Climates of Hunger: Mankind and the World's Changing Weather

4. One Thousand Years in Iceland

One Thousand Yt~ars in Iceland CHAPTER 4 AsTHE FWW of westedies around the North Pole expanded in the twelfth and thirteenth centuries A.D., bringing drought to the Mill Creek people, and damp winters in western Europe, what happened in other parts of the Northern Hemisphere? What happened, for instance, in the North Atlantic-where ice moving southward from the Arctic meets thf: Gulf Stream? An expansion of the westerlies-in effect, an expansion of the Arctic-should bring sea ice farther south. Conditions should have changed measurably in the border region between arctic and Gulf Stream waters. This is an area as sensitive to climatic change as the Iowa ecotone between tall prairie and steppe. The North Atlantic, Greenland, Iceland, would be another plac€: to take the measure of climatic change, to see if we understand its mechanisms and dimensions . The object of our research on Mycenae and Mill Creek was not simply to solve ancient mysteries. We wantj!d to know how quickly 47 Figure 4.1. The North Atlanf Ie. oe/ c';<./ _-~r'-t~ICELAND , Reykjavik North Atlantic Ocean , One Thouslmd Years in Iceland 49 climate could change, and by how much. And we wanted to know the causes of climatic change. Iceland, about the size of Indiana, lies just south of the Arctic Circle. Fish and agriculture, mostly herding and wool production, are its staples. Despite its high latitude, the Gulf Stream makes possible the growing of potatoes, turnips, hay, and in certain times barley and other grains. About 750 miles to the east lies Norway. Greenland is a like distance to the west (see figure 4.1). Iceland is of particular interest to us because it has a recorded history of more than 1,000 years, chronicles not so much of the doings of armies and diplomats as of the lives of individual Icelanders and families. The first sagas "Saga" has come to mean any long narrative, but the original sagas were Icelandic prose narratives of the twelfth and thirteenth centuries. Today these are a source of information about climate. In references to those times, the terms "Viking" and "Norseman" are often interchanged. If original use is followed, "Norse" applied to any ancient ScandinaVian-anyone from Norway, Sweden, or Denmark , or people of Scandinavian stock who settled in Iceland or Greenland. "Viking" had a narrower meaning. The Vikings were the pirates, the raiders and plunderers of the whole northern European and North Atlantic area for about three centuries beginning in the eighth century A.D. Early in this period the Vikings used longboats, graceful and useful in coastal waters, but not built for the open seas. Later they adopted the Norse knorr, stubby in appearance but extremely seaworthy. Although a widespread belief exists that a knorr could sail only with the wind, these ships, in fact, did very well both ways. The North Atlantic lands, including Iceland and Greenland, were settled by a mixture of Irish, Norse, and other Europeans trying to escape the Vikings, and by the Vikings themselves. In 1914, Otto Pettersson published a paper on "Climatic variations in historic and prehistoric time." Pettersson drew on the sagas and on recorded indications of climate from the time of the Vikings until his day. 50 II Our Climates since A.D. 900 Pettersson reported: The earliest information we possess regarding the climate of Iceland is derived from the record of the monk Dicuil of Ireland in 825. He describes a visit some 30 years earlier by some Irish ecclesiastics to the Island of Thyle (Iceland). At that time, about a century before its colonisation by the Norsemen, Iceland was visited and inhabited by the Irish. The sagas call them "Papar," which indicates that they were monks or hermits.... Dicuil narrates the description of the island given by his fellow monks, who had been there from February to August.... [Po 7] Dicuil's report, as Pettersson quotes it, was that Iceland was not surrounded by ice, even in the severe months of winter and early spring, but that "after a day's journey to the north they found a frozen sea." As Scandinavians began to occupy Iceland in the late ninth century, the climate remained favorable. Pettersson states that the sagas "nowhere mention that driftice hindered the norsemen in their [early] journeys to and from the island" (p. 7). This ice-free time lasted hundreds of years, Pettersson says: Although weather conditions are often mentioned in the older annals and sagas I cannot find that the annual ice-drift to the shores of Iceland is spoken of before the 13th century. In the 13th century Iceland began to get blocked by driftice from Greenland. The blockade was much more severe then than now [1914] , although even now the northcoast is frequently blocked and sometimes, though not often, the east and southeast coasts. Owing to the influence of the Irminger current, the westcoast in our time is nearly always free from ice.... The ice conditions in Greenland are intimately connected with those in Iceland. The advance of ice out of Nordbotn in the 12-13th centuries, of which Bardsson speaks, proved fatal to the old Norse colonies in Greenland because it cut off communication with their mother-country. [Po 8] So the twelfth and thirteenth centuries were times ofcatastrophe not only for the Mill Creek people, but for cultures thousands of miles away. The Arctic did expand, bringing increased westerlies and dry western air to Mill Creek, and bringing icy death to Greenland. Otto Pettersson's contribution to our understanding of climates, past, present, and future has generally gone unnoticed, in part because it happened to be written at the wrong time. The climate One Thousand Years in Iceland 51 was especially favorable in the decades following 1914, in Iceland, Greenland, Sweden, and most of the hemisphere. Climatically, times were good; who cared if Eric the Red's colony froze back in the fifteenth century? Pettersson's work, valuable as it is, wa:s only a beginning. The ancient records held much more than he knew, and it remained for others to write a more detailed history of climate in the North Atlantic. The others included Thorvaldur Thoroddsen, a contemporary of Pettersson, and Sigurdur Thorarinsson, a later scientist. The work of these men has been compiled and added to by PaIl Bergth6rsson (1962, 1969) of the Icelandic Meteorological Office. Tropical waters and ice Iceland's strategic location in the meeting place of arctic waters and the Gulf Stream makes its I,OOO-year record one of special interest-and in fact .makes such a record possible. It is easy to get a 1000year record of temperatures in Iceland; in fact, thermometer records there go back to 1846. But to go back further, Bergth6rsson and others depend on reports of drift ice floating by the island. Drift ice is carried from the arctic ice pack and the waters north of Iceland by ocean currents. In cold times arctic waters carry the ice southward. In warmer times, the warm Gulf Stream dominates the Iceland area, keeping drift ice away. Iceland and Greenland are far enough north to observe this ice and sometimes to be blockaded by it; yet they are far enough south so that in historic times they have not been surrounded by solid ice. And in the warmer spells of historic times, little or no ice floats by them. The drift ice is easy to see, both from shore and from ships. From early days, down to the sinking of the Titanic, and since then, it has imperiled ships and affected commerce, and Icelanders have long noted it. They have done so in a systematic way in modern times, in some detail back to the sixteenth century; the sagas and chronicles before that mention drift ice occasionally. Drift ice, then, is a thermometer of the North Atlantic. But the readings must be interpreted. How do we calibrate records of driftice sightings, including some incomplete records, in terms of degrees of temperature? How cold was it when heavy sea ice threatened ships traveling to Norse homelands? 52 II. Our Climates since A.D. 900 Bergth6rsson and others have applied some clever techniques to this problem. While the answers they get are for average temperatures covering a decade or longer, and not for the daily readings we are accustomed to, the results are both fascinating and instructive. A basic principle in the interpretation of old, indirect climatic records is that of calibration with the detailed weather records of modern times. We have already seen examples of this process. For instance, the pollen records from Mill Creek can tell about climate because we know, from current scientific observation and understanding , which plants need wet weather and which ones have an advantage in dry weather. With drift ice, and with many other climatic indicators, modern records can be put alongside precise scientific records-in the case of drift ice, thermometer readings-from the same time. These precise records are a yardstick with which to measure the old records. In Iceland, we know the number of months drift ice appeared, and the temperature, for every year since 1846. So it's possible to make a chart of temperatures based on drift-ice information. Bergth6rsson worked with the average temperature for each decade, and in the records of the past century he found a direct and rather simple relationship between temperature and ice sightings. For example, if ice is sighted in 20 months of one decade, and 22 months of the next decade, that second decade is about O.loC (0.2°F) colder than the first. The rule, however, must be applied carefully; it does not hold as well in warm times. That makes sense; at the point when no drift ice is sighted at all, any additional warming can't make a difference. For the time before 1780, however, there aren't many records that discuss drift ice. The reconstruction of average temperatures in Iceland is, therefore, based on four types of information available for various times (Bergth6rsson, 1969, pp. 94-95). 1846 to the present thermometer readings 1781 to 1845 a chart of drift-ice sightings in each year, compiled by Thoroddsen 1591 to 1780 other weather information compiled by Thoroddsen from historical records (driftice records are sketchy) 900 to 1590 records of famines and severe years One Thouso'znd Years in Iceland 53 Bergth6rsson, then, had to establish the correlation between drift ice and thermometer readings; going further back he had to make some interpretations of recorded general comments on the weather; and (for the oldest period) he had to correlate severe years with drift ice (which in turn links them to temperature). Reconstructing a decade We can get an idea of how Bergth6rsson went about this work from one decade, the 1750s, in the period for which, on the basis of general weather comments and a few comments about drift ice, he had to estimate drift ice for each year. He did so as follows (1962, pp.3-4): 1751 No mention of drift ice [in the chronicles], but extensive "fast-ice" said to be at west coast. Comparison with the years 1866 and 1881, when fast-ice was rather similar, indicates drift ice duration of 5-6 months. 1752 No mention of ice. Cold from late F~:bruary until Easter. Cold spring. The frost was so severe that the earth cracked in places. The same phenomenon occurred in 1918, when drift ice duration was 1~ months. Furthermore, average ice duration after a severe ice year like the preceding one is two months. 1753 No mention of ice. There must, however, have been some drift ice, since in 1758 it is remarked that no drift ice was sighted and that this had not happened for time immemorial. Estimate : one month. 1754 A very severe winter. Drift ice lying well into the summer. Estimate: six months. 1755 Drift ice disappeared September 3, which is very late. Very extensive ice in the spring. Estimate: seven months. 1756 Drift ice from early March until August 25. Estimate: six months. 1757 Prolonged drift ice until early June. Estimate: three months. 1758 Statement that no ice was sighted. 1759 Severe drift ice off the north coast and even reaching the S.W. coast. Under usual conditions this would mean a drift ice duration of three to seven months, but since the winter, spring, and summer were not severe, the estimate is three months. 1760 No mention of ice. Estimate one month. With these estimates of drift ice, then, Bergth6rsson applied his formula that converts ice duration to temperature. 54 II. Our Climates since A.D. 900 For the time before 1591, BergthOrsson did not even have the skimpy information above. He did, however, have some references to bad years, years of famine. He says: There is hardly any event related to weather which is more worth recording than famine years. These are very much associated with coldness and ice in Iceland. The usual succession of events is the following . The springs and summers were cold and the hay-making failed frequently ; Then a severe winter killed a considerable part of the sheep, horses, and cattle. This was possibly not serious for the richer farmers who were usually better supplied with food and hay, but the poorer farmers quite often had very limited supplies and therefore their losses were heavier. The only possibility was then to go out and beg. The number of wandering beggars increased tremendously in such years, but at the same time the generosity of others decreased at the same rate. This resulted in starvation of the lowest classes, people dropping dead on their way between the farms, of hunger directly or diseases associated with hunger. In addition to this it seems that the fishery failed preferably in cold periods. With increasing coldness the frequency of severe famine years rose at an increasing rate. [1962, pp. 9-10] Bergth6rsson, then, uses historical references to severe years to extend his temperature chart back. Of course, this requires calibration of famine years in more recent (and hence better known) times with drift ice, and therefore with temperature. He does have a small amount of information about drift ice off the southwestern coast of Iceland from before 1591. Ice appears there only in very bad times; the main flows are along the eastern coast of the island. Since the population is concentrated at the southwest, as is the fishing industry, ice is mentioned even when not much else is. Bergth6rsson brought all this work together in the graph of figure 4.2. The solid part of the curve is taken from actual thermometer readings; the section before the mid-nineteenth century is estimated. For the early fifteenth century, BergthOrsson felt he did not have enough data to estimate temperatures. Bergth6rsson's picture shows that from the settlement of Iceland, about A.D. 900, until the late twelfth century, the times were warm, One Thousand Years in Iceland 55 OF 0c 41 40 39 38 . 4.5 :-. \ : ~ . .. 4.0 -. :-.. : ~ : : :: : .... . 3.5 .. :::: : _.: e.- ~.... ..: 3.0 ~~: 900 1100 1300 ............ ...J\ .. .... . .. ... . .- ...... .. 1500 1700 1900 Figure 4.2. A thousand years of Icelandic temperatures. The dotted line shows estimated temperatures (with the early part of the fifteenth century omitted because of insufficient data); the solid line shows temperatures based on thermometer readings: both from Bergthorsson, 1969, p. 98. The dashed line shows the change in recent years for the Northern Hemisphere; based on Bryson 1974a. though variable. Then Iceland was quite cold for almost 200 years. After the fourteenth century came a time of less drift ice and fewer beggars, though we are left somewhat unsure of the early fifteenth century. From the late sixteenth through the nineteenth century, the best years in Iceland were only as good as the worst years the Vikings had seen. Then, in our century, a great warming set in. How much confidence can we place in this chart, a record of temperatures based on such incomplete information? In some cases the reading for a whole decade or more comes from just a mention or two of famine. Precision and poetry Such a reconstruction offends those who in this age of science and technology, of precision in measurements, expect to know rainfall to the nearest one-hundredth of an inch. We have not a single thermometer reading from Iceland before the nineteenth century. But Bergth6rsson did have the drift-ice and famine records, as well as literary records like the following poem by Olafur Einarsson (15731659 ), a pastor in eastern Iceland: Formerly the earth produced all sorts of fruit, plants and roots. But now almost nothing grows.... 56 II. Our Climates since A.D. 900 Then the floods, the lakes and the blue waves brought abundant fish. But now hardly one can be seen. The misery increases more. The same applies to other goods.... Frost and cold torment people The good years are rare. H everything should be put in a verse Only a few take care of the miserables.... [Bergthorsson, 1962, p. 21] Does such a poem convey less of reality than a temperature chart? Perhaps it gives more. Whatever poetry does tell us, we would like to know the arithmetic of past climates. In the next two chapters, we will set a number of details against this chart. For now, we can consider it a hypothesis , a proposition to be tested. We can use it as a possible climate of the last 1,000 years, and seek other evidence to confirm or refute it. ...