Greenland

The best paleoclimatic information exists for Greenland, where during the last 30-40 years quite a large number of ice-cores have been drilled, beginning with the oldest one (Camp Century) and ending with the series of 13 drilled during the summers of 1993-1995 along the North-Greenland-Traverse. The routine analyses of ice cores include measurements of iso-topic composition (51B0, 8D), content of greenhouse gases (CO, and CIi4), dust content, chemical composition, electricity conductivity, annual ice accumulation, etc. The advantage of this kind of proxy data is mainly the high time resolution, which allows researchers to investigate even seasonal changes and the wide spectrum of information available about climate and environmental changes. However, in recent years some scientists have expressed scepticism as to whether analyses of ice cores may be considered reliable because of the artificial contamination and disturbances of ice cores during the drilling (for details see Jaworowski et at. 1990, 1992), More recently a new source of information about past temperatures in Greenland has become available. This new source is based on temperature profiles measured down through an ice sheet in deep boreholes. This information is then used to reconstruct past surface temperatures. This is possible because temperatures down through the ice depend on the geothermal heat flow density, the ice-flow pattern, and the past surface temperatures and accumulation rates (Dahl-Jensen et al. 1998). Since the beginning of 1970s, this method has very often been used for the non-glaciated areas based on temperature measurements in wells (see e.g. Cermak 1971; Lachenbruch and Marshall 1986; Pollack and Chapman 1993; Majorowicz et al. 1999, and references therein). The main weakness of this method is its low time resolution, which decreases the further back in time one investigates. This means that the high-frequency changcs are not registered. In the distant past even such prominent climatic events like Bolling/Allerod and cold Younger Dryas periods are not resolved (see Figure 10.4).

Holocene climate history for the central part of Greenland is presented in Table 10.1 and Figures 10.2-10.4. Generally speaking, there is good correspondence in the timing of the occurrence of warm and cold periods on a millennial time scale, distinguished from different sources. Some differences are connected with different time resolutions of the sources mentioned, different sensitivity to environmental and climate-forming factors, and probably errors in the dating of some of them. The first warm period (not noted by the borehole temperatures) occurs in the early Holocene about 10.0 to 8.56 ka BP, The greatest and longest wanning is clearly evident in reconstructed temperatures from GRIP borehole, lasting from 8 to 5-4 ka BP and this can be referred to as the Climatic Optimum (Dahl-Jensen et al. 1998). However, more high-time resolution data show thai during this time at least one to three cold periods also occurred (see Table 10.1 and Figures 10.2 and 10.3). Dahl-Jensen etal. (1998) calculated that the surface temperature during this period was about 2.5°C warmer than the present temperature (Figure 10.4b). A period of maximum postglacial warmth between approximately 8 and 4 ka BP has been proposed from the Camp Century ice core (Dansgaard et al. 1971).

The next warm period had a rather short duration and began about 2.5 ka and ended between 2-1.5 ka BP. For this time there is the greatest difference in data received from the borehole temperature measurements (see Table 10.1 and Figures 10.2-10.4), which show clear cooling (0.5°C below present temperature) about 2 ka BP. Analysis of reconstructed temperature for the Dye 3 core (Dahl-Jensen et al. 1998) also provides similar results. All paleoclimatic proxy data reveal, on the other hand, the existence of a Medieval Warm Period, which in Greenland, as may be seen from Table 10.1 and Figures 10.2 10.4, started about 1.4-1.5 ka BP, thus significantly earlier than in Europe. According to reconstructed temperatures by Dahl-Jensen et al. (1998), the maximum warming occurred about 900 A.D. (Figure 10.4c) and was l°C warmer than at present in Greenland.

Table 10.1. Warm and cold periods (ka BP) in the Holocene based on measurements of oxygen isotope S^O, an ice accumulation rate and chemical fluxes at GISP2 core as well as based on borehole temperanires at GRIP core

^—^Measured element

<S'"0>*

Ice accumulation

Chemical

Borehole temperature

Periods

rate*

fluxes**

at GRIP core***

9.5-8.5

9.2-8.5

10.6-9.3

8.0-7.6

8.1-7.3

7.9-6.3

8.0-1.2

7.0-6.6

Warm

5.3-4.7 3.6-3.1

5.0-4.2

2.5-2.0

2.5-1.9

2.7-1.5

1.0-0.8

1.3-0.8

0.96-0.6!

1.5-0.8 0.05-0.0

> 11.3

11.6-9.5

8.5-8.0

8.5-8.0

8.8-7.8

7.5-7,0

Cold

6.5-6.0 4.7-4.3

6.0-5.2

6.1-5.0

3.0-2.5

3.1-2.4

3.0-1.5

1.9-1.1

1.9-1.3

0,8 0.0

0.8-0.0

0.61-0.0

0,7-0.1

Author interpretation based on figures presented at the following papers: * - Meese et al. (1994); ** - O'Brien et al. (1995); *** - Dahl-Jensen et al. (1998)

Author interpretation based on figures presented at the following papers: * - Meese et al. (1994); ** - O'Brien et al. (1995); *** - Dahl-Jensen et al. (1998)

200 400 600 300 1000 1200 1400 1600 1800 Years A.D.

Figure 10.4. The contour plots of all the GRIP temperature histograms as a function of time describes the reconstructed temperature history (solid lines) and its uncertainty. The temperature history is the history at the present elevation (3240 m) of the summit of the Greenland Ice Sheet. The dashed curves are the standard deviations of the reconstruction. The present temperature is shown as a horizontal line. (A) - the last lOOky BP, (B) - the last lOky BP and (C)- the last 2000 years. Reprinted with the permission from Dahl-Jensen D., Mosegaard K... Gundestrup N., Clow C. D„ Johnscn S. J., Hansen A. W. and Balling N., 'Past temperatures directly from the Greenland Ice Sheet', Science, 282, 268-271. Copyright 1998 American Association for the Advancement of Science.

200 400 600 300 1000 1200 1400 1600 1800 Years A.D.

Figure 10.4. The contour plots of all the GRIP temperature histograms as a function of time describes the reconstructed temperature history (solid lines) and its uncertainty. The temperature history is the history at the present elevation (3240 m) of the summit of the Greenland Ice Sheet. The dashed curves are the standard deviations of the reconstruction. The present temperature is shown as a horizontal line. (A) - the last lOOky BP, (B) - the last lOky BP and (C)- the last 2000 years. Reprinted with the permission from Dahl-Jensen D., Mosegaard K... Gundestrup N., Clow C. D„ Johnscn S. J., Hansen A. W. and Balling N., 'Past temperatures directly from the Greenland Ice Sheet', Science, 282, 268-271. Copyright 1998 American Association for the Advancement of Science.

From the start of the Holocene to 1 ka BP, between 4 and 6 cold periods can be distinguished. The first one occurred in the transition period from Younger Dryas to the Holocene, when the temperatures were generally colder than average conditions in the Holocene. The second deterioration of climate occurred from 8.5 ka to 8.0 ka BP, i.e. just before the start of the Climatic Optimum. Between one and three short cold events occurred during the time of Optimum. Both the ice accumulation rate and changes in chemical fluxes show that this cooling occurred from 6 to 5 ka BP (Table 10.1 and Figures 10.2 and 10.3). On the other hand, the oxygen isotopes reveal three colder periods observed in three other periods (7.5-7.0 ka, 6.5-6.0 ka, and 4.7-4.3 ka BP). Oxygen isotope and borehole temperantre data show the clear coolness of climate from 3 ka to about 1.5-1.1 ka BP with a very small warming spell between 2.5-2.0 ka BP. The remaining sources reveal not so long-lasting cooling (see Table 10.1 and Figures 10.2-10.4). The maximum cooling during this period occurred around 2 ka BP and was about 0.5°C lower than the present climate (see Figure 10.4b).

O'Brien et ai (1995) found that the cold events identified in their glaciochemical series correspond in timing to records of the world-wide Holocene glacier advances (Denton and Karlen 1973) and to cold events in paleoclimate records from Europe, North America, and the Southern Hemisphere (Harvey 1980), as determined by combining glacier advance, oxygen isotope (5I80), pollen count, tree ring width, and ice core data (Figure 10.5). They also reveal quite a good correspondence between the timing of cold periods and periods of low solar output, as identified in residual tree ring radiocarbon (8I4C) age measurements (Stuiver and Braziunas 1989) (Figure 10.3). Moreover, they also found almost the same quasi-cycles of 8I4C climate (2500 years) and cold periods identified in the G1SP2 record (-2600 years).

Years BP

0 2000 4000 6000 8000 10000

-UC years BP

Figure 10.5. Paleoclimate cold events: GISP2 Holocene EOF!; world-wide glacial expansions and their relative magnitude (Denton and Karlen 1973); synthesis of various climate proxy records from Europe, Greenland, North America, and the Southern Hemisphere showing cold periods (Harvey 1980); the Cockburn Stade (Andrews and Ives 1972); and the YD event (Mayewski etal. 1993). Reprinted with permission from O'Brien S. R„ Mayewski P. A., Meeker L. D., Meese D. A., Twickler M. S. and Whitlow S. I , 'Complexity of Holocene climate as reconstructed from a Greenland ice core', Science, 270, 1962 -1964. Copyright 1995 American Association for the Advancement of Science.

From the proxy data presented here, the precipitation changes during the Holocene are best represented by the ice accumulation rate. Meese et al.

(1994) found, however, that the accumulation and oxygen isotopes correlate significantly at GISP2. From Figure 10.2 one can see that this correlation is positive. Mostly the precipitation is greater in warmer periods and lower in colder periods. There arc, however, some exceptions, such as about 6.8 ka or 1.2-1.0 ka BP.

Dahl-Jcnsen et al. (1998) comparing the results presented here with those from the Dye 3 borehole (865 km further south from GRIP), found that the Dye 3 temperature is similar to the GRIP history, but has an amplitude 1.5 times greater, indicating higher climatic variability there. They concluded that the difference in amplitudes observed between the two sites is a result of their different geographic location in relation to the variability of atmospheric circulation, even on the time scale of a millennium. The importance of regional influences on environmental changes, especially in the second half of the Holocene, is also revealed by O'Brien et al. (1995). They concluded that this complexity in Holocene climate makes distinguishing a natural from an anthropogenically-altcrcd climate a formidable task.

10.1.2 Canadian High Arctic

Proxy data concerning Holocene climatic change in the Canadian high Arctic largely comprises ice core, glacial, and sea-ice/ice-shclf components and less geomorphological and chronological evidence (Evans and England 1992). Ice-core analyses from the Agassis, Mcighen, and Devon Ice Caps (e.g., Koerner and Paterson 1974; Koerner 1977a, b, 1979, 1992; Paterson etal. 1977; Fisher and Koerner 1980, 1983; Koerner and Fisher 1985; Koerner et al. 1990) provide an important record of high latitude climatic change. Fisher and Koerner (1980) found for Devon Island a period of increasing postglacial warmth from 10 to 8.3 ka BP. Then the temperature showed small oscillations until 4.3 ka BP, when the maximum postglacial temperatures occurred. Since 4.3 ka BP there has been a progressive cooling. The end of the postglacial optimum occurred between 4.5 and 3 ka BP. This scheme of the climatic changes on Devon Island is in good correspondence with the reconstructed temperature history from the GRIP borehole (Figure 10.4b). Variations in the abundance of stranded driftwood, which are also used as indicator of Holocene climate change in the high Arctic, generally correlate very well with both the above-mentioned series of data (Figure 10.6). The interpretation of this Figure is as follows: the greater the abundance, the wanner the summer temperatures and the lighter the sea-ice conditions, which allow for drifting of wood. Bradley (1990) summarising the proxy data for the Holocene paleoclimate of the Queen Elizabeth Islands, identified two basic climatic periods: 1) the early-mid Holocene when summer temperatures were comparable or higher than at present, and 2) the last 3500 ± 500 years over which summer temperature dropped significantly. Evans and England (1992), analysing proxy data from northern Ellesmere Island, generally give a similar reconstruction.

This paper

Stewart and England 1963

This paper

Stewart and England 1963

Years BP * 10J

Figure 10.6. Histogram of driftwood radiocarbon dates from the Canadian and Greenland high Arctic based on data from Stewart and England (1983). Evans (1988), Lemmen (1988) and Blake (1987) (after Evans and England 1992). I, 2 and 3 in the upper part of the figure denote the periods of significant differences in the abundance of stranded driftwood.

Years BP * 10J

Figure 10.6. Histogram of driftwood radiocarbon dates from the Canadian and Greenland high Arctic based on data from Stewart and England (1983). Evans (1988), Lemmen (1988) and Blake (1987) (after Evans and England 1992). I, 2 and 3 in the upper part of the figure denote the periods of significant differences in the abundance of stranded driftwood.

10.t.3 Eurasian Arctic Islands

The Greenland Ice Sheet and the Canadian high Arctic represent the typical continental climate, while on the other hand, the Eurasian Arctic islands (from Svalbard to Sevemaya Zemlya), and particularly Svalbard, characterise the part of the Arctic with the most maritime climate. For this part of the Arctic there exist some ice-core analyses from Svalbard, Zemlya Frantsa Josifa, and Sevcrnaya Zemlya, but most of them do not cover the last thousand years (Tarussov 1988, 1992; Vaikmae 1990). Only the ice core from the Vavilov ice dome (Severnaya Zemlya) supplies proxy data for almost the entire Holocene period (Figure 10.7). Unfortunately, there are some uncertain ties in the dating of the ice core; the ice age is evidently overestimated, even in the upper part of the profile (Tarussov 1992). Tarussov further notes that the interpretation of this core is problematic due to the strange absence of a correlation between the chloride and 5IS0 curves of the same core. Based on the review of literature concerning the history of glacier advances and retreats during the Holocene in Svalbard, Novaya Zemlya, Zemlya Frantsa Josifa and Scvernaya Zemlya, estimated using geomorphological and glacier investigations (e.g., Bazhcv and Bazheva 1968; Szupryczyriski 1968; Grossvald 1973; Baranowski 1977b; Werner 1993; Lubinski et al 1999), and for the last thousand years also using ice-core analyses (Vaikmae and Punning 1982; Tanissov 1988, 1992; Vaikmae 1990), one can conclude that there exists a good correspondence between climatic changes in this region of the Arctic. Therefore, the history of the Holocene climate will be presented here using mostly data from Svalbard, for which this history is best known.

Figure in. 7. Variations in 8lsO and CI concentrations for the Severnaya Zemlya ice core (after Vaikmae 1990).

Recently, Werner (1993) has provided a review of our current knowledge concerning the climatic changes in Svalbard in the Holocene and has also supplied a new Holocene moraine chronology for central and northern Spitsbergen. He presents evidence for multiple Neoglacial advances in this area. The fragmentary moraine record indicates two Little Ice Age advances and two older Neoglacial advances. The oldest moraines had stabilised by ca, 1.5 ka BP, and a second group of moraines by ca. 1.0 ka BP. The first group of moraines, according to Werner (1993) may correspond to the advance of glaciers between 3.5 and 2.0 ka BP, reported by many authors (e.g. Szupryczynski 1968; Baranowski 1975, 1977a, b; Baranowski and Karlen 1976; Punning el al. 1976; Lindner et al. 1982; Nicwiarowski 1982; Marks 1983). The advance of glaciers indicated by moraines dated to ca. 1.0 ka BP is not recognised in the stratigraphy of southern Spitsbergen. The moraine chronology proposed by Werner (1993) compares well with other proxy climate records on Spitsbergen, summarised in his Figure 10. From this, it can be seen that Climatic Optimum occurred between 7 and 4 ka BP, the same as in Greenland and in the Canadian Arctic. During this period the reduced sea ice (Haggblom 1982), increased the production of local pollen (Hyvarinen 1972) and the occurrence of thermophitus marine molluscs (Feyling-Hanssen and Olsson 1960) were observed. There are also no traces indicating an advance of the glaciers. The proxy climatic records further show evidence for late-Holocene (4-2 ka BP) climatic deterioration. In the next 1000 years, there is evidence mainly in sea ice and glacial records of some wanning of the climate, similar to the reconstructed temperature histories for the GRIP borehole (Figure 10.4b). More recently, Svendsen and Mangemd (1997) have obtained a generally similar history of the Holocene climate based on investigations of sediment cores from the proglacia! lake Linnevatnet, west Spitsbergen.

Summarising all the proxy data from the Arctic presented in this section we can say that:

1. The Holocene climate until 1 ka BP was wanner than today, except during the early part and the period about 2 ka BP.

2. A Climatic Optimum occuned between 8 and 5-4 ka BP with temperatures being 2-2.5°C higher than present,

3. A drop in temperature was noted between 4 and 2 ka BP (minimum) and then an amelioration of climate was observed with a temperature maximum of about 900-1000 A.D.,

4. A significant similarity of climatic changes was noted in the entire Arctic analysed as well as in the areas bordering the Norwegian and Greenland seas (Iceland, Jan Mayen, and Scandinavia) (Werner 1993). Therefore we can probably state that the remaining part of the Arctic (not presented here) also had a similar climate history during the Holocene.

Generally speaking in the history of the climate of the last I ka years, three periods have most often been distinguished: the Medieval Warm Period (MWP), the Little Ice Age (LIA) and the Contemporary Global Wanning (CGW). The latter period will be described in the next section. Thus, what do the proxy data tell us about the climate in the Arctic during the first two periods? The most detailed answers are given by ice-core analyses.

10.2.1 Greenland

The best ice-core analyses from the whole Arctic are available for the Greenland Ice Sheet. In the first half of the 1990s, the two longest ice corcs (GRIP and GISP2) were drilled in the Summit (central part of Greenland) as well as 13 shallow ones (covering the last 500-1000 years) along the North-Greenland-Traverse. In addition, as was mentioned in the previous section, measurements of the borehole temperatures allow the reconstruction of the surface temperature histories for GRIP and Dye 3 areas. Let us start with the analysis from the proxy data giving the most averaged history. The borehole temperatures confirm the MWP and the LIA as having existed in Greenland, As was mentioned in the previous section, the MWP occurred here earlier than in other parts of the world (this fact allows Vikings to have built settlements in the southern part of Greenland). Maximum warmth is centred between 900 and 1000 A.D., but the beginning of this period can be dated between 500 and 600 A.D. and the end about 1200 A.D. (Figure 10.4c). From this Figure it can be seen that temperatures at this time were about 1°C greater than they are at present in Greenland. This period with temperatures higher than normal in Greenland lasted from about 200 A.D. to 1300 A.D. The MWP is also clearly seen in the ice accumulation rate data (Figure 10.2). The duration is the same as is shown by borehole temperatures but the greatest maximum of ice accumulation occurred around 800 A.D. On the other hand, the secondary maximum of accumulation corresponds very well with the maximum temperature from about 900-1000 A.D. The average accumulation from A.D. 620 to 1150 was 0.26 m of ice per year, 8% higher than the average Holocene accumulation rate and the highest rate recorded in the last 12 ka years (Figure 10.8 and Meese et at. 1994). In coastal Greenland, the MWP began as early as 800 A.D. (Lamb 1977). Proxy data available to Lamb were, of course, not as precise as those presented here. For this reason, and because all of Greenland reacts equally to factors determining climatic changes at present (see Przybylak 1996a, 2000a), it seems that the start of the MWP in coastal parts should be shifted to about 600 A.D. The historical records, mainly from northwestern Europe, describe an MWP occurring anywhere between A.D. 800 and A.D. 1300 (Lamb 1977; Houghton et at. 1990, 1996) with dates varying by as much as 200 years. This means that in Europe the MWP started about 200 years later than is indicated by the G1SP2 record.

5O0 750 1000 1250 1500 1750 2000

Years A.D.

5O0 750 1000 1250 1500 1750 2000

Years A.D.

Figure 10.8. The 100-year smoothed accumulation record from the GISP2 core for the period A.D. 500 to the present. The arrows show locations of visually identified melt layers in the ice core. Reprinted with the permission from Meese D. A., Gow A. J.. Grooles P.. Mayewski P. A., Ram M„ Stuiver M., Taylor K. C„ Waddington E. D. and Zielinski G. A., 'The accumulation record from the G1SP2 core as an indicator of climate change throughout the Holocenc', Science, 266, 1680-1682. Copyright 1994 American Association for the Advancement of Science.

The LIA is not as well defined as the MWP in the literature. The first views, based on rather low-resolution proxy data, assumed that the LIA was one long, sustained cold period with dates ranging from A.D. 1200 to 1800 or A.D. 1350 to 1900 (for details see e.g. Lamb 1977, 1984; Starkel 1984; Grove 1988). The new, high-resolution, data reveal that this opinion was wrong, and that the climate during this period underwent significant fluctuations from cold to warm and warm to cold conditions. However, through most of the period a cold climate occurred. The climatic changes in the LIA period in Greenland is shown in Figures 10.4 and 10.8-10.10. The reconstructed temperature for GRIP and the mean isotope record from northern Greenland (Figure 10.4 and Figure 10.9) clearly show the existence of an LIA in Greenland. According to borehole temperatures, the LIA lasted from about A.D. 1400 to 1900. Throughout this period, except for a few decades around 1700 A.D., the temperature was colder than at present in Greenland. Dahl-Jensen et al. (1998) distinguished two cold periods centred at 1550 and 1850 A.D, with

1500 1600 1700 1800 1900 2000

Figure 10.9. a) stacked isotope record of core BI8, B2I and B29 for time span 1480-1969: Thin line represents the average of the triannual data sets, thick line and shading using dots the mean and standard deviation of the spline approximations after subtracting the core averages, b) stacked record of Na+ concentrations in core BI6, B18 and B21. To allow for different absolute sea salt level in each core, which are largely caused by the different altitude of the drill sites, Na+ concentrations were normalised to the individual core average, c) three years intervals of reconstructed solar irradiance for the time span 1612 to 1913 (Lean et al. 1995), d) SO^ concentration above background (thin line) and frequency in a 30 year interval (dotted bars) of stratosphcrically derived volcano horizons in the annual record of core B21 (after Fischer et al. 1998).

temperatures 0.5 and 0.7°C below the present, respectively, The mean isotope rccord from northern Greenland (Figure 10.9) shows that in this part of Greenland the LIA ended earlier, at about 1850 A.D. Based on an oxygen isotope record from Camp Century (Figure 1 in Johnsen et al. 1970), which lies in the same part of Greenland but has a longer record, we can assume that in the northern Greenland, similar to central Greenland, the start of the LIA occurred about 1400 A.D. In this record the two minima are also quite evident, but the times of their occurrence are different. The first minimum was centred around 1680 A.D. and the second around 1820-1830 A.D. The temperature during these periods was estimated to be I°C lower than present (Figure 10.9).

1500 1600 1700 1800 1900 2000

Figure 10.9. a) stacked isotope record of core BI8, B2I and B29 for time span 1480-1969: Thin line represents the average of the triannual data sets, thick line and shading using dots the mean and standard deviation of the spline approximations after subtracting the core averages, b) stacked record of Na+ concentrations in core BI6, B18 and B21. To allow for different absolute sea salt level in each core, which are largely caused by the different altitude of the drill sites, Na+ concentrations were normalised to the individual core average, c) three years intervals of reconstructed solar irradiance for the time span 1612 to 1913 (Lean et al. 1995), d) SO^ concentration above background (thin line) and frequency in a 30 year interval (dotted bars) of stratosphcrically derived volcano horizons in the annual record of core B21 (after Fischer et al. 1998).

The break in the prolonged cooling, observed particularly from about A.D. 1600 to 1850, occurred in the second half of the 18lh century. The greatest peculiarity of this record, not observed in other records (see Figures 10.4, 10.8 and 10.10), is the fact that the highest temperatures during the whole period recorded occurred at the end of the 19th century. The ice accumulation record from GISP2 (Figures 10.8 and 10.10) does not show so clearly the existence of the LIA in Greenland. However, this kind of data has a lower reliability than the two previous kinds. This is probably due to the fact that in the short-term scale (LIA) the positive correlation found by Meese et al. (1994) between precipitation (accumulation) and temperature (5!*0) is significantly less than in the long-term scale (the Holocene). Przybylak (1996a), working on the basis of the instrumental observations, found that in the warmer and colder periods both above and below normal precipitation can occur in the Arctic. From this account, the mean accumulation over the last 800 years was slightly higher than normal (about 3%), but there was significant variability on decadal and century time scalcs (Figure 10.10). Surprisingly, however, the accumulation variations agree quite well with surface temperatures reconstructed for the GRIP borehole. The greatest discrepancies are the facts that: I) the first accumulation minimum occurred about 100 years earlier than the minimum temperature and 2) the accumulation after 1850 A.D. does not show any rise.

I 026

0.23 022

1650 1700 1750 1 800 1850 1900 1950 2000

Years AD

Figure 10.10. The 25-year smoothed accumulation record from the GISP2 core from A.D. 1650 to the present. The dates above the arrows correspond to years of decreased accumulation thai correlate with dated glacial advances or cold periods in Greenland and elsewhere (Grove 1988). Reprinted with the permission from Meese D. A., Gow A. J., Grootes P., Mayewski P. A.. Ram M.. Stuiver M„ Taylor K. C., Waddington E. D. and Zielinski G. A., 'The accumulation record from the G1SP2 core as an indicator of climate change throughout the Holocene', Science. 266, 1680-1682. Copyright 1994 American Association for the Advancement of Science.

1650 1700 1750 1 800 1850 1900 1950 2000

Years AD

Figure 10.10. The 25-year smoothed accumulation record from the GISP2 core from A.D. 1650 to the present. The dates above the arrows correspond to years of decreased accumulation thai correlate with dated glacial advances or cold periods in Greenland and elsewhere (Grove 1988). Reprinted with the permission from Meese D. A., Gow A. J., Grootes P., Mayewski P. A.. Ram M.. Stuiver M„ Taylor K. C., Waddington E. D. and Zielinski G. A., 'The accumulation record from the G1SP2 core as an indicator of climate change throughout the Holocene', Science. 266, 1680-1682. Copyright 1994 American Association for the Advancement of Science.

Was this article helpful?

0 0

Post a comment