommer et al., 2008), the distribution of moose has gradually retreated northwards (Schmölcke and Zachos, 2005). Interspecific competition between moose and red deer is not considered to be particularly important (Ahlén, 1965; Mysterud, 2000), however, few studies have investigated these species living in the same area. We would not exclude competition as a contributing factor, but climatic change and habitat alterations because of expanding agriculture seem to be more likely candidates. Much of the present distribution of wild ungulates in Europe was established after the Middle Ages when high hunting pressure and habitat alterations extirpated many populations (Linnell and Zachos, 2011; Schmölcke and Zachos, 2005). It is therefore likely that much of what we consider climatic boundaries for the distribution of many species are rather an effect of several different historic processes.

Conclusions

The European landscape has a long history of human modifications and exploitation. Even so, the European fauna has retained a distinct biogeographical distribution of species indicating that climate is still one of the main controlling factors (Heikinheimo et al., 2007; Rueda et al., 2010). Direct or indirect effects of future climatic warming are expected to alter the distribution of wild ungulates, increasing the range of temperate species and limiting the distribution of northern species (Mysterud and Sæther, 2011). Our results from western Norway show that the transformation from a forested landscape to a more open cultural landscape produced changes in the ungulate fauna contrary to expectations based solely on climate change. This highlights the importance of also considering changes in land use and forest structure when predicting future responses of fauna to climate change.

Footnotes

Acknowledgements

We are grateful to Astrid B Lorentzen for aid in producing the maps, Øystein Wiig and Lars Groseth for information regarding two of the sites, and to Arne B Johansen for valuable comments and discussion.

Funding

Funds for the project were provided by the Directorate for Nature Management.

References

Aaris-Sørensen

(2009) Diversity and dynamics of the mammalian fauna in Denmark throughout the last glacial–interglacial cycle, 115–0 kyr BP. Fossils and Strata 57: 1–59.

Aas

Faarlund

(1988) Postglasiale skoggrenser i sentrale sørnorske fjelltrakter. 14C-datering av subfossile furu- og bjørkerester. (Postglacial forest limits in central south Norwegian mountains. Radiocarbon datings of subfossil pine and birch specimens). Norsk Geografisk Tidsskrift 42: 25–61.

Aas

Faarlund

(1995) Skoggrenseutviklingen i Norge, særlig i det 20. århundre. AmS-Varia 24: 89–100.

Ahlén

(1965) Studies on the red deer, Cervus elaphus L., in Scandinavia. III. Ecological investigations. Viltrevy 3: 177–376.

Andersen

Herfindal

Sæther

(2004) When range expansion rate is faster in marginal habitats. Oikos 107: 210–214.

Austrheim

Solberg

Mysterud

(2011) Spatio-temporal variation in large herbivore pressure in Norway during 1949–1999: Has decreased grazing by livestock been countered by increased browsing by cervids? Wildlife Biology 17: 286–298.

Bakka

(1972) Forntida i Stord. In: Høyland

(ed.) Stord bygdebok I. Stord i gamal og ny tid. Stord: Stord Herad, pp. 70–144.

Bergsvik

Hufthammer

(2009) Stability and change among marine hunter-fishers in western Norway 7000–4500 cal BC. Results fromthe excavations of two rockshelters in Hardanger. In: Crombé

Van Strydonck

Sergant

. (eds) Chronology and Evolution Within the Mesolithic of North-west Europe. Cambridge: Cambridge Scholars Publishing, pp. 435–449.

Bjune

(2005) Holocene vegetation history and tree-line changes on a north–south transect crossing major climate gradients in southern Norway – Evidence from pollen and plant macrofossils in lake sediments. Review of Palaeobotany and Palynology 133: 249–275.

10.

Bjune

Bakke

Nesje

. (2005) Holocene mean July temperature and winter precipitation in western Norway inferred from palynological and glaciological lake-sediment proxies. The Holocene 15: 177–189.

11.

Blehr

(1973) Traditional reindeer hunting and social change in the local communities surrounding Hardangervidda. Norwegian Archaeological Review 6: 104–112.

12.

Bommen

(2009) Bruken av hellerar i eldre jernalder i Sunnhordland. Institutt for arkeologi, historie, kultur- og religionsvitenskap, Masters thesis.

13.

Brinkmann

Shetelig

(1920) Ruskenesset: En stenalders jagtplass. Norske Oldfund 3.

14.

Brøgger

(1910) Vestnorske hulefund fra ældre jernalder. Bergens Museums Aarbok 16: 1–22.

15.

Brøgger

(1911) Jæderske stenaldersbosteder. Naturen 35: 313–334.

16.

Bøe

(1934) Boplassen i Skipshelleren på Straume i Nordhordaland. Bergen: A.S. John Griegs Boktrykkeri.

17.

Cannon

(2001) Archaeofaunal relative abundance, sample size, and statistical methods. Journal of Archaeological Science 28: 185–195.

18.

Clutton-Brock

Guinness

Albon

(1982) Red Deer. Behavior and Ecology of Two Sexes. Chicago: The University of Chicago Press.

19.

Collett

(1909) Hjorten i Norge (Cervus elaphus, atlanticus), nogle biologiske meddelelser. Bergens Museums Aarbog 6: 1–31.

20.

Collett

(1912) Norges pattedyr. Kristiania: Aschehoug.

21.

Degerbøl

(1951) Det osteologiske materiale. Knoglemateriale fra en ny udgravning i Vistehulen i Norge. In: Lund

(ed.) Fangst-boplassen i Vistehulen: på Viste, Randaberg, Nord-Jæren. Stavanger: Stavanger Museum, pp. 52–84.

22.

Eide

Birks

Bigelow

. (2006) Holocene forest development along the Setesdal valley, southern Norway, reconstructed from macrofossil and pollen evidence. Vegetation History and Archaeobotany 15: 65–85.

23.

Formozov

(1946) Snow Cover as an Integral Factor of the Environment and its Importance in the Ecology of Mammals and Birds. Edmonton: Occasional Publication 1, Boreal Institute, University of Alberta.

24.

Giesecke

Bjune

Chiverrell

. (2008) Exploring Holocene continentality changes in Fennoscandia using present and past tree distributions. Quaternary Science Reviews 27: 1296–1308.

25.

Gordon

(1988) Facilitation of red deer grazing by cattle and its impact on red deer performance. Journal of Applied Ecology 25: 1–10.

26.

Graham

(1986) Responses of mammalian communities to environmental changes during the Late Quaternary. In: Diamond

Case

(eds) Community Ecology. New York: Harper & Row, pp. 300–313.

27.

Grayson

(1989) Bone transport, bone destruction, and reverse utility curves. Journal of Archaeological Science 16: 643–652.

28.

Grayson

Delpech

(1998) Changing diet breadth in the Early Upper Palaeolithic of Southwestern France. Journal of Archaeological Science 25: 1119–1129.

29.

Grøndahl

Hufthammer

Dahl

. (2010) A Preboreal elk (Alces alces L., 1758) antler from south-eastern Norway. Fauna Norvegica 30: 9–12.

30.

Groseth

(2004) Regulering av Numedalslågen. Vedlegg til rapport fra arkeologiske registreringer i Pålsbufjorden september/oktober 2002.

31.

Gunnarsdóttir

(1996) Holocene vegetation history and forest-limit fluctuations in Smådalen, eastern Jotunheimen, South Norway. Paläoklimaforschung 20: 233–255.

32.

Haldane

JBS

(1956) The relation between density regulation and natural selection. Proceedings of the Royal Society of London B 145: 306–308.

33.

Haug

(2012) Rockshelters in central Norway: Long-term changes in use, social organization and production. In: Bergsvik

Skeates

(eds) Caves in Context: The Cultural Significance of Caves and Rockshelters in Europe. Oxford: Oxbow Books, pp. 39–47.

34.

Heikinheimo

Fortelius

Eronen

. (2007) Biogeography of European land mammals shows environmentally distinct and spatially coherent clusters. Journal of Biogeography 34: 1053–1064.

35.

Hjelle

Halvorsen

Overland

(2010) Heathland development and relationship between humans and environment along the coast of western Norway through time. Quaternary International 220: 133–146.

36.

Hjelle

Hufthammer

Bergsvik

(2006) Hesitant hunters: A review of the introduction of agriculture in western Norway. Environmental Archaeology 11: 147–170.

37.

Hofgaard

(1997) Inter-relationships between treeline position, species diversity, land use and climate change in the central Scandes Mountains of Norway. Global Ecology and Biogeography Letters 6: 419–429.

38.

Hofreiter

Stewart

(2009) Ecological change, range fluctuations and population dynamics during the Pleistocene. Current Biology 19: R584–R594.

39.

Høgestøl

Prøsch-Danielsen

(2006) Impulses from agro-pastoralism in the 4th and 3rd millennia BC on the south-western coastal rim of Norway. Environmental Archaeology 11: 19–34.

40.

Hougen

(1922) Gjeitalemen. En stenalders boplass i Førde, Søndhordaland. Bergens Museums Aarbok 1920–1921: 1–15.

41.

Hufthammer

(1992a) De osteologiske undersøkelsene fra Kotedalen. In: Fasteland

(ed.) Kotedalen – En boplass gjennom 5000 år. Bind 2. Naturvitenskapelige undersøkelser. Bergen: Historisk Museum, Universitetet i Bergen, pp. 9–64.

42.

Hufthammer

(1992b) Prehistorical and early historical distribution of Roe deer Capreolus capreolus (L.) in Norway. Fauna Norvegica Series A 13: 35–37.

43.

Hufthammer

(2006) The vertebrate fauna of eastern Norway – From the Ice Age to the Middle Ages. Skrifter (Kulturhistorisk Museum) 4: 191–202.

44.

Hufthammer

Aaris-Sørensen

(1998) Late- and postglacial European roe deer. In: Andersen

Duncan

Linnell

JDC

(eds) The European Roe Deer: The Biology of Success. Oslo: Scandinavian University Press, pp. 47–69.

45.

Indrelid

(1978) Mesolithic economy and settlement patterns in Norway. In: Mellars

(ed.) The Early Postglacial Settlement of Northern Europe: An Ecological Perspective. London: Duckworth, pp. 147–176.

46.

Indrelid

(1994) Fangstfolk og bønder i fjellet. Bidrag til Hardangerviddas førhistorie 8500– 2500 år før nåtid. Oslo: Universitetets Oldsaksamlings Skrifter, Ny rekke nr. 17.

47.

Jansen

(1998) Grønehelleren, en boplass i Solund. Årbok for Sogn 44: 73–86.

48.

Johannesen

(1998) Fiskevær og fiskebuer i vestnorsk jernalder. En analyse av strandtufter i Hordaland. Arkeologiske avhandlinger og rapporter fra Universitetet i Bergen 2.

49.

Juhl

(2001) Austbø på Hundvåg gennem 10000 år. AmS-Varia 38: 1–101.

50.

Kaland

(1986) The origin and management of Norwegian coastal heaths as reflected by pollen analysis. In: Fasteland

(ed.) Anthropogenic Indicators in Pollen Diagrams. Rotterdam: A.A. Balkema, pp. 19–36.

51.

Klein

(1989) Why does skeletal part representation differ between smaller and larger bovids at Klasies River Mouth and other archaeological sites? Journal of Archaeological Science 6: 363–381.

52.

Kuiters

Groot Bruinderink

WTA

Lammertsma

(2005) Facilitative and competitive interactions between sympatric cattle, red deer and wild boar in Dutch woodland pastures. Acta Theriologica 50: 241–252.

53.

Lande

Engen

Sæther

B-E

(2003) Stochastic Population Dynamics in Ecology and Conservation. Oxford: Oxford University Press.

54.

Langvatn

(1998) Hjortens erobring av Norge. In: Brox

(ed.) Brennpunkt Natur 98/99. Trondheim: Tapir, pp. 49–71.

55.

Lenarz

Nelson

Schrage

. (2009) Temperature mediated moose survival in northeastern Minnesota. Journal of Wildlife Management 73: 503–510.

56.

Liberg

Wahlström

(1995) Habitat stability and litter size in theCervidae – A comparative analysis. In: Wahlström

(ed.) Natal Dispersal in Roe Deer: An Evolutionary Perspective. PhD thesis, Stockholm University, pp. 1–60.

57.

Lie

(1991) Elgens historie i Norge. Elgen 1991: 9–13.

58.

Liljegren

Ekström

(1996) The terrestrial late glacial fauna in South Sweden. Acta Archaeologica Lundensia. Series in 8 24: 135–139.

59.

Linnell

JDC

Zachos

(2011) Status and distribution patterns of European ungulates: Genetics, population history and conservation. In: Putman

Apollonio

Andersen

(eds) Ungulate Management in Europe: Problems and Practices. Cambridge: Cambridge University Press, pp. 12–53.

60.

Loe

Mysterud

Veiberg

. (2009) Negative density-dependent emigration of males in an increasing red deer population. Proceedings of the Royal Society B 276: 2581–2587.

61.

Lyman

(1985) Bone frequencies: Differential transport, in situ destruction, and the MGUI. Journal of Archaeological Science 12: 221–236.

62.

Markgren

(1966) A study of hand-reared moose calves. Viltrevy 4: 1–42.

63.

Mitchell-Jones

Amori

Bogdanowicz

. (1999) The Atlas of European Mammals. London: Poyser.

64.

Moe

(1979) Tregrense-fluktuasjoner på Hardangervidda etter siste istid. In: Nydal

Westin

Hafsten

(eds) Fortiden i søkelyset: 14C datering gjennom 25 år. Trondheim: Laboratoriet for Radiologisk Datering, pp. 199–208.

65.

Moe

Vorren

K-D

Alm

. (1996) Norway. In: Berglund

Birks

HJB

Ralska-Jasiewiczowa

(eds) Palaeoecological Events During the Last 15 000 Years: Regional Syntheses of Palaeoecological Studies of Lakes and Mires in Europe. Chichester: Wiley, pp. 153–213.

66.

Moen

(1999) National Atlas of Norway: Vegetation. Hønefoss: Norwegian Mapping Authority.

67.

Myhre

(2004) Agriculture, landscape and society ca. 4000 BC – AD 800. In: Gjerdåker

Lunden

Myhre

(eds) Norwegian Agricultural History. Trondheim: Tapir Academic Press, pp. 12–77.

68.

Mysterud

(2000) Diet overlap among ruminants in Fennoscandia. Oecologia 124: 130–137.

69.

Mysterud

(2010) Still walking on the wild side? Management actions as steps towards ‘semi-domestication’ of hunted ungulates. Journal of Applied Ecology 47: 920–925.

70.

Mysterud

Sæther

B-E

(2011) Climate change and implications for the future distribution and management of ungulates in Europe. In: Putman

Apollonio

Andersen

(eds) Ungulate Management in Europe: Problems and Practices. Cambridge: Cambridge University Press, pp. 349–375.

71.

Mysterud

Stenseth

Yoccoz

. (2001) Nonlinear effects of large-scale climatic variability on wild and domestic herbivores. Nature 410: 1096–1099.

72.

Nesje

(2009) Latest Pleistocene and Holocene alpine glacier fluctuations in Scandinavia. Quaternary Science Reviews 28: 2119–2136.

73.

Nummedal

(1913) Bjørneremsfundet: Forhistoriske hulefund fra Mien i Romsdalen. DKNVS Skrifter 1912: 1–41.

74.

Olsen

(1976) Skipshelleren. Osteologisk materiale. Bergen: Zoologisk Museum, Universitetet i Bergen.

75.

Olsen

Shetelig

(1934) Runekammen fra Setre. Bergens Museums Årbok 1933: 1–92.

76.

Olsson

Austrheim

Grenne

(2000) Landscape change patterns in mountains, land use and environmental diversity, mid-Norway 1960–1993. Landscape Ecology 15: 155–170.

77.

Overland

Hjelle

(2009) From forest to open pastures and fields: Cultural landscape development in western Norway inferred from two pollen records representing different spatial scales of vegetation. Vegetation History and Archaeobotany 18: 459–476.

78.

Overvoll

Wiers

(2004) Viltet i Vaksdal – kartlegging av viktige viltområde og status for viltartane. MVA-rapport 8/2004: 1–51.

79.

Perkins

Daly

(1968) A hunters’ village in Neolithic Turkey. Scientific American 219: 97–106.

80.

Petersen

(1936) Gamle gårdsanlegg i Rogaland. Fortsettelse: Utsira, Lyngaland, Håvodl, Birkelandstølen, Hanaland. Oslo: Aschehoug.

81.

Petersen

(1910) Hestneshulen: Beretning om undersøkelsen av en forhistorisk boplads paa Hitteren. DKNVS Skrifter 1910: 1–50.

82.

Prescott

(1991) Late Neolithic and Bronze Age developments on the periphery of southern Scandinavia. Norwegian Archaeological Review 24: 35–48.

83.

Reimer

Baillie

MGL

Bard

. (2009) IntCal09 and Marine09 radiocarbon age calibration curves, 0–50,000 years cal BP. Radiocarbon 51: 1111–1150.

84.

Renecker

Hudson

(1986) Seasonal energy expenditures and thermoregulatory responses of moose. Canadian Journal of Zoology 64: 322–327.

85.

Rolstad

Framstad

Gundersen

. (2001) Naturskog i Norge. Definisjoner, økologi og bruk i norsk skog- og miljøforvaltning. Aktuelt fra skogforskningen 1: 1–53.

86.

Root

Price

Hall

. (2003) Fingerprints of global warming on wild animals and plants. Nature 421: 57–60.

87.

Rosvold

Andersen

(2008) Wild boar in Norway – Is climate a limiting factor? NTNU Vitenskapsmuseet Rapport Zoologisk Serie 1: 1–23.

88.

Rosvold

Halley

Hufthammer

. (2010) The rise and fall of wild boar in a northern environment: Evidence from stable isotopes and subfossil finds. The Holocene 20: 1113–1121.

89.

Rosvold

Røed

Hufthammer

. (2012) Reconstructing the history of a fragmented and heavily exploited red deer population using ancient and contemporary DNA. BMC Evolutionary Biology 12: 191.

90.

Rueda

Rodríguez

MÁ

Hawkins

(2010) Towards a biogeographic regionalization of the European biota. Journal of Biogeography 37: 2067–2076.

91.

Schmölcke

Zachos

(2005) Holocene distribution and extinction of the moose (Alces alces, Cervidae) in Central Europe. Mammalian Biology 70: 329–344.

92.

Semiadi

Holmes

Barry

. (1996) Effects of cold conditions on heat production by young sambar (Cervus unicolor) and red deer (Cervus elaphus). Journal of Agricultural Science 126: 221–226.

93.

Sexton

McIntyre

Angert

. (2009) Evolution and ecology of species range limits. Annual Review of Ecology Evolution and Systematics 40: 415–436.

94.

Simpson

Webster

AJF

Smith

. (1978) Energy and nitrogen metabolism of red deer ( Cervus elaphus) in cold environments: A comparison with cattle and sheep. Comparative Biochemistry and Physiology 60: 251–256.

95.

Skjølsvold

(1977) Slettabøboplassen: Et bidrag til diskusjonen om forholdet mellom fangst- og bondesamfunnet i yngre steinalder og bronsealder. Arkeologisk museum i Stavanger - Skrifter 2: 1–340.

96.

Sommer

Zachos

Street

. (2008) Late Quaternary distribution dynamics and phylogeography of the red deer (Cervus elaphus) in Europe. Quaternary Science Reviews 27: 714–733.

97.

Speed

JDM

Austrheim

Hester

. (2010) Experimental evidence for herbivore limitation of the treeline. Ecology 91: 3414–3420.

98.

Stacey

Taper

(1992) Environmental variation and the persistence of small populations. Ecological Applications 2: 18–29.

99.

Stuiver

Reimer

(1993) Extended C-14 data base and revised CALIB 3.0 C-14 age calibration program. Radiocarbon 35: 215–230.

100.

Telfer

Kelsall

(1979) Studies of morphological parameters affecting ungulate locomotion in snow. Canadian Journal of Zoology 57: 2153–2159.

101.

Travis

JMJ

Dytham

(1999) Habitat persistence, habitat availability and the evolution of dispersal. Proceedings of the National Academy of Sciences USA 266: 723–728.

102.

van Beest

Van Moorter

Milner

(2012) Temperature-mediated habitat use and selection by a heat-sensitive northern ungulate. Animal Behaviour 84: 723–735.

103.

Walther

Post

Convey

. (2002) Ecological responses to recent climate change. Nature 461: 389–395.

104.

Webb

Bartlein

(1992) Global changes during the last 3 million years: Climatic controls and biotic responses. Annual Review of Ecology and Systematics 23: 141–173.

105.

Williams

Ives

Applegate

(2003) Population dynamics across geographical ranges: Time-series analyses of three small game species. Ecology 84: 2654–2667.

106.

Wolfe

(1974) An overview of moose coactions with other animals. Naturaliste Canadien 101: 437–456.