Introduction The Jostedalsbreen and Jotunheimen regions of southern Norway have been foci of research on Holocene glacier variations for many years. As a result, a detailed chronology of glacier and climatic variations has revealed both similarities and differences between these two regions, which are some 50 km apart (e.g. Matthews etai, 2000; Nesje etal., 2000, 2001). Little is known, however, about the glaciers and their fluctuations in the relatively remote mountain area of Breheimen, east of Jostedalsbreen and northwest of Jotunheimen (Fig. 1). This study represents the first part of an investigation of the Holocene glacier and climatic history of this area, which occupies a transitional location on a steep climatic gradient between the relatively maritime and continental climates respectively of Jostedalsbreen and Jotunheimen. The project as a whole is using a range of methods, including the use of moraine mapping and surface dating, and glaciofluvial and glaciolacustrine stratigraphy, which may be characterised as a 'multiproxy' approach. The Neoglacial maximum extent of glaciers in southern Norway is widely accepted to have occurred in the Little Ice Age (Matthews, 1991; Winkler, 1996a; Dahl etal., 2002). Within the Little Ice Age, most glaciers in the Jostedalsbreen region (0strem etal., 1976; Grove, 1985, 1988; Erikstad and Sollid, 1986; Bogen etal., 1989; Bickerton and Matthews, 1993) and in Jotunheimen (Matthews, 1974, 1977; Matthews and Shakesby, 1984; Erikstad and Sollid, 1986) attained their maximum extents in the eighteenth century. In this paper, we focus on the moraine-ridge sequences in front of the low-altitude', warm-based (temperate) glaciers of the Breheimen region. Our investigations of the higher altitude 396 glaciers, which are of the cold-based (polythermal or subpolar) type (Paterson, 1 994; Benn and Evans, 1998), and of the distal glaciofluvial and glaciolacustrine sedimentary sequences, will be reported separately. This paper has five main aims: 1 to describe the moraine-ridge sequences of the low-altitude glaciers; 2 to identify the Little Ice Age moraines and differentiate them from any older Holocene moraines; 3 to date the moraines as accurately as possible using a combination of lichenometric dating, Schmidt hammer 'R-values' and radiocarbon dating; 4 to detect any synchroneity of glacier response to decadal-scale climatic fluctuations during the Little Ice Age interval; 5 to compare the pattern and timing of Little Ice Age glacier variations in Breheimen with those already established for Jostedalsbreen and Jotunheimen. Study area Breheimen refers to the region between Lustrafjord and the major valleys of Jostedalen, Ottadalen and Boverdalen (Fig. 1). Although a number of mountain peaks rise to over 2000 m above sea-level, most of the region consists of high plateau areas around 1500-1800 m underlain by granitic and migmatic gneiss (Sigmond etai, 1984; Lutro and Tveten, 1996) dissected by glaciated valleys such as Bratadalen/Mysubyttdalen, Lundadalen and Fortunsdalen. The seven low-altitude glaciers under investigation here (Table 1 and Fig. 1; cf. 0strem et a/., 1988) were selected because they have the best developed moraine-ridge sequences. These glaciers are outlets from either small ice-caps or cirque glaciers, and five of them descended during the Little Ice Age on to valley floors between 1000-1350 m a.s.l. As the forelands of these glaciers lie above the regional tree line in the alpine zone, they are closer in size and character to those of Jotunheimen than to the larger outlet glaciers of Jostedalsbreen. Although no meteorological data are available from the core of Breheimen, data from the nearest meteorological stations are summarised in Table 2 (Aune, 1 993; Forland, 1 993). Gjeilo i Skjak and Brata, to the northeast and north, respectively, being located in one of the driest regions of southern Norway, are unlikely to be representative of the climate of Breheimen. Stations to the west and southwest (Bjorkehaug i Jostedalen and Fortun), being exposed to a more maritime climate, appear to be similarly unrepresentative but are indicative of the precipitation gradient. Short-term glaciometeorological measurements carried out at Harbardsbreen in central Breheimen in the late 1990s (Kjollmoen, 2000) suggest fairly Table 1 Characteristics of the seven 'low-altitude' glaciers investigated in Breheimen (modified from 0strem etai, 1988) Glacier Area Glacier Glacier Aspect Number (km2) snout foreland of altitude altitude3 moraines (m) (m) Tverreggibreen 1.77 1400 1050 Northeast 8 Storegrovbreen 9.13 1380 1300 South 13 Greinbreen 1.22 1180 1240 North 8 Heimste Breen 3.55 1580 1200 North 8 Ytste Breen 3.00 1560 1200 North 7 Vesldalstindbreen 0.13 1440 1320 Notheast 2 Nordre Holäbreen 9.87 1600 1320 Southeast 6 a Lowest altitude of outermost moraine. Copyright © 2003 John Wiley & Sons, Ltd. 397 Table 2 Climatic data from meteorological stations close to Breheimen (from Aune, 1993; Forland, 1993) Station Altitude Mean air temperatu re (X) Mean annual (m) precipitation January July Year (mm) Bjorkhaug 324 -4.9 13.4 3.7 1380 Fortun 27 -5.1 14.2 4.4 739 Sognefjell 1413 -10.7 5.7 -3.1 860 B rátá 712 -8.6 11.5 1.3 548 Gjeilo 378 -9.4 13.9 2.8 295 close correspondence with the data from Sognefjell to the south, which has a mean annual precipitation of 860 mm and a mean annual temperature of —3.1 "C at an altitude of 1413 m (Aune, 1993; Forland, 1993). The short-term glacio-logical studies at Harbardsbreen also confirm the character of the glaciers in Breheimen with a lower glacier mass turnover than in western Norway, but no significant net balance loss as in Jotunheimen. During fieldwork, no signs of recent glacier advances were indicated in the morphology of several glacier snouts visited in Breheimen. Methods The glacier forelands and individual moraine ridges were identified and mapped on vertical aerial photographs (e.g. Fig. 2) and checked in the field. Three geochronological methods were applied to the moraine ridges: lichenometric dating, relative-age dating based on the degree of surface weathering of boulders as indicated by Schmidt-hammer readings, and radiocarbon dating of buried soils and associated organic material. Lichenometry The moraines were dated in detail by lichenometry (cf. Innes, 1985a; Worsley, 1990; Matthews, 1994) with the aim of detecting synchronous moraine formation and hence possible Little Ice Age climatic fluctuations. The longest axes of specimensof the yellow-green crustose lichensof Rhizocarpon subgenus (Innes, 1985b; Poelt, 1988) were measured to the nearest millimetre from 25-m lengths of moraine proximal to the crest. Formerly known as Rhizocarpon geographicum agg., the lichens measured include specimens of both section Rhizocarpon and section Alpicola. The taxonomic group and procedures used ensured comparability with previous work in southern Norway (Matthews, 1974, 1994; Erikstad and Sollid, 1986; Bickerton and Matthews, 1992, 1993). Application of lichenometry in Breheimen posed two main problems. First, there is no documentary evidence for the control points required for constructing local or regional lichenometric dating curves. Curves constructed for Jostedalsbreen and Jotunheimen (Matthews, 1974; Erikstad and Sollid, 1986; Bickerton and Matthews, 1992) therefore had to be used. Second, as the moraines were commonly short, fragmented and/or located in suboptimal positions for lichen growth (e.g. on unstable slopes or where disturbance by avalanches and fluvial activity had occurred), 25-m lengths of undisturbed moraine surfaces were difficult to find. Consequently, it was not possible to utilise the most Figure 2 Vertical aerial photograph of the glacier foreland of Storegrovbreen (compare with the map in Figure 5) extensively used approach based on five 25-m long sites (search areas) per moraine, which is generally considered to provide the most reliable lichenometric dates (e.g. Matthews, 1974, 1975). Taking into account these problems, and the greater degree of accordance in general of Breheimen climatic conditions to those of Jotunheimen, most reliance has been placed on the lichenometric dating curve (1.5 using the notation of Matthews, 1974) from Storbreen, Jotunheimen, based on the five largest lichens from the single site on each moraine with the largest lichens (equation 1): log(y + 60) = 1.8601 + 0.0054x (1) where x is lichen size in millimetres and y is moraine age in years. This avoids potential problems of unrepresentative single lichens and is available for all the moraines and moraine fragments under investigation. We have, however, dated all the moraines using a range of other curves, especially those defined by equations (2)-(4): log(y + 30) = 1.6597 + 0.0065* (2) ln(y+ 150) = 5.0299+ 0.0071 x (3) log(y+ 123) = 2.080+ 0.0039x (4) Equation (2) is based on the single largest lichen per moraine, from Storbreen, Jotunheimen (1.1 from Matthews, 1974). Equation (3) is based on single largest lichens from moraines and other control points of known age in the Jostedalsbreen region (one of the 'western curves' of Erikstad and Sollid, 1986). Equation (4) is based on the five largest lichens from 398 the optimum single site per moraine at Nigardsbreen, a major outlet glacier of the Jostedalsbreen ice-cap for which historical evidence of moraine age has been firmly established (1.5 from Bickerton and Matthews, 1992). Results and interpretation Results for each individual glacier foreland are considered as a prelude to construction of a composite Little Ice Age moraine chronology for all seven Breheimen low-altitude glaciers. Schmidt hammer 'R-values' The Schmidt hammer is used here to identify possible pre-Little Ice Age moraines, an approach that was pioneered in Jotunheimen (Matthews and Shakesby, 1984; McCarroll, 1989a, 1991, 1994). Measurements were carried out on samples of 50 boulders (one impact per boulder) on outermost moraine ridges, and on additional sites both outside the glacier foreland (normally comprising boulder surfaces but sometimes bedrock outcrops) and on selected inner moraines. Unstable boulders were avoided and near-horizontal, lichen-free boulder surfaces were selected where possible. The gneissic rock types dominating Breheimen provided, except for the Hestbrepiggane granitic area, widely jointed blocky boulders favourable for Schmidt hammer use (cf. Winkler and Shakesby, 1995; Winkler, 2000). If the outermost moraine on a glacier foreland dates from the Little Ice Age, the mean 'R-value' (rebound value) from the outermost moraine will be indistinguishable from unweathered boulders from inner moraines of relatively recent origin. If, on the other hand, the mean is similar to that obtained for a sample of boulders from outside the outermost moraine, the latter could date from much earlier in the Holocene, with the exception of any bulldozed boulders incorporated into the moraine (e.g. Matthews and Shakesby, 1984; McCarroll, 1989b). Statistical significance of differences between means was assessed using 95% confidence intervals (Matthews, 1981). Radiocarbon dating Excavations were made at seven sites in the distal slope of the outermost moraine at Tverreggibreen, Storegrovbreen and Greinbreen to search for datable buried organic material. No suitable sites were found at the other four glaciers. Buried arctic-alpine podsols, brown soils and humic regosols (Ellis, 1979, 1980) were radiocarbon dated to provide maximum estimates of moraine age (Matthews, 1985, 1993a). At both Tverreggibreen and Greinbreen, the stratigraphy of distal mires close to the outermost moraines revealed inwashed minerogenic layers associated with the timing of the Little Ice Age glacial maximum. For the former glacier, organic material buried beneath fluvioglacial sediments near the outermost terminal moraine provided additional evidence for the timing of glacier variations prior to the Little Ice Age maximum. Radiocarbon dating was carried out using conventional gas-proportional counting techniques and acid-washed, 1 -cm-thick samples unless stated otherwise. Thin slices of organic material in soils and peats were used to investigate age-depth gradients near the buried surfaces in the manner described by Matthews (1980, 1993a). The dates were calibrated using the computer program of Stuiver and Reimer (1993) and the data set in Stuiver etal. (1998). Tverreggibreen Officially '0vreOtta No. 24' (0strem eta/., 1988), this glacier is the northeastern outlet of a rapidly thinning, small ice-cap on Tverreggi (1 768 m; Fig. 1). The present glacier snout descends toca. 1400 m but the outermost moraineof the glacier foreland reaches the valley bottom of Mysubyttdalen at about 1050 m. A sequence of eight arcuate recessional moraines, some fragmented and partially buried by glaciofluvial and colluvial activity, occurs on the glacier foreland (Fig. 3). Lichen sizes and predicted moraine ages are summarised in Table 3. Lichen sizes on the outermost moraine ridge (M1) suggest a date of ad 1734 according to the Storbreen lichenometric dating curve (1.5; equation 1) but the lichens on the second moraine (M2) are some 1 0 mm larger yielding a date earlier in the late-seventeenth century. The difference in predicted ages of M1 and M2 is clearly indicative of the potential scale of the dating errors resulting from local differences in microclimate and other factors affecting lichen growth. Even larger differences in predicted age result from use of the lichenometric dating curves from the Jostedalsbreen region (equations 3 and 4), which that suggest local and regional differences in lichen growth also occur within Breheimen. There is nevertheless a consistent pattern of lichen sizes and predicted dates across the remaining moraines in the sequence. Only short fragments of M3 and M4 are undisturbed, however, which casts doubt on the accuracy of the dates estimated. Steep bedrock cliffs behind moraine M9 (predicted date ad 1908) would have prevented the deposition of any younger moraines after this date. The location of this relatively low-altitude glacier, with a northeastern aspect near the western margin of Breheimen, may mean that the local conditions on the foreland of Tverreggibreen are relatively favourable for lichen growth and hence may exaggerate moraine age when estimates are based on lichenometric dating curves from Jotunheimen. Comparison of Schmidt hammer readings from M1 and M2 and on a bedrock outcrop outside the glacier foreland do not suggest any pre-Little Ice Age moraine formation (Table 4). There is no statistically significant difference (p > 0.05) between the mean R-values for sites on M1 and M2, both of which are significantly higher than the mean for the well-weathered bedrock site by at least 10 R-value points but similar to those obtained from unweathered, recently exposed rock surfaces. Thus, according to the Schmidt hammer results, all moraines were apparently deposited during the Little Ice Age. Radiocarbon dates from four sites are summarised in Table 5 and Figs 3 and 4. At site 1 (Fig. 4a), the outermost moraine is located at the edge of a small, shallow mire, and seven radiocarbon dates here relate to samples from five sedimentary units recognised within the mire. At the base, unit 1 is a buried humus-iron podsol consisting of a thin (1 cm thick), dark brown, surface AD horizon underlain by a 5-cm-thick pinkish-grey, leached A2 horizon, below which are well-developed illuvial Bh (black) and Bs (orange-brown) i 11 u via I horizons. Alternating minerogenic and organic units 2-5 lie 399 Tverreggibreen 100 m ^ Glacier ice | Snow Glacier foreland ~I\ Vegetated area outside :—' the glacier foreland T"| Bedrock with or without a sparse cover of surface deposits M1 - 8 Numbered moraine ridges Boulders Cobbles Gravel Sand Lake or river !,l'l Moraine ridges Fluted moraine Bedrock cliffs and erosional scarps Streams and dry gullies v«+ Levee I site 1| Site of excavation Figure 3 The g acier foreland and moraine sequence of Tverregg breen Table 3 Lichen sizes and lichenometric dating results from seven low-altitude j glaciers in Breheimen: note that predicted dates usint equation (1) have been used n Fig. 11 Glacier and Largest Mean of the Predicted dates (ad) Number of moraine lichen five largest from numbered equations3 sites per (mm) lichens (mm) moraine 1 2 3 4 Tverreggibreen M1 138 120.6 1734 1668 1742 1767 6 M2 149 133.8 1677 1604 1709 1722 6 M3 112 97.0 1817 1785 1810 1835 1 M4 105 89.4 1839 1809 1825 1854 1 M5 101 93.2 1828 1822 1836 1844 3 M6 100 93.6 1827 1825 1838 1843 4 M7 91 80.8 1861 1851 1857 1874 8 M8 65 58.8 1908 1908 1906 1918 9 Storegrovbreen M1 118 98.4 1813 1762 1796 1831 2 M2 117 97.8 1815 1766 1798 1833 2 M3 110 87.2 1845 1792 1815 1859 2 M4 83 77.2 1870 1871 1873 1882 1 M5 80 75.2 1874 1878 1879 1886 3 M6 82 70.8 1884 1873 1875 1895 6 M7 77 70.6 1885 1884 1885 1895 2 M8 66 64.0 1898 1906 1905 1908 2 M9 64 59.0 1908 1910 1908 1918 6 M10 48 42.8 1936 1935 1934 1945 2 M11 53 46.0 1931 1928 1926 1940 7 M12 32 30.8 1953 1955 1957 1963 1 M13 29 27.6 1957 1958 1961 1968 1 (continued overleaf) Copyright © 2003 John Wiley & Sons, Ltd. 400 Table 3 (Continued) Glacier and Largest Mean of the Predicted dates (ad) Number of moraine lichen five largest from numbered equations3 sites per (mm) lichens (mm) - moraine 12 3 4 Greinbreen M1 111 104.6 1793 1798 1813 1814 4 M2 89 70.6 1885 1856 1861 1895 2 M3 81 69.0 1888 1875 1877 1899 1 M4 77 67.8 1891 1984 1885 1901 1 M5 60 56.0 1914 1917 1915 1923 2 M6 73 64.6 1897 1893 1892 1907 2 M7 43 42.0 1937 1942 1941 1947 1 M8 42 36.4 1945 1943 1943 1955 1 Heimste Breen (Hestbrepiggane) M1 124 107.2 1784 1737 1780 1807 7 M2 100 93.6 1817 1825 1838 1843 1 M3 112 100.4 1807 1785 1810 1826 3 M4 80 73.4 1879 1878 1879 1890 2 M5 64 60.2 1906 1910 1908 1916 3 M6 68 60.2 1906 1903 1901 1916 2 M7 52 44.4 1933 1930 1928 1943 4 M8 42 39.8 1940 1943 1943 1950 2 Ytste Breen (Hestbrepig igarie) M1 131 124.8 1717 1705 1761 1753 2 M2 128 115.4 1755 1719 1770 1783 2 M3 107 102.6 1799 1802 1822 1820 2 M4 94 85.2 1850 1842 1851 1864 3 M5 70 67.2 1892 1899 1898 1902 1 M6 84 75.8 1873 1868 1871 1885 1 M7 87 81.2 1860 1861 1865 1873 1 LM1b 53 52.0 — — — — 2 LM2 135 126.0 1712 1684 1750 1749 2 LM3 28 25.0 — — — — 2 Vesldalstindbreen M1 112 100.2 1807 1785 1810 1826 7 M2 93 81.8 1859 1845 1853 1871 5 Nordre Holäbreen M1 125 112.2 1767 1732 1776 1793 7 M2 110 103.2 1798 1792 1815 1818 5 M3 115 101.4 1808 1774 1803 1823 6 M4 101 99.2 1810 1822 1836 1829 5 M5 98 84.6 1852 1831 1842 1865 6 M6 82 80.6 1862 1873 1875 1874 6 a Equations (1)-(4) are given in the text. b Lateral moraine ridges high on plateau (see Fig. 9). above the buried soil. Units 4 (light brownish-grey, organic-stained medium sand) and 2 (light-grey, fine sandy silt) are separated by a 2-cm-thick brown peat (unit 3). The 11-cm-thick surface organic layer of the mire (unit 1) is largely unhumified orange-brown peat. The radiocarbon dates from site 1 are consistent with mire development on the podsolic soil when Little Ice Age moraine deposition caused ponding of water against a bedrock knoll. Prior to moraine formation, the presence of a well developed podsolic soil at the site would have required prolonged, undisturbed soil formation. It is probable that the formation of this mature humo-ferric podsol would have required several thousand years to form in this arctic-alpine environment (cf. Ellis and Matthews, 1984). Colluvial deposition of sandy minerogenic sediments from the distal slope of the moraine certainly began some time after 1 720 ± 50 14C yr BP (SWAN-436). This date is unlikely to represent a close estimate of the time elapsed since moraine formation because even thin surface organic horizons of well-developed arctic-alpine podsols can contain carbon with a considerable apparent mean residence time (Matthews and Dresser, 1983; Matthews, 1993a,b). This interpretation is borne out by the dates of 200 ±50 and 330 ±50 14C yr BP (SWAN-435 and SWAN-438, respectively) obtained from the base of unit 3, which represent close maximum estimates for a short-term cessation of colluvial deposition. As the difference between these two estimates does not differ statistically, they can be combined to produce a mean date of 265 ± 35 14C yr BP, which, when calibrated, yields an intercept age of 303 cal. yr BP or ad 1647 (upper and lower 2a age limits are 430 and 153 cal. yr BP, respectively, or ad 1 520 and 1 797). Following the short interval of stabilisation represented by unit 3, renewed minerogenic deposition (unit 2) occurred: as the sediments are not stained with organic matter and are relatively fine grained and well sorted, they probably represent a low-energy glaciofluvial deposit rather than colluvium. This deposition event had ceased by 165 ± 35 14C yr BP (a mean of SWAN-433 and SWAN-437, derived from the base of unit Copyright © 2003 John Wiley & Sons, Ltd. 401 Table 4 Schmidt hammer results for outermost moraines, inner moraines and sites outside the glacier forelands: values are mean 'R-values' and 95% confidence intervals (n = 50) for sites at seven low-altitude glaciers in Breheimen Glacier Outside sitea Outermost moraine Inner morainesb Tverreggibreen 47.06 ± 2.1 9C 47.66 ±2.28c 57.58±2.06 59.76 ±1.86 (M2) Storegrovbreen 43.92 ±2.91 50.44 ± 3.05 (M1) 57.26±2.74 (M1) 53.54 ±2.94 (M2) 54.62 ± 3.01 (M2) 63.50 ± 1.67 (M2) 63.1 6 ±1.80 (M10) 63.1 6 ±1.60 (M11) Greinbreen 38.22 ±2.73 50.30 ±2.73 51.22 ±2.39 51.54±2.64 51.56 ± 2.18 52.52 ±2.45 (M2) 53.40 ±2.99 (M2) 50.62 ±2.28 (M3) 55.12 ±2.46 (M4) 54.14±2.58(M5) Heimste Breen 49.64 ±3.50 50.62 ± 3.08 52.96±2.92 60.86 ±2.24 No inner site recordedd Ytste Breen 46.48 ± 3.49 60.86 ±2.1 8 57.64 ±2.40 (LM2) 49.72 ± 3.00 57.62 ± 3.77 57.62 ± 1.76 (LM1) 59.56± 1.82 (LM1) 57.94 ±2.10 (LM2) 58.50 ±2.1 7 (LM3) 59.26 ±2.27 (LM3) 60.48 ± 1.73e 60.86±2.31e Vesldalstindbreen 47.37 ±2.44 53.52 ± 1.99 55.98± 1.87 55.52 ±1.82 (M2) Nordre Holäbreen 48.54 ± 3.47 58.24 ± 1.63 59.26± 1.59 57.1 7 ± 1.43 (M6) a Sites outside the glacier foreland. b Recessional moraines on the glacier foreland (excluding the outermost moraine, M1). c Bedrock surfaces outside the glacier foreland (not boulders). d Inner sites from Ytste Breen considered sufficient. e Boulders inside the glacier foreland (but not on a particular moraine ridge). Table 5 Radiocarbon dating results from Tverreggibreen, Storegrovbreen and Greinbreen Glacier and site Laboratory number l4Cage (yrBP) Material Deptr i (cm) <5nC (%) Calibrated agea (cal. yr BP) Tverreggibreen Site 1 SWAN-432 170 ±50 Peat 9.0- 10.0 -28.0 305 (2 74, 1 86, 180, 1 75, 148, 11, 4) 0b SWAN-433 100 ±50 Peat 10.0- -11.0 -28.0 282 (240, 232, 126, 125, 65, 38, 0) 0b SWAN-437 230 ±50 Peat 10.0- -11.0 -27.4 428 (291) 2 SWAN-434 150 ±40 Peat 20.0- -21.0 -26.7 291 (268, 216, 144, 19, 3) 0b SWAN-435 200 ± 50 Peat 21.0- -22.0 -27.0 313 (282, 168, 155) 0 SWAN-438 330 ±50 Peat 21.0- -22.0 -27.3 506 (428, 377, 323) 291 SWAN-436 1720 ±50 Podsol A0 28.0- -29.0 -25.4 1732 (1687, 1675, 1612) 1524 Site 2 SWAN-439 20 ±50 Podsol A0 20.0- -21.0 -26.2 257 (0b) 0b SWAN-440 1100 ±50 Podsol A0 21.0- -22.0 -25.8 1168 (1046, 1040, 974) 928 Site 3 SWAN-431 290 ±40 Podsol A0 6.0- -7.0 -24.6 467 (310) 160 Site 4 SWAN-427 200 ± 40 Immature soil 11.0- -11.5 -25.5 308 (282, 168, 155) 1 SWAN-428 660 ± 50 Immature soil 21.0- -22.0 -25.4 675 (651, 575, 575) 546 SWAN-429 680 ± 60 Immature soil 22.0- -23.0 -25.2 707 (654) 546 SWAN-430 1250 ±70 Immature soil 26.0- -27.0 -24.2 1295 (1175) 987 Storegrovbreen Site 1 SWAN-422 680 ± 50 Brown soil 15.0- -16.0 -24.9 688 (654) 551 SWAN-424 800 ± 50 Brown soil 16.0- -17.0 -24.7 791 (694) 659 SWAN-425 2050 ±50 Brown soil 20.0- -22.0 -23.3 2146 (1995) 1885 Site 2 SWAN-426 460 ± 50 Brown soil 26.0- -27.0 -24.1 548 (511) 341 Greinbreen Site 1 CAR-1398 860 ± 60 Eriophorum sp. 25.0- -28.0 -25.9 926 (758, 751, 742) 669 CAR-1399 680 ± 50 Peat 12.0- -15.0 -25.4 688 (654) 551 CAR-1400 960 ± 60 Peat 15.0- -15.5 -26.6 969 (916) 733 CAR-1401 680 ± 60 Peat 15.5- -16.0 -26.5 707 (654) 546 Site 2 SWAN-366 750 ± 50 Humic regosol 27.0- -28.0 -26.2 756 (672) 571 a Calibrated ages include intercept ages in brackets and 2cr range. b Possible influence of 'bomb' carbon. Copyright © 2003 John Wiley & Sons, Ltd. 402 TVERREGGIBREEN o. E Q — North 4C dates (yr BP) 170 ±50 100 ± 50 ; 230 ±50 150 ±40 200 ± 50 ; 330 ± 50 B Site 2 Units South ® 1 North. S-West |<^x o ff 5) -a o £ o ff oj ff oj oj ff m E s> 'o & x > oj ff .a T3 2 1950 - 1900 - s 1850 < < o < g 1800 1750 - 1700 -- <2 MORAINE FREQUENCY 6 4 2 0 ...... m 2/3 M W- 3 one moraine- i i i i i i 6 4 2 B MORAINE CLUSTERS e3 n 2/3^ 1 il i ta 1 i n I 3/4^ m 4/5 y □ 4/5 i 3 i 1 1 c cl) c cd c 0) _Q 0) E co E 'oj co 'oj nh ed sz ff co O o qq ~d - 1950 - 1900 i r i i 1850 - 1800 - 1750 1700 Figure 11 Lichenometric dates and the composite Little Ice Age moraine chronology for Breheimen: (A) dated moraine sequences at seven individual glaciers; (B) histogram of moraine frequency (weighted); (C) moraine clusters (further explanation in the text) Copyright © 2003 John Wiley & Sons, Ltd. 409 data). Second, owing to the lack of surfaces of known age from within Breheimen, the 1.5 lichenometric dating curve from Storbreen, Jotunheimen, was used (equation 1). Third, all dated moraines have been included in the histogram and moraine clusters, rather than only those with a minimum of four sites per moraine (although moraines with only one site per moraine are differentiated and treated with extreme caution). This method has led to the identification of eight main clusters (shaded in Fig. 11C), which contain 20 of the 45 moraine ridges included in the study. Five of the eight clusters are each represented at three or four glaciers. Two more clusters include moraines from three glaciers if moraines consisting of only one site are included. One apparent cluster around ad 1830 of two moraines has been excluded as these moraines occur on the same glacier foreland. The oldest cluster dates from ad 1793 to 1799. Although earlier moraines exist at three glaciers there is insufficient evidence to establish synchronous glacier behaviour. Moraines dating from the ad 1 750s or 1 760s occur, however, at two glaciers (Yste Breen and Nordre Holabreen). In addition, older moraine ridges, possibly dating from the late seventeenth and early eighteenth centuries, occur at Tverreggibreen and Yste Breen, respectively. Five clusters occur within the nineteenth century at ad 1807-1813, 1845-1852, 1859-1862, 1879-1885 and 1897-1898. Two clusters occur in the early twentieth century—ad 1 906-1908 and 1 931 -1 933—despite the steep slopes of the upper valley sides that were not conducive to moraine deposition at that time. Discussion The ages of the recessional moraines and their climatic implications The composite Breheimen chronology appears to have successfully identified a regional glacier response. Broad synchroneity in glacier behaviour can be used in support of the general accuracy of the lichenometric dates. Two lines of independent evidence also support the accuracy of the composite chronology. First, the two youngest moraine clusters (ad 1906-1908 and 1931-1933) occur at the same time as known glacier advances of the Jostedalsbreen icecap, which have been documented historically and were closely associated with moraine-ridge deposition (Rekstad, 1902; Fasgri, 1950; Bickerton and Matthews, 1993; Winkler, 1996b). Moraines dating from about ad 1930 are also well known in Jotunheimen (Hoel and Werenskiold, 1962; Erikstad and Sol lid, 1986). Second, as there are commonly around eight moraine ridges on glacier forelands in Jotunheimen (Erikstad and Sollid, 1986; Matthews, unpublished), where glaciers are approximately the same size, recognition of eight moraine-date clusters in Breheimen is unlikely to be coincidental. Comparison of the Breheimen moraine clusters (Fig. 11 C) with those from Jostedalsbreen (Fig. 11 E) and Jotunheimen (Fig. 11D) reveals some further similarities with the timing of particular moraine clusters but there is no general agreement with either chronology from these neighbouring regions. Breheimen moraines dated ad 1793-1799, 1807-1813, 1845-1852, 1859-1862 and 1897-1898, correspond most closely with moraines in Jotunheimen dated to ad 1796-1802, 1812-1818, 1845-1854, 1860-1868 and 1886-1898, respectively. Breheimen moraines dating to ad 1879-1886 and 1906-1908 correspond most closely with Jostedalsbreen moraines dated to ad 1882-1892 and 1906-1911, respectively. Only the clusters around ad 1930, and possibly those around 1810 and the 1850s appear to be in close agreement across all three regions. The timing of recessional moraine formation in Breheimen is likely to represent a response to both summer temperature and winter precipitation. Bickerton and Matthews (1993) discussed the close association of formation of the Little Ice Age recessional moraines of Jostedalsbreen with runs of cool summers and summer temperature minima of some 1.0-1.5 "C below the average for the period ad 1700-1950. They concluded that the short-term glacier advances responsible for moraine-ridge formation represented a near-immediate response to climate of the glacier tongues, which was superimposed upon the longer term dynamic response of the ice-cap. The latter is clearly affected by the different reaction times of the longer and shorter tongues, which varies from 3 to 4 yr for the shorter tongues, such as Brigsdalsbreen, to 23-27 yr for the longer tongues, such as Nigardsbreen (Nesje, 1989; Winkler etal., 1997), and a variable influence of winter accumulation. Bogen etal. (1989) suggested that the outlets of Jostedalsbreen responded synchronously during the first three decades of the twentieth century but reacted differently to extensive melting and local changes in winter precipitation in later decades. Especially since the 1980s, increased winter precipitation has led to a positive net mass balance of the ice-cap and the largest glacier advances of the twentieth century (Nesje etal., 1995b; Winkler etal., 1997). It also has been demonstrated that the net balance for the maritime glaciers of southern Norway is more influenced by the winter balance than the summer balance, whereas the opposite holds for the more continental glaciers, such as Jotunheimen (Nesje etal., 1995b; Winkler etal., 1997; Nesje and Dahl, 2000). It is probable, therefore, that regional differences between the glacier response of Breheimen and that of Jostedalsbreen on the one hand and Jotunheimen on the other, is affected by differences in the relative importance of temperature and precipitation operating on more than one timescale (cf. Winkler and Nesje, 2000). Although these differences between regions may be partly attributed to inaccuracies in dating (see below), they must also be influenced by differences in response between glaciers within the Breheimen region. The latter include effects of the west-east climatic gradient across the region and local differences between glaciers, which include morphological, aspect and altitudinal differences. Seven glaciers are not sufficient to analyse this within-region variation in detail. However, the greatest differences within Breheimen are found between Storegrovbreen and Vesldalstindbreen—a western ice-cap that faces south and an eastern cirque glacier with a northerly aspect. Dating the Little Ice Age maximum and implications for Holocene glacier variations Lichenometric dating, radiocarbon dates and Schmidt hammer R-values together convincingly demonstrate that the low-altitude glaciers in Breheimen attained their Neoglacial maximum extent in the Little Ice Age. Differences between glaciers in the precise timing of the Little Ice Age maximum are suggested by the lichenometric dates. Although some differences reflect limitations of the method and are considered below, other differences are real and must reflect such factors as glacier size, morphology and dynamics. The conventional Copyright © 2003 John Wiley & Sons, Ltd. 410 view of a mid-eighteenth century Little Ice Age maximum seems at first sight to apply only at Nordre Holabreen. At three other glaciers (Storegrovbreen, Greinbreen and Heimste Breen) underestimation of moraine age owing to suboptimal lichen growth conditions is a possibility so that a mid-eighteenth century maximum cannot be ruled out. The smallest glacier (Vesldalstindbreen) seems to have attained its Little Ice Age maximum early in the nineteenth century, whereas Tverreggibreen and Ytste Breen may have attained their maxima in the late-seventeenth or early-eighteenth centuries. At Tverreggibreen, where the outermost moraine (M1) is extensive and a significantly older date is predicted, it is particularly difficult to argue for a mid-eighteenth century Little Ice Age maximum. Such methodological problems appear less important in relation to the younger recessional moraines. During the Little Ice Age, the glaciers overrode mature arctic-alpine soils that require several millennia to develop and the minimum age of which is shown by the oldest radiocarbon dates, whereas thin slices from the surface of these buried soils provide maximum estimates for the burial event and hence moraine age (cf. Matthews, 1985, 1991, 1993a). Although these radiocarbon-dating techniques are unable to define the timing of the Little Ice Age maximum exactly, the youngest radiocarbon dates provide precise limiting dates and show that it normally occurred late in the Little Ice Age, which is compatible with the lichenometric dating of the outermost moraines. In a few cases (SWAN-432, SWAN-433, SWAN-434, SWAN-439) the possibility of contamination of shallow-buried material by 'bomb effect' carbon cannot be ruled out (cf. Matthews, 1985; Dresser, 2001) but the likelihood of such young contaminants affecting these samples through root penetration or other methods is considered low in view of the number of consistent dates from a variety of materials from different sites and depths. Where the dates are older, there are three possible explanations: first, the apparent mean residence time may be greater at depth in mature soils (SWAN-436, SWAN-440, SWAN-425); second, the apparent mean residence time may be greater near the surface of mature, mid-alpine brown soils (SWAN 422 and SWAN 424); third, the dates actually relate to relatively early events (SWAN-428, SWAN-429, SWAN-430). The only radiocarbon-dating evidence pointing towards a pre-Little Ice Age, Neoglacial maximum is from Greinbreen, where the range of the calibrated dates for the outermost moraine is ad 981-1399. These dates are not fully explained, however, as they are contradicted by the lichenometric evidence and are difficult to explain in terms of contamination by relatively old material. Although no pre-Little Ice Age Neoglacial maximum has been established in Breheimen, there is evidence, based on the buried immature soils at Tverreggibreen (site 4), for relatively early glacier advances about ad 655-963 and ad 1277-1396. These short-lived soils have yielded reasonably precise dates for glaciofluvial burial events, which occurred before the glacier attained its Little Ice Age maximum and may have been associated with glacier advances. This conclusion is supported by glaciolacustrine studies in central Jotunheimen, where the Bovertun II Event indicates a glacier expansion episode about ad 550-1 150, prior to glacier contraction in the Mediaeval Warm Period (Matthews etal., 2000). Evidence for early Little Ice Age advances of glaciers (after the Mediaeval Warm Period) has also been found in the Jostedalsbreen region (Nesje and Dahl, 1991 a; Nesje and Rye, 1993) and elsewhere in southern Norway (Elven, 1978; Nesje etal., 1995a) but this invariably relates to glaciers that were smaller, usually considerably smaller, than at their Little Ice Age maxima. Further methodological implications In the application of lichenometric dating to dating moraines in the relatively remote region of Breheimen where conditions for the technique are not ideal, several problems have been highlighted. First, the use of lichenometric dating curves from a neighbouring region inevitably has reduced the general level of accuracy of predicted dates. The importance of local dating control has been emphasised in previous applications of lichenometry (Innes, 1985a; Worsley, 1990; Matthews, 1994). This study has demonstrated that well-dated lichenometric curves from neighbouring regions provide an alternative approach. Second, predicted dates are likely to be less accurate where searched sections of moraines are short (because optimal growth conditions may be rare). Third, individual moraines may be missing, either because they never existed on steep valley sides or have been destroyed, fragmented or disturbed owing to colluvial activity and glaciofluvial erosion. In the reconstruction of the composite moraine chronology, however, it has proved possible to extract a general, regional glacial response even where moraines are short, disturbed and, in some cases, missing. The regional response has been detected despite heterogeneous glacier morphology and the existence of local and regional climatic gradients. The fact that 25 individual dated moraines in Fig. 11A fall outside the moraine clusters identified in Fig. 11C can be attributed to these problems. The lichenometric techniques used in this study should prove useful in similar circumstances elsewhere. A combination of radiocarbon dating and Schmidt hammer R-values provides a 'multi-proxy' approach that enables the conclusions of lichenometry to be tested and the glacial chronology to be extended. Thus, although radiocarbon dates are less accurate than lichenometric dates over the Little Ice Age timescale, they provide stratigraphical insights into early and pre-Little Ice Age events that are not possible with the surface dating techniques. Although the Schmidt hammer is much less accurate as a chronological tool, it has the potential to differentiate Little Ice Age from earlier Holocene moraines. There is, however, no current evidence relating to mid-or early Holocene glacier expansion episodes in Breheimen. Such glacier advances have been widely reported elsewhere in southern Norway, mainly from distal glaciolacustrine or glaciofluvial depositional sequences (e.g. Nesje and Dahl, 1991b, 1994; Nesje and Kvamme, 1991; Nesje etal., 1991, 1994, 2000, 2001; Karlen and Matthews, 1 992; Matthews and Karlen, 1992; Dahl and Nesje, 1994, 1996; Matthews etal., 2000) and, in some cases, the existence of early Holocene moraines have been located beyond Little Ice Age glacier limits (see especially Dahl etal., 2002). In the present study, the Schmidt hammer results would have been the most likely to identify such relatively old moraines but provide little or no evidence for them. Conclusions 1 Seven low-altitude (temperate) glaciers in Breheimen are fronted by well developed Little Ice Age moraine sequences, each characterised by 2-13 but most commonly eight moraine ridges. 2 There is little or no evidence for pre-Little Ice Age moraines associated with these glacier forelands. Based on radiocarbon dating, Schmidt-hammer R-values and lichenometry, the Breheimen glaciers reached their Neoglacial maximum during the Little Ice Age. Copyright © 2003 John Wiley & Sons, Ltd. 411 3 Radiocarbon dating of thin (1 cm thick) samples of soils and peat, either buried beneath outermost moraine ridges or associated with adjacent mires and glaciofluvial sites affected by moraine formation, provide close maximum age estimates for the timing of the Neoglacial maximum. These estimates range from the fifteenth century to ad 1693 at Tverreggibreen and Storegrovbreen, but a pre-Little Ice Age maximum dating from ad 981 is 1399 is a possibility at Greinbreen. 4 Close estimates of the age of immature soils buried by glaciofluvial sediments suggest relatively minor glacier advances about ad 655-963 and ad 1277-1396, during the build-up of Tverreggibreen towards its Little Ice Age maximum. Several dates up to about 2000 cal. yr BP obtained from buried mature soils provide minimum estimates (but not close estimates) of the period of undisturbed soil development prior to burial. 5 Mean Schmidt hammer R-values from boulders on outermost moraine ridges are consistent with a late-Neoglacial glacier maximum and the absence of earlier Holocene moraines: some relatively low R-values causing greater variability being attributed to heterogeneous petrology and/or the incorporation of highly weathered boulders by glacier push mechanisms. 6 According to the lichenometric data, there are no pre-Little Ice Age moraines at these low-altitude glaciers, but the precise timing of the Little Ice Age maximum appears to have varied between glaciers, ranging from the late-seventeenth century to the early-nineteenth century. A conventional mid-eighteenth century date cannot be ruled out for at least three glaciers. 7 A lichenometry-based composite moraine chronology indicates broadly synchronous formation of recessional moraine ridges during the following time intervals: ad 1793-1799, 1807-1813, 1845-1852, 1859-1862, 1879-1885, 1897-1898, 1906-1908 and 1931-1933. These are attributed to short-term glacier advances during the long-term recession from the Little Ice Age glacier maximum. 8 The response of the temperate glaciers of Breheimen is indicative of their transitional position on the regional climatic gradient between Jostedalsbreen and Jotunheimen. Although the regional response is closer to that in Jotunheimen, complexity is introduced by local climatic gradients, aspect and morphological differences between glaciers. Regional differences in the timing of moraine formation are likely to reflect a response to differences in both summer temperature and winter precipitation. 9 Methodological implications of the study include: (i) the potential of a 'multi-proxy' approach in the context of Holocene moraine stratigraphy, where lichenometric dating, radiocarbon dating and Schmidt hammer R-values can each play a role; and (ii) the sensitivity of the composite moraine chronology to short-term regional glacier advances and hence climatic fluctuations based on Little Ice Age recessional moraines. Acknowledgements Field research was carried out on the University of Wales Swansea, Jotunheimen Research Expeditions 1998-2000 with the assistance of Alexandra Barwaldt, David Brett, Susanne Buhn and Kathrine Falch. We are also grateful to Dr P.Q. Dresser for carrying out the radiocarbon dating, and to Nicola Jones and Anna Ratcliffe for drawing the figures. Stefan Winkler is grateful for a personal grant from Deutsche Forschungsgemeinschaft (DFG), which enabled him to work on this project. This paper constitutes Jotunheimen Research Expeditions, Contribution No. 150. References Andersen JL, Sol I id J L. 1971. Glacial chronology and glacial geomorphology of the glaciers Midtdalsbreen and Nigardsbreen, south Norway. Norsk Geografisk Tidsskrift 25: 1 -38. Aune B. 1993. Temperatur normaler, normalperiode 1961-90. Rapport 02/93, Den norske meteorologiske institutt: Oslo. Benn Dl, Evans DJA. 1998. Glaciers and Glaciation. Arnold: London. Bickerton RJ, MatthewsJA. 1992. On the accuracy of lichenometric dates: an assessment based on the 'Little Ice Age' moraine sequence of Nigardsbreen, southern Norway. The Holocene 2: 227-237. Bickerton RJ, Matthews JA. 1993. 'Little Ice Age' variations of outlet glaciers from the Jostedalsbreen ice-cap, southern Norway: a regional lichenometric-dating study of ice-marginal moraine sequences and their climatic significance. Journal of Quaternary Science 8: 45-66. Bogen J, Wold B, Ostrem G. 1989. Historic glacier variations in Scandinavia. In Glacier Fluctuations and Climatic Change, OerlemansJ (ed.). Kluwer: Dordrecht; 109-128. Da hi SO, Nesje A. 1994. Holocene glacier fluctuations at Hardangerjokulen, central southern Norway: a high-resolution composite chronology from lacustrine and terrestrial deposits. The Holocene 4: 269-277. Dahl SO, Nesje A. 1996. A new approach to calculating Holocene winter precipitation by combining glacier equilibrium-line altitudes and pine-tree limits: a case study from Hardangerjokulen, central southern Norway. The Holocene 6: 381-398. Dahl SO, Nesje A, Lie 0, Fjordheim K, Matthews JA. 2002. Timing, equilibrium-line altitudes and climatic implications of two early-Holocene glacier readvances during the Erdalen Event at Jostedalsbreen, western Norway. The Holocene 12: 17-25. Dresser PQ. 2001. Bomb effect. In The Encyclopaedic Dictionary of Environmental Change, Matthews JA, Bridges EM, CaseldineCJ, Luckman AJ, Owen G, Perry AH, Shakesby RA, Walsh RPD, Whittaker RJ, Willis KJ (eds). Arnold: London; 60. Ellis S. 1979. The identification of some Norwegian mountain soil types. Norsk Geografisk Tidsskrift 33: 205-212. Ellis S. 1980. Soil-environmental relationships in the Okstindan Mountains, north Norway. Norsk Geografisk Tidsskrift 34: 1 67-1 76. Ellis S, Matthews JA. 1984. Pedogenic implications of a l4C-dated palaeopodzolic soil at Haugabreen, southern Norway. Arctic and Alpine Research 16: 77-91. Elven R. 1978. Subglacial plant remains from the Omnsbreen glacier area, South Norway. Boreas 7: 83-89. Erikstad L, Sol lid J L. 1986. Neoglaciation in South Norway using lichenometric methods. Norsk Geografisk Tidsskrift 40: 85-105. Faegri K. 1950. On the variations of western Norwegian glaciers during the last 200 years. Association Internationale d'Hydrologie Scientifique, Assemblee Generale d'Oslo 1948 2: 293-303. Forland EJ. 1993. Nedbornormaler normalperiode 1961-90. Rapport 39/93, Den norske meteorologiske institutt: Oslo. Grove JM. 1985. The timing of the Little Ice Age in Scandinavia. In The Climatic Scene, Tooley MJ, Sheail GM (eds). George Allen and Unwin: London; 132-153. Grove JM. 1988. The Little Ice Age. Methuen: London. Haines-Young RH. 1983. Size variation of Rhizocarpon on moraine slopes in southern Norway. Arctic and Alpine Research 15: 295-305. Haines-Young RH. 1985. Discussion of 'Size variation of Rhizocarpon on moraine slopes in southern Norway': a reply. Arctic and Alpine Research 17: 212-216. Hoel A, Werenskiold W. 1962. Glaciers and snowfields in Norway. Norsk Polarinstitutt Skrifter 126: 1-242. InnesJL. 1985a. Lichenometry. Progress in Physical Geography 9: 187-254. InnesJL. 1985b. A standard Rhizocarpon nomenclature for lichenometry. Boreas 14: 83-85. Karlen W, Matthews JA. 1992. Reconstructing Holocene glacier variations from glacial lake sediments: studies from Nordvestlandet and Jostedalsbreen-Jotunheimen, southern Norway. Geografiska Annaler 74A: 327-348. Kjollmoen B (ed.). 2000. Glasiologiske undersokelser i Norge 1999. Rapport 2/2000, Norges Vassdrags- og Energidirektorat: Oslo. Copyright © 2003 John Wiley & Sons, Ltd. 412 Lutro O, Tveten E. 1 996. Geologiske kart over Norge, bergrunnskart Ardal, Malestokk 1 -.250000. Norges Geologiske Undersokelse: Trondheim. Matthews JA. 1974. Families of lichenometric dating curves from the Storbreen gletschervorfeld, Jotunheimen, Norway. Norsk Geografisk Tidsskrift 28: 215-235. Matthews JA. 1 975. Experiments on the reproducibility and reliability of lichenometric dates, Storbreen gletschervorfeld, Jotunheimen, Norway. Norsk Geografisk Tidsskrift 29: 97-109. Matthews JA. 1977. A lichenometric test of the 1750 end-moraine hypothesis: Storbreen gletschervorfeld, southern Norway. Norsk Geografisk Tidsskrift 31: 129-1 36. Matthews JA. 1980. Some problems and implications of l4C dates from a podzol buried beneath an end moraine at Haugabreen, southern Norway. Geografiska Annaier 62A: 185-208. Matthews J A. 1981. Quantitative and Statistical Approaches to Geography: a Practical Manual. Pergamon Press: Oxford. Matthews JA. 1985. Radiocarbon dating of surface and buried soils: principles, problems and prospects. In Geomorphology and Soils, Richards KS, Ellis S, Arnett RR (eds). George Allen and Unwin: London; 269-288. Matthews JA. 1991. The late Neoglacial ('Little Ice Age') glacier maximum in southern Norway: new l4C-dating evidence and climatic implications. The Holocene 1:21 9-233. Matthews JA. 1993a. Radiocarbon dating of arctic-alpine palaeosols and the reconstruction of Holocene palaeoenvironmental change. In Climate Change and Human Impact on the Landscape, Chambers FM (ed.). Chapman and Hall: London; 84-96. Matthews J A. 1993b. Radiocarbon dating of buried soils with particular reference to Holocene solifluction. In Solifluction and Climatic Variation in the Holocene, Frenzel B, Matthews JA, Glaser B (eds). Palaoklimaforschung Bd 11, Gustav Fischer Verlag: Stuttgart; 309-324. Matthews JA. 1994. Lichenometric dating: a review with particular reference to 'Little Ice Age' moraines in southern Norway. In Dating in Exposed and Surface Contexts, Beck C (ed.). University of New Mexico Press: Albuquerque, NM; 185-212. Matthews JA, CaseldineCJ. 1987. Arctic-alpine Brown Soils as a source of palaeoenvironmental information: further l4C dating and palynological evidence from Vestre Memurubreen, Jotunheimen, Norway. Journal of Quaternary Science 2: 59-71. Matthews JA, Dresser PQ. 1983. Intensive l4C dating of a buried palaeosol horizon. Geologiske Foreningens i Stockholm Forhandlingar 105: 59-63. Matthews JA, Karlen W. 1992. Asynchronous neoglaciation and Holocene climatic change reconstructed from Norwegian glaciolacustrine sedimentary sequences. Geology 20: 991 -994. Matthews JA, Shakesby RA. 1984. The status of the 'Little Ice Age' in southern Norway: relative-age dating of Neoglacial moraines with Schmidt hammer and lichenometry. Boreas 13: 333-346. Matthews JA, Dahl SO, Nesje A, Berrisford MS, Andersson C. 2000. Holocene glacier variations in central Jotunheimen, southern Norway based on distal glaciolacustrine sediment cores. Quaternary Science Reviews 19: 1625-1647. McCarroll D. 1989a. Potential and limitations of the Schmidt hammer for relative-age dating: field tests on Neoglacial moraines, Jotunheimen, southern Norway. Arctic and Alpine Research 21: 268-275. McCarroll D. 1989b. Schmidt hammer relative-age evaluation of a possible pre-'Little Ice Age' Neoglacial moraine, Leirbreen, southern Norway. Norsk Geologisk Tidsskrift 69: 125-1 30. McCarroll D. 1991. The age and origin of Neoglacial moraines, Jotunheimen, southern Norway: new evidence from weathering-based data. Boreas 20: 283-295. McCarroll D. 1994. The Schmidt hammer as a measure of degree of rock surface weathering and terrain age. In Dating in Exposed and Surface Contexts, Beck C (ed.). University of New Mexico Press: Albuquerque, NM; 29-46. Nesje A. 1989. Glacier-front variations of outlet glaciers from Jostedalsbreen and climate in the Jostedalsbre region of western Copyright © 2003 John Wiley & Sons, Ltd. Norway in the period 1901-80. Norsk Geografisk Tidsskrift 43: 3-17. Nesje A, Dahl SO. 1991a. Late Holocene glacier fluctuations in Bevringsdalen, Jostedalsbreen region, western Norway. The Holocene 1: 1-7. Nesje A, Dahl SO. 1991b. Holocene glacier variations of Bläisen, Hardangerjokulen, central southern Norway. Quaternary Research 35: 25-40. Nesje A, Dahl SO. 1994. Late Glacial and Holocene glacier fluctuations and climatic variations in western Norway: a review. Quaternary Science Reviews 12: 255-261. Nesje A, Dahl SO. 2000. Glaciers and Environmental Change. Arnold: London. Nesje A, Kvamme M. 1991. Holocene glacier and climatic variations in western Norway: evidence for early Holocene glacier demise and multiple Neoglacial events. Geology 19: 610-612. Nesje A, Rye N. 1993. Late Holocene glacier activity at Sandskardfonna, Jostedalsbreen area, western Norway. Norsk Geografisk Tidsskrift 47: 21 -28. Nesje A, Dahl SO, Lovlie R, SulebakJR. 1994. Holocene glacier activity at the southwestern part of Hardangerjokulen, central southern Norway: evidence from lacustrine sediments. The Holocene 4: 377-382. Nesje A, Dahl SO, Lovlie R. 1995a. Late Holocene glaciers and avalanche activity in the Alfotbreen area, western Norway: evidence from a lacustrine sedimentary record. Norsk Geologisk Tidsskrift 75: 120-126. Nesje A, Kvamme M, Rye N, Lovlie R. 1991. Holocene glacier and climatic history of the Jostedalsbreen region, western Norway: evidence from lake sediments and terrestrial deposits. Quaternary Science Reviews 10: 87-114. Nesje A, Johannessen T, Birks HJB. 1995b. Briksdalsbreen, western Norway: climatic effects on the terminal response of a temperate glacier between ad 1 901 and 1 994. The Holocene 5: 343-347. Nesje A, Dahl SO, Andersson C, Matthews JA. 2000. The lacustrine sedimentary sequence in Sygneskardvatnet, western Norway: a continuous, high-resolution record of the Jostedalsbreen ice cap during the Holocene. Quaternary Science Reviews 19: 1047-1065. Nesje A, Matthews JA, Dahl SO, Berrisford MS, Andersson C. 2001. Holocene glacier fluctuations of Flatebreen and winter-precipitation changes in the Jostedalsbreen region, western Norway, based on glaciolacustrine sediment records. The Holocene 11: 267-280. Ostrem G, Liestol O, Wold B. 1976. Glaciological investigations at Nigardsbreen, Norway. Norsk Geografisk Tidsskrift 30: 187-209. Ostrem G, Selvig KD, Tandberg K. 1 988. Atlas over breer i Sor-Norge. Norges Vassdrags- og Energiverk Vassdragsdirektoratet: Oslo. Paterson WSB. 1994. The Physics of Glaciers, 3rd edn. Pergamon: Oxford. PoeltJ. 1988. Rhizocarpon Ram. em. Th. Fr. subgen. Rhizocarpon in Europe. Arctic and Alpine Research 20: 292-298. RekstadJ. 1902. lagttagelser fra braeer i Sogn og Nordfjord. Norges Geologiske Indersogelses Aarbog 1902 3: 1 -48. Sigmond EMO, Gustavson M, Roberts D. 1984. Bergrunnskart over Norge. Malestokk 1: 1 million. Norges Geologiske Undersokelse: Trondheim. Stuiver M, Reimer PJ. 1993. Extended l4Cdata base and revised CALIB 3.0 l4C age calibration program. Radiocarbon 35: 21 5-230. Stuiver M, Reimer PJ, Bard E, BeckJW, Burr GS, Hughen KA, Kramer B, McCormac FG, Van der plicht J, Spurk M. 1 998. INTCAL 98: radiocarbon-age calibration 24,000-0 cal BP. Radiocarbon 40: 1041-1083. Winkler S. 1996a. Frührezente und rezente Gletscherstandsschwankungen in Ostalpen und West- /Zentralnorwegen. Trierer Geographische Studien 15, Geographische Gesellschaft Trier, Trier. Winkler S. 1996b. Front variations from outlet glaciers of Jostedalsbreen, western Norway, during the 20th century. Norges Geologiske Undersokelse Bulletin 431: 33-47. Winkler S. 2000. Der "Schmidt-Hammer" als geochronologische Methode - Anwendungsmöglichkeiten und Problematik aufgezeigt an Beispielen aus Neuseeland und Norwegen. Trierer Geographische Studien 23: 123-146. Winkler S, Nesje A. 2000. Gletscher und Klima in Norwegen während der letzten 10.000 Jahre. Oldenburger Geoökologische Arbeiten 4: 189-228. J. Quaternary Sei., Vol. 18(5) 395-413 (2003) 413 Winkler S, Shakesby RA. 1995. Anwendung von Lichenometrie und Schmidt-Hammer zur relativen Altersdatierung prä-frührezenter Moränen, am Beispiel der Vorfelder von Guslar-, Mitterkar-, Rofenkar- und Vernagtferner, Otztaler Alpen, Osterreich. Petermanns Geographische Mitteilungen 139: 283-304. Winkler S, Haakensen N, Nesje A, Rye N. 1997. Glaziale Dynamik in Westnorwegen—Ablauf und Ursachen des aktuellen Gletschervorstoßes am Jostedalsbreen. Petermanns Geographische Mitteilungen 141: 43-63. Worsley P. 1990. Lichenometry. In Geomorphological Techniques, 2nd edn, Goudie A (ed.). Unwin Hyman: London; 422-428. Copyright © 2003 John Wiley & Sons, Ltd.