OTHER RECORDS OF PAST CLIMATES
Diaries, annals, chronicles, etc.
In spite of the importance of reports of the weather by the people of the time — particularly any obviously seriously intentioned weather diaries, farm and estate records, mentions of weather difficulties in audited accounts and reports of building operations and repairs, etc. — little or no attempt seems to have been made to submit them to meteorological analysis until quite recent years.
Doubtless the main reason for this neglect has been the formidable size of the task.
A massive amount of material could ultimately be collected for Europe and the Far East and a few other regions; but it is dispersed in many kinds of archives and literature and problems are presented by ancient styles of handwriting and earlier forms of the many languages used, establishing the trustworthiness of the reporter, interpreting the dates, and so on.
(Fig. 29 shows part of a fourteenth century scroll, an ‘account roll’, from the manor of Knightsbridge, London, referring to a great drought in the summer of 1342.)
The dates were commonly given in state and other papers as the xth year of the reign of this or that prince, often princes of territories which have long since been merged or absorbed in the bigger countries of present-day Europe.
The difficulties produced by numerous differences of practice over the date from which each new year started, and in the time at which various countries introduced the modern (Gregorian) calendar, have also led to errors in ascribing the year of certain events or misplacing their exact date by several days in the texts derived by historians and others from the original sources.
Nevertheless, the wealth of material is such that every season of any kind of dramatic character in Europe since about AD 1100 is probably either known already or could finally be determined, especially now that the possibilities of verification from various kinds of proxy data have been greatly enlarged by modern techniques.
More documentary reports could certainly be unearthed if the effort could only be organized.
Many great archives have hardly yet been tapped for this purpose.
And where lengthy historical manuscripts have been printed and published, any weather information has commonly been omitted in the interest of economy, so that the original manuscripts should be sought out.
Reports exist, in fact, of a considerable number of remarkable seasons and severe events in much earlier times — the inscribed stones of ancient Babylon and the Near East and many centuries of records from China and India also await attention — and these, too, are open to comparisons with relevant fossil evidence.
The full list of historical sources of weather reports and references to ‘parameteorological’ events, such as great floods, parched ground, shipwrecks and damage to coasts is a long one.
It includes medieval monastic chronicles such as those of the Venerable Bede at Durham in Saxon England and Matthew Paris at St Albans (20 miles north of London), the audited accounts of great estates and their farms, legal and government papers e.g. reporting harvest difficulties, hunger and cattle raids, epidemics of disease and so on — harbour records, bridge repairs and many others.
The use of audited accounts is valuable: in documents such as these, where the weather is only mentioned as an explanation or excuse for expenditure and losses, false reports might be entered unless there was a system of checking by persons who could not be misled while the real character of the seasons concerned was still common knowledge.
The manors belonging to one famous English abbey were indeed found to have falsified their accounts in this way in the fourteenth century.
A few meteorologists and others of a historical turn of mind have published collections of these reports covering various parts of Europe, Iceland, eastern North America, Chile, China and Japan, some of which go back over many centuries.
A few published compilations of this sort appeared from the sixteenth century onwards in England and central Europe.
Famous later ones included Thomas Short’s A General Chronological History of the Air, Weather, Seasons, Meteors, in Sundry Places and Different Times etc. (London, 1749), a huge collection by the French meteorologist F. Arago published by the Academy of Sciences in 1858, R. Hennig’s Katalog bemerkenswerter Witterungsereignisse von den dltesten Zeiten bis zum Jahre 1800 (Berlin, Royal Prussian Academy, 1904) and C. Easton’s Les hivers dans l’Europe occidentale (Leyden, 1928).
The latest in the series are the many-volumed work by C. Weickinn Quellentexte zur Witterungsgeschichte Europas von der Zeitwende bis zum Jahre 1850 (Berlin, 1958–63) and the most thoroughly researched and verified compilation by M.K.E. Gottschalk of North Sea storm floods and river floods in the Netherlands from early times to the year 1700, in three volumes, published 1971–5 (Assen, van Gorcum).
All these compilations, except perhaps the last, contain mistakes, and in using them care is particularly needed to detect any multiple entries of the same flood or frost, etc. due to errors in the transcription of dates by the compilers or in earlier works that the compilers used.
On the other hand, the British compiler C.E. Britton (A Meteorological Chronology to AD 1450, London, Meteorological Office, 1937), confronted with the numerous reports of severe weather in various parts of the British Isles in the winters of the 1430s, addressed himself to the problem of which winter they all meant, and he seems never to have considered seriously the possibility that there might have been such frequent severe spells that the dates reported might be right.
Mapping the reports together with those from the continent appears in fact to substantiate spells of several weeks of widespread severe weather in seven, perhaps even eight, of the winters of that decade — an experience matched only in the 1690s so far as present knowledge goes.
The only safe recourse is to check the reports quoted in the compilations back to the original source documents, but the quantity of data available for Europe is such that placing the reports for each given season on maps shows up many of the faulty ones.
Despite the need for critical scrutiny, the compilers have not only effectively revealed the possibility of ultimately producing maps of the prevailing weather, season by season, each year back to the Middle Ages in Europe (and, perhaps, for eastern North America from the early 1600s); they have also provided the basis on which reasonable estimates of the prevailing temperatures and rainfall in successive half-century periods can be achieved.
Numerical indices were designed, and calculated, expressing for each decade the relative numbers (best expressed as a ratio) of reports of mild or cold months in winter and wet or dry months in summer.
The decade values of the winter and summer indices were then compared with the measured temperatures and rainfalls in the winters and summers in England in the case of the decades when instrument observations were available, after 1700.
From the statistical relationships revealed, temperature values and rainfall estimates could be derived for each decade back to the early Middle Ages.
Because of uncertainty about the completeness and accuracy of the reporting, however, it is unwise to rely on the figures for the individual decades, though the more extreme decades are probably correctly identified.
Taken in blocks of fifty years, the results are thought to be reliable as a first approximation to the temperature and rainfall history of England since AD 1100.
These derived histories are shown in figs. 30 and 31, on which estimates of the 100-year means for the eleventh century and the 200-year means for the period AD 800–1000 (derived in the same way but from scarcer reports) have been added.
The diagrams in this form are designed to show the range of uncertainty of the individual values.
What stands out is the certainty of a warmer period that lasted several centuries in the high Middle Ages and of an equally long period of colder climate culminating in the seventeenth century.
The rapidity of the decline in the fourteenth to fifteenth centuries and again in the late sixteenth century is verified as a striking feature.
Neither the decline between the years 1300 and 1600 nor the recovery from around 1700 to the present century were smooth, uninterrupted processes.
Even in the fifty-year means a time of easier conditions around 1500–50 and a reversion to colder conditions around 1800 come clearly to light.
The rainfall changes seem to indicate a long period when the totals were about 10 per cent below modern experience: this may be explained by colder seas and therefore less water vapour taken up into the atmosphere during the time of colder climate.
During the medieval warmth up to 7 per cent higher average yearly total rainfalls are indicated, but usually lower rainfall in the summer-time: this probably means that the warm summers were commonly influenced by the anticyclone belt, as is true of the occasional warm summers we get in England nowadays.
There is also an appearance of an interesting oscillation, whereby the summers of the second half of most centuries were wetter than those of the first half.
Other methods of converting descriptive reports to numerical values have led to long series of data for parts of the world outside Europe.
Thus, ‘content analysis’ applied by Catchpole and Moodie at the University of Manitoba to the linguistic terms used in the journals kept by the Hudson’s Bay Company’s staff at their fur trading posts near river mouths in the southern and southwestern parts of the Bay from the early eighteenth century onwards, and a little later on other rivers in northern Alberta, Saskatchewan and Manitoba, has produced year-by-year series of the dates of formation and break-up of the ice on the rivers.
The trend of the dates, averaged over five years at a time, shows quite good correspondence with the trend of the temperatures observed in Europe, including the warmth of the 1730s and the cold climax around 1880; though, unlike the experience in Europe, the decade 1810–19 seems to have produced mostly mild winters.
Rather similarly, H.Arakawa succeeded in extracting from Japanese documents a year-by-year series of the dates of freezing of the small Lake Suwa in central Japan from 1440.
This Japanese series shows less close correspondence with Europe than do the series from central Canada, though the fifty-year means are correlated significantly with those of Europe (the central England temperatures).
Just how much detail of the individual years can be taken as trustworthy from the descriptive reports available in any part of the world must be established either by checking back to the sources — and, preferably, some knowledge of the character and motives of the observer — or by comparison with other independent reports for the same season on a map capable of meteorological analysis.
A further alternative check may be provided by support from some kind of fossil data.
A number of handwritten journals which can more or less be described as weather diaries and some strictly daily registers of the weather, often with wind directions reported, are available: the quality and completeness of some of them, such as that kept by the Reverend Father Merle mostly at Driby in Lincolnshire (but also on his travels to and from Oxford) from 1337 to 1344, and that of the Danish astronomer, Tycho Brahe, from 1582 to 1597, are a testimony to the serious intention of the observer, so that errors, exaggeration or falsification seem most unlikely.
From the mid-sixteenth century onwards there are daily reports of the weather from somewhere or other in western or central Europe with few years omitted.
And from about 1670 there are in various archives many daily reports in the logs of ships in diverse ports and on the high seas (fig. 32).
For climatic research this is a vast treasure trove waiting to be used — for instance, to improve our knowledge and understanding of the severest phase of our climate in recent centuries, in the 1690s.
TO BE CONTINUED ...
CLIMATE, HISTORY AND THE MODERN WORLD
Re: CLIMATE, HISTORY AND THE MODERN WORLD
Grain prices records
Another type of unbroken year-by-year series is provided by the prices of wheat and sometimes other crops in various countries.
From early times until the agricultural advances in the eighteenth century — in some countries until about 1800 or after — the fluctuations in the price of grain, or of bread, responded essentially to the yield of the latest harvest.
This means that they registered each year's weather, except when the harvest was affected by wars, civil commotion or other disasters such as loss of manpower through disease epidemics.
W.H.Beveridge in the 1920s collected and analysed the wheat prices in England (at Exeter) from AD 1316 to 1820.
He found evidence of certain periodicities or cyclical tendencies of which, for instance, the statistically significant one around 5.1 years length and most of the others suspected (those of 12, 20 and 55 years) are close to period lengths commonly identified since in long series of climatic data.
Even in the smoothed version of the long histories of wheat prices seen in fig. 33a, the major price rises around AD 1300 and 1550–1650 can probably be largely attributed to climate, and that around 1800 probably owes something to climatic difficulties as well as to the Napoleonic wars.
In the year-by-year record of the German rye prices (fig. 33b) the peaks nearly all correspond to years of particularly unfavourable weather, the troughs to runs of good harvests: this is true even during and after the Napoleonic period.
(The remarkable year 1816, when the sun was dimmed by a thick veil of volcanic dust from the great eruption of Tamboro in the East Indies the previous year, became known in Europe and North America as ‘the year without a summer’.)
But in grain prices we are using what must really be classified as proxy data.
TO BE CONTINUED ...
Another type of unbroken year-by-year series is provided by the prices of wheat and sometimes other crops in various countries.
From early times until the agricultural advances in the eighteenth century — in some countries until about 1800 or after — the fluctuations in the price of grain, or of bread, responded essentially to the yield of the latest harvest.
This means that they registered each year's weather, except when the harvest was affected by wars, civil commotion or other disasters such as loss of manpower through disease epidemics.
W.H.Beveridge in the 1920s collected and analysed the wheat prices in England (at Exeter) from AD 1316 to 1820.
He found evidence of certain periodicities or cyclical tendencies of which, for instance, the statistically significant one around 5.1 years length and most of the others suspected (those of 12, 20 and 55 years) are close to period lengths commonly identified since in long series of climatic data.
Even in the smoothed version of the long histories of wheat prices seen in fig. 33a, the major price rises around AD 1300 and 1550–1650 can probably be largely attributed to climate, and that around 1800 probably owes something to climatic difficulties as well as to the Napoleonic wars.
In the year-by-year record of the German rye prices (fig. 33b) the peaks nearly all correspond to years of particularly unfavourable weather, the troughs to runs of good harvests: this is true even during and after the Napoleonic period.
(The remarkable year 1816, when the sun was dimmed by a thick veil of volcanic dust from the great eruption of Tamboro in the East Indies the previous year, became known in Europe and North America as ‘the year without a summer’.)
But in grain prices we are using what must really be classified as proxy data.
TO BE CONTINUED ...
Re: CLIMATE, HISTORY AND THE MODERN WORLD
Varieties of ‘fossil’ records showing yearly layers
There is no space here for any approach to a full coverage of the many kinds of
indirect data from which information about climate in the past may be derived.
Because weather and climate have an impact on nearly every aspect of our environment, the lines of evidence of past events are innumerable.
The reader who wishes for a more comprehensive survey must be referred to other works, such as the authors treatise Climate: Present, Past and Future—Volume 2: Climatic History and the Future, pp. 1–279 (London, Methuen, 1977) or M. Schwarzbach’s Das Klima der Vorzeit (Stuttgart, Enke, 1961).
A very brief overview of the main categories of evidence of past climate, including instrumental measurements, historical manuscripts and fossil data, both physical and biological, is given here in table 1 at the end of this chapter.
Some of the pieces of evidence left by past climates that differed from today’s could individually be interpreted in other ways, even attributed to primitive man’s disturbance of the landscape.
But many items, such as old moraines left by glaciers which have since shrunk or disappeared and old shorelines, are unequivocal.
Even in these cases, however, there are complexities to be unravelled: was the former glacier expansion mainly due to lower temperatures or to increased snowfall on the heights?
And has the lake or sea level in the area where the old shoreline is found been affected by warping of the Earth’s crust, that is by land-sinking or the reverse?
Problems arise with the interpretation of tree ring variations: at the site concerned how much was due to temperature and how much to moisture variations?
In all cases there is a problem to be solved by mathematical or statistical techniques of how far we can arrive at clear numerical estimates of the past temperatures, rainfalls, etc. within definable limits of error from the types of evidence available.
But, taking all the independent types of evidence together, there is no longer any doubt about the major features of the climatic history which they reveal.
Here we can only enlarge a little on a few things.
Particular importance must be attached to those items which register, however indirectly, the weather of each individual year — tree rings, year-layers in ice-sheets and glaciers, also year-layers (varves) in lake bed deposits — and to techniques of dating evidence of whatever kind.
Evidence of the year-by-year sequences just before, and during, the times of most rapid climatic change should be of great interest; but little has so far been done in this line of research.
On the other hand, special importance must also be accorded to those types of evidence from which most has already been learnt of the long history of climate; these include most notably pollen analysis, marine-biological studies of the deposits on the ocean bed, and oxygen isotope studies.
The oldest year-by-year record that has come down to us is the flood levels of the River Nile in lower Egypt, a variable which depends mostly on the summer monsoon rains over Ethiopia.
Yearly gauge readings at Cairo are available from the time of Mahomet, and some records inscribed on stone go back to the first dynasty of the pharaohs around 3100 BC.
There are problems, particularly in connection with the prolonged silting of the river bed which effectively changes the zero level, but enough data exist to show that, there was a sharp drop about 2800 BC to persistently lower flood levels than before.
The floods seem also to have been generally lower between the AD 600s and 1400, and again for a while around 1500, than in the seventeenth, eighteenth and nineteenth centuries before falling again to the present century.
From AD 622 the yearly gauge readings also record the seasonal low level of the Nile, which registers the flow of the White Nile fed by the equatorial rains over east Africa.
This seems to have been lowest in the AD 700s and around the 1600s and particularly high in and around the 1100s, 1450–1500 and about 1840–90.
Another splendid series is provided by the measurements of the thickness of the yearly mud layers in the bed of the small Lake Saki in the Crimea (near 45°N 33 1/2°E) from about 2300 BC; these are considered to register more or less directly variations in the (mainly summer) rainfall in that region.
A smoothed version of this record is seen in fig. 34.
A somewhat irregular fluctuation of about 200 years length seems to be an element in the story, but a more striking feature is the evidence of wetness in the Crimea that accompanied the periods of warmer and more genial climate in west and northwest Europe (and apparently over most of the northern hemisphere) before 2000 BC and in the early Middle Ages.
(The fluctuations of rainfall and run-off in the Crimea and much of southeast Europe are believed to be generally inverse to those over most of northern, western and central Europe.)
A number of features of the individual years’ record in the Crimea also look important.
There are signs of a fifty-year cycle, and perhaps a 20–23 year one, both of which are known to be present in the incidence of blocking anticyclones in the northeast Atlantic-northern Europe sector; this is a development which certainly affects rainfall over Russia.
But also extraordinary deviations in individual years in the Crimea seem to occur at times of major, long-lasting change of regime.
Just before the first big drop of the curve there were two years (2177 and 2150 BC) that produced mud layers of respectively five and ten times the normal thickness which had been typical of the wet regime that was just ending.
The further decline after 2000 BC may have been signalled by two runs of eight to eleven consecutive years in the nineteenth century BC producing extraordinarily thin mud layers.
Similarly extreme deviations in either direction — in AD 805 the only other occurrence of a thick layer approaching that of 2150 BC and a run of very thin layers in the AD 1280s — mark off the beginning and end of the medieval period of moist climate in the Crimea.
Even longer series of year-layers, or varves, in lake sediments are available for study from the beds of lakes and former lakes that have since disappeared, which formed around the margin of the former North American and Scandinavian ice sheets during the post-glacial melting.
Some of these series in Wisconsin and northern Sweden are nine thousand to ten thousand years long.
The original Swedish series, worked out by De Geer many years ego, spans the whole fifteen thousand years or so of the ice retreat and postglacial time: it is, however, a composite structure, built up from shorter overlapping series derived from lakes which individually had a shorter history; some uncertainty and controversy has developed about its structure.
The year-layers in the great land-based ice-sheets still present in Antarctica and Greenland also provide material for study.
The layers are made identifiable by seasonal changes in the density and texture of the snow, caused by temperature changes and wind.
Most simply, the thickness of each layer — after allowing for compression by the overlying younger ice — indicates the amount of snow which accumulated each year.
Allowance must be made for the slow movement of the ice, such that the older layers (except near the ice-sheet crest) have arrived by plastic flow from some distance from their present site, naturally from a somewhat higher level.
Measurements of the proportion of the heavy isotope of oxygen, the form with an atomic weight of 18 instead of the usual 16, present in the H20 of which the ice is composed, indicate the temperature of the snow at the time (however long ago) when it formed by condensation from the water vapour in the atmosphere.
Oxygen-18 measurements therefore make it possible to recognize the seasonal changes of temperature, and so they can help identification of the year-layers in cores taken from the ice-sheet.
This has been done for about the last 1500 years of the fossil record.
They also identify long lasting temperature changes in the course of changes of climate and make possible estimates of the magnitude and rate of change.
The Greenland record in fig. 35 indicates the whole course of the last ice age and post-glacial time — mostly dated by other methods - as well as much of the last interglacial.
A close analysis of the last thousand years of the record from this site in far northwest Greenland indicates a great deal about the changes of temperature and the downput of snow there, but it must not be assumed that the temperature sequence is identical with that anywhere in Europe.
The medieval warmth clearly reached its climax, and also ended, earlier there than in Europe.
The Little Ice Age affected north Greenland too, as fig. 36 indicates, but there were some differences of phasing.
Indeed, the climatic history obtained by various techniques from central longitudes of Canada appears to parallel that in western and northern Europe more closely than that from Greenland.
This is a discovery which was not altogether unexpected on meteorological grounds: it is doubtless related to the big waves in the westerlies.
TO BE CONTINUED ...
There is no space here for any approach to a full coverage of the many kinds of
indirect data from which information about climate in the past may be derived.
Because weather and climate have an impact on nearly every aspect of our environment, the lines of evidence of past events are innumerable.
The reader who wishes for a more comprehensive survey must be referred to other works, such as the authors treatise Climate: Present, Past and Future—Volume 2: Climatic History and the Future, pp. 1–279 (London, Methuen, 1977) or M. Schwarzbach’s Das Klima der Vorzeit (Stuttgart, Enke, 1961).
A very brief overview of the main categories of evidence of past climate, including instrumental measurements, historical manuscripts and fossil data, both physical and biological, is given here in table 1 at the end of this chapter.
Some of the pieces of evidence left by past climates that differed from today’s could individually be interpreted in other ways, even attributed to primitive man’s disturbance of the landscape.
But many items, such as old moraines left by glaciers which have since shrunk or disappeared and old shorelines, are unequivocal.
Even in these cases, however, there are complexities to be unravelled: was the former glacier expansion mainly due to lower temperatures or to increased snowfall on the heights?
And has the lake or sea level in the area where the old shoreline is found been affected by warping of the Earth’s crust, that is by land-sinking or the reverse?
Problems arise with the interpretation of tree ring variations: at the site concerned how much was due to temperature and how much to moisture variations?
In all cases there is a problem to be solved by mathematical or statistical techniques of how far we can arrive at clear numerical estimates of the past temperatures, rainfalls, etc. within definable limits of error from the types of evidence available.
But, taking all the independent types of evidence together, there is no longer any doubt about the major features of the climatic history which they reveal.
Here we can only enlarge a little on a few things.
Particular importance must be attached to those items which register, however indirectly, the weather of each individual year — tree rings, year-layers in ice-sheets and glaciers, also year-layers (varves) in lake bed deposits — and to techniques of dating evidence of whatever kind.
Evidence of the year-by-year sequences just before, and during, the times of most rapid climatic change should be of great interest; but little has so far been done in this line of research.
On the other hand, special importance must also be accorded to those types of evidence from which most has already been learnt of the long history of climate; these include most notably pollen analysis, marine-biological studies of the deposits on the ocean bed, and oxygen isotope studies.
The oldest year-by-year record that has come down to us is the flood levels of the River Nile in lower Egypt, a variable which depends mostly on the summer monsoon rains over Ethiopia.
Yearly gauge readings at Cairo are available from the time of Mahomet, and some records inscribed on stone go back to the first dynasty of the pharaohs around 3100 BC.
There are problems, particularly in connection with the prolonged silting of the river bed which effectively changes the zero level, but enough data exist to show that, there was a sharp drop about 2800 BC to persistently lower flood levels than before.
The floods seem also to have been generally lower between the AD 600s and 1400, and again for a while around 1500, than in the seventeenth, eighteenth and nineteenth centuries before falling again to the present century.
From AD 622 the yearly gauge readings also record the seasonal low level of the Nile, which registers the flow of the White Nile fed by the equatorial rains over east Africa.
This seems to have been lowest in the AD 700s and around the 1600s and particularly high in and around the 1100s, 1450–1500 and about 1840–90.
Another splendid series is provided by the measurements of the thickness of the yearly mud layers in the bed of the small Lake Saki in the Crimea (near 45°N 33 1/2°E) from about 2300 BC; these are considered to register more or less directly variations in the (mainly summer) rainfall in that region.
A smoothed version of this record is seen in fig. 34.
A somewhat irregular fluctuation of about 200 years length seems to be an element in the story, but a more striking feature is the evidence of wetness in the Crimea that accompanied the periods of warmer and more genial climate in west and northwest Europe (and apparently over most of the northern hemisphere) before 2000 BC and in the early Middle Ages.
(The fluctuations of rainfall and run-off in the Crimea and much of southeast Europe are believed to be generally inverse to those over most of northern, western and central Europe.)
A number of features of the individual years’ record in the Crimea also look important.
There are signs of a fifty-year cycle, and perhaps a 20–23 year one, both of which are known to be present in the incidence of blocking anticyclones in the northeast Atlantic-northern Europe sector; this is a development which certainly affects rainfall over Russia.
But also extraordinary deviations in individual years in the Crimea seem to occur at times of major, long-lasting change of regime.
Just before the first big drop of the curve there were two years (2177 and 2150 BC) that produced mud layers of respectively five and ten times the normal thickness which had been typical of the wet regime that was just ending.
The further decline after 2000 BC may have been signalled by two runs of eight to eleven consecutive years in the nineteenth century BC producing extraordinarily thin mud layers.
Similarly extreme deviations in either direction — in AD 805 the only other occurrence of a thick layer approaching that of 2150 BC and a run of very thin layers in the AD 1280s — mark off the beginning and end of the medieval period of moist climate in the Crimea.
Even longer series of year-layers, or varves, in lake sediments are available for study from the beds of lakes and former lakes that have since disappeared, which formed around the margin of the former North American and Scandinavian ice sheets during the post-glacial melting.
Some of these series in Wisconsin and northern Sweden are nine thousand to ten thousand years long.
The original Swedish series, worked out by De Geer many years ego, spans the whole fifteen thousand years or so of the ice retreat and postglacial time: it is, however, a composite structure, built up from shorter overlapping series derived from lakes which individually had a shorter history; some uncertainty and controversy has developed about its structure.
The year-layers in the great land-based ice-sheets still present in Antarctica and Greenland also provide material for study.
The layers are made identifiable by seasonal changes in the density and texture of the snow, caused by temperature changes and wind.
Most simply, the thickness of each layer — after allowing for compression by the overlying younger ice — indicates the amount of snow which accumulated each year.
Allowance must be made for the slow movement of the ice, such that the older layers (except near the ice-sheet crest) have arrived by plastic flow from some distance from their present site, naturally from a somewhat higher level.
Measurements of the proportion of the heavy isotope of oxygen, the form with an atomic weight of 18 instead of the usual 16, present in the H20 of which the ice is composed, indicate the temperature of the snow at the time (however long ago) when it formed by condensation from the water vapour in the atmosphere.
Oxygen-18 measurements therefore make it possible to recognize the seasonal changes of temperature, and so they can help identification of the year-layers in cores taken from the ice-sheet.
This has been done for about the last 1500 years of the fossil record.
They also identify long lasting temperature changes in the course of changes of climate and make possible estimates of the magnitude and rate of change.
The Greenland record in fig. 35 indicates the whole course of the last ice age and post-glacial time — mostly dated by other methods - as well as much of the last interglacial.
A close analysis of the last thousand years of the record from this site in far northwest Greenland indicates a great deal about the changes of temperature and the downput of snow there, but it must not be assumed that the temperature sequence is identical with that anywhere in Europe.
The medieval warmth clearly reached its climax, and also ended, earlier there than in Europe.
The Little Ice Age affected north Greenland too, as fig. 36 indicates, but there were some differences of phasing.
Indeed, the climatic history obtained by various techniques from central longitudes of Canada appears to parallel that in western and northern Europe more closely than that from Greenland.
This is a discovery which was not altogether unexpected on meteorological grounds: it is doubtless related to the big waves in the westerlies.
TO BE CONTINUED ...
Re: CLIMATE, HISTORY AND THE MODERN WORLD
Radiocarbon and its role in dating evidence
For times beyond the range of identifiable year-layers in ice-sheets and lake sediments, and beyond the longest yearly tree ring chronologies or human records in any part of the world, dates can be estimated by radiometric methods or, in the case of material in sediments, by assuming a broad constancy of sedimentation rate.
For the periods with which human history and archaeology are concerned the most important of the radiometric methods is radiocarbon dating.
This depends upon precise measurement of the radioactivity produced by the minute amount of the unstable isotope of carbon present in the carbon constituent of organic matter.
This radioactive isotope 14C, distinguished by an atomic weight of 14 — normal carbon has atomic weight 12 — is produced in the atmosphere by the effect of cosmic ray bombardment on the nitrogen atoms with which they collide.
Carbon 14 is assimilated into the structure of the living vegetation with the carbon dioxide breathed in from the atmosphere.
About 1 per cent of the carbon in living wood is the unstable, radioactive isotope; and the atoms of it decay, producing on average about fifteen disintegrations per minute per gram of carbon present.
After the death of the vegetation, which means cessation of the absorption of atmospheric carbon dioxide, its store of radioactivity is no longer renewed.
The activity therefore decays.
In scientific language, we say that the half-life of radiocarbon (14C) is 5730 years: this means that the activity falls by a half every 5730 years.
In practice therefore the amount of radioactivity dwindles and ultimately becomes very difficult to measure — and the errors produced by any contamination become greater — the older the material to be dated.
The effective limit of radiocarbon dating is about fifty thousand years.
Estimates of the margins of error arising from experimental difficulties are always quoted.
There is an additional source of error, however, established by radiocarbon dating of objects of known age and attributed to the fact that the amount of radioactive carbon in the atmosphere has evidently not been precisely constant down the ages.
The variations give rise to an error amounting to 500–1000 years in middle post-glacial times and to over 100 years in another period as recent as about AD 1400–1800.
These errors can, however, in most cases be corrected by using a calibration curve relating apparent radiocarbon ages to true ages obtained from specimens of the wood of bristlecone pine dated by its rings.
Because of these difficulties dates in the times before Christ based on radiocarbon tests with the results unadjusted for the calibration errors referred to are commonly distinguished in the newer literature by small letters, as for instance a date of 930 ± 220 be for the latest occurrence so far identified of a pine-tree stump significantly higher than the present upper treeline on the mountains of Scotland.
In this notation, be means before Christ and the ± 220 years is the estimate of the range of experimental error about the indicated date of 930.
Capital letters BC are used for firm calendar dates not resting on radiocarbon or other deductive methods and also for radiocarbon dates which have been corrected by applying the bristlecone pine calibration.
With the latter the estimated range of experimental error will be quoted.
In specifying the age of radiocarbon-tested material, the letters bp (or BP where the calibration has been applied) mean years ‘before present’, actually years before AD 1950 which has been adopted as a fixed datum at the beginning of radiocarbon work.
What is learnt from studying past glacier variations by dating organic matter and tree-stumps, etc. buried in old moraines will be referred to at appropriate points in the text of later chapters.
Here it is sufficient to note that there have been great variations of the mountain glaciers in all parts of the world during post-glacial time.
The course of major fluctuations eight to ten thousand years ago during the melting of the former ice-sheets can be followed.
Later moraines which may have been formed during the warmest post-glacial times have generally been obliterated by glacier advances since the time of minimum extent, which in some places was nearly six thousand years ago but more generally around 2000 BC.
TO BE CONTINUED ...
For times beyond the range of identifiable year-layers in ice-sheets and lake sediments, and beyond the longest yearly tree ring chronologies or human records in any part of the world, dates can be estimated by radiometric methods or, in the case of material in sediments, by assuming a broad constancy of sedimentation rate.
For the periods with which human history and archaeology are concerned the most important of the radiometric methods is radiocarbon dating.
This depends upon precise measurement of the radioactivity produced by the minute amount of the unstable isotope of carbon present in the carbon constituent of organic matter.
This radioactive isotope 14C, distinguished by an atomic weight of 14 — normal carbon has atomic weight 12 — is produced in the atmosphere by the effect of cosmic ray bombardment on the nitrogen atoms with which they collide.
Carbon 14 is assimilated into the structure of the living vegetation with the carbon dioxide breathed in from the atmosphere.
About 1 per cent of the carbon in living wood is the unstable, radioactive isotope; and the atoms of it decay, producing on average about fifteen disintegrations per minute per gram of carbon present.
After the death of the vegetation, which means cessation of the absorption of atmospheric carbon dioxide, its store of radioactivity is no longer renewed.
The activity therefore decays.
In scientific language, we say that the half-life of radiocarbon (14C) is 5730 years: this means that the activity falls by a half every 5730 years.
In practice therefore the amount of radioactivity dwindles and ultimately becomes very difficult to measure — and the errors produced by any contamination become greater — the older the material to be dated.
The effective limit of radiocarbon dating is about fifty thousand years.
Estimates of the margins of error arising from experimental difficulties are always quoted.
There is an additional source of error, however, established by radiocarbon dating of objects of known age and attributed to the fact that the amount of radioactive carbon in the atmosphere has evidently not been precisely constant down the ages.
The variations give rise to an error amounting to 500–1000 years in middle post-glacial times and to over 100 years in another period as recent as about AD 1400–1800.
These errors can, however, in most cases be corrected by using a calibration curve relating apparent radiocarbon ages to true ages obtained from specimens of the wood of bristlecone pine dated by its rings.
Because of these difficulties dates in the times before Christ based on radiocarbon tests with the results unadjusted for the calibration errors referred to are commonly distinguished in the newer literature by small letters, as for instance a date of 930 ± 220 be for the latest occurrence so far identified of a pine-tree stump significantly higher than the present upper treeline on the mountains of Scotland.
In this notation, be means before Christ and the ± 220 years is the estimate of the range of experimental error about the indicated date of 930.
Capital letters BC are used for firm calendar dates not resting on radiocarbon or other deductive methods and also for radiocarbon dates which have been corrected by applying the bristlecone pine calibration.
With the latter the estimated range of experimental error will be quoted.
In specifying the age of radiocarbon-tested material, the letters bp (or BP where the calibration has been applied) mean years ‘before present’, actually years before AD 1950 which has been adopted as a fixed datum at the beginning of radiocarbon work.
What is learnt from studying past glacier variations by dating organic matter and tree-stumps, etc. buried in old moraines will be referred to at appropriate points in the text of later chapters.
Here it is sufficient to note that there have been great variations of the mountain glaciers in all parts of the world during post-glacial time.
The course of major fluctuations eight to ten thousand years ago during the melting of the former ice-sheets can be followed.
Later moraines which may have been formed during the warmest post-glacial times have generally been obliterated by glacier advances since the time of minimum extent, which in some places was nearly six thousand years ago but more generally around 2000 BC.
TO BE CONTINUED ...
Re: CLIMATE, HISTORY AND THE MODERN WORLD
Pollen analysis and vegetation history
By far the most of what was known of post-glacial climatic history until 1950 was contributed by pollen analysis.
As long ago as 1876 the Norwegian botanist Axel Blytt first detected the broad post-glacial sequence of vegetation history, and this and the succession of climatic regimes which it points to were outlined in his Essay on the Immigration of the Norwegian Flora (Christiania).
As the ice melted, at first tundra vegetation was established, spreading from regions which had always been south of the ice, then birch trees infiltrated the tundra, and later birch and pine forest was established.
This in its turn was replaced by the mixed forest of broadleafed trees in and beyond all the regions now occupied by this forest type, which is now known to have been dominated by the warmth-loving elm and lime (linden) trees for some long time before about 3500 BC in the zone of present-day oak forest.
Each forest type presumably corresponded to the same sort of climate as that in which we find it today, except that in Europe generally there seem to have been delays of up to some thousands of years in the arrival of each tree species and forest type into territories that had earlier been too cold for it.
Further decades, centuries or longer had to pass before each new forest type became dominant in all the areas which had become suitable.
We can detect these delays by the quicker response of the insect populations to climatic change.
The evidence of the beetle species present at each stage is particularly well preserved.
No similar delays seem to have occurred over the broad plains of North America east of the Rocky Mountains, and it is clear that in Europe the advance of the successive forest types was hindered because the ice age refuges of the trees were south of the great mountain barriers.
Other complications in interpreting the results of pollen analysis concern the later retreat of the vegetation types from the greatest extent which they attained towards the north and on the heights in mid post-glacial times.
Soil deterioration set in and peatbogs replaced the forest in many persistently wet places, a development which can be attributed to the cumulative effect of a long period of wet climatic regime preceding.
In many places it does not imply a change of climate at the time when bog was replacing the forest.
In some places, moreover, this change of the landscape may have been assisted by man’s activities, making clearings in forest that was already under stress and grazing his animals on the cleared areas.
In southwest England and west Wales the times of clearance and cultivation by Neolithic farmers, evidenced by charcoal layers and the pollen of weeds like plantain associated with cultivation, often seem to coincide with the earliest layers of peat formation.
And from that time on, roughly covering the last five thousand years, in this part of Europe the problem of making deductions about climate from evidence of the prevailing vegetation character is increasingly complicated by man’s management of the land.
Another limitation of pollen analysis as a tool for reconstructing the climatic record is that, apart from the pollens in varved sediments and a few deposits of peat or lake bed sediments which may have grown very fast, it is seldom possible to achieve a time resolution, or fix, closer than a hundred years.
On the other hand, some very long records can be produced.
Two from Europe, one from Alsace and one from Macedonia, have provided records which start more than 125,000 years ago, indicating the entire course of the last interglacial from its early stages, and continuing right through the last ice age and post-glacial time.
The correspondence of the large-scale features of the sequence in Europe with the Greenland isotope curve, including the early shocks which heralded the end of the interglacial and beginning of the ice age, is highly satisfactory.
Professor R.G. West and his co-workers at Cambridge have performed similar analyses of the vegetation history of eastern England through several previous interglacial periods, making it possible to identify common features of the climatic development and, in very broad terms, its timing.
A model of how some specific indications of prevailing temperatures may be derived from evidence of plant distributions was demonstrated in a classic work by J. Iversen of the Danish Geological Survey in 1944.
By plotting the long-term average temperatures of the warmest and coldest months of the year (scaled along the x and y axes of graph paper) at all sites where a given plant was present (or, plotting in a different colour where it was absent) in the present climate of Denmark, Iversen was able to show in the case of holly (Ilex aquifolium), ivy (Hedera helix) and mistletoe (Viscum album) that the limits were quite strictly defined by certain temperature values.
TO BE CONTINUED ...
By far the most of what was known of post-glacial climatic history until 1950 was contributed by pollen analysis.
As long ago as 1876 the Norwegian botanist Axel Blytt first detected the broad post-glacial sequence of vegetation history, and this and the succession of climatic regimes which it points to were outlined in his Essay on the Immigration of the Norwegian Flora (Christiania).
As the ice melted, at first tundra vegetation was established, spreading from regions which had always been south of the ice, then birch trees infiltrated the tundra, and later birch and pine forest was established.
This in its turn was replaced by the mixed forest of broadleafed trees in and beyond all the regions now occupied by this forest type, which is now known to have been dominated by the warmth-loving elm and lime (linden) trees for some long time before about 3500 BC in the zone of present-day oak forest.
Each forest type presumably corresponded to the same sort of climate as that in which we find it today, except that in Europe generally there seem to have been delays of up to some thousands of years in the arrival of each tree species and forest type into territories that had earlier been too cold for it.
Further decades, centuries or longer had to pass before each new forest type became dominant in all the areas which had become suitable.
We can detect these delays by the quicker response of the insect populations to climatic change.
The evidence of the beetle species present at each stage is particularly well preserved.
No similar delays seem to have occurred over the broad plains of North America east of the Rocky Mountains, and it is clear that in Europe the advance of the successive forest types was hindered because the ice age refuges of the trees were south of the great mountain barriers.
Other complications in interpreting the results of pollen analysis concern the later retreat of the vegetation types from the greatest extent which they attained towards the north and on the heights in mid post-glacial times.
Soil deterioration set in and peatbogs replaced the forest in many persistently wet places, a development which can be attributed to the cumulative effect of a long period of wet climatic regime preceding.
In many places it does not imply a change of climate at the time when bog was replacing the forest.
In some places, moreover, this change of the landscape may have been assisted by man’s activities, making clearings in forest that was already under stress and grazing his animals on the cleared areas.
In southwest England and west Wales the times of clearance and cultivation by Neolithic farmers, evidenced by charcoal layers and the pollen of weeds like plantain associated with cultivation, often seem to coincide with the earliest layers of peat formation.
And from that time on, roughly covering the last five thousand years, in this part of Europe the problem of making deductions about climate from evidence of the prevailing vegetation character is increasingly complicated by man’s management of the land.
Another limitation of pollen analysis as a tool for reconstructing the climatic record is that, apart from the pollens in varved sediments and a few deposits of peat or lake bed sediments which may have grown very fast, it is seldom possible to achieve a time resolution, or fix, closer than a hundred years.
On the other hand, some very long records can be produced.
Two from Europe, one from Alsace and one from Macedonia, have provided records which start more than 125,000 years ago, indicating the entire course of the last interglacial from its early stages, and continuing right through the last ice age and post-glacial time.
The correspondence of the large-scale features of the sequence in Europe with the Greenland isotope curve, including the early shocks which heralded the end of the interglacial and beginning of the ice age, is highly satisfactory.
Professor R.G. West and his co-workers at Cambridge have performed similar analyses of the vegetation history of eastern England through several previous interglacial periods, making it possible to identify common features of the climatic development and, in very broad terms, its timing.
A model of how some specific indications of prevailing temperatures may be derived from evidence of plant distributions was demonstrated in a classic work by J. Iversen of the Danish Geological Survey in 1944.
By plotting the long-term average temperatures of the warmest and coldest months of the year (scaled along the x and y axes of graph paper) at all sites where a given plant was present (or, plotting in a different colour where it was absent) in the present climate of Denmark, Iversen was able to show in the case of holly (Ilex aquifolium), ivy (Hedera helix) and mistletoe (Viscum album) that the limits were quite strictly defined by certain temperature values.
TO BE CONTINUED ...
Re: CLIMATE, HISTORY AND THE MODERN WORLD
The post-glacial record and evidence from beetles
The smooth curves in fig. 37, which outline the history of the prevailing temperatures in England since the depths of the last ice age about twenty thousand years ago, are an example of what may be deduced about the climate by way of pollen analysis used to reconstruct the vegetation history.
These curves are thought to be a reasonable approximation to the 500-or 1000-year mean temperatures of the high summer and winter months.
Points added about the right-hand end of each curve indicate the amount of deviation apparently distinguishing the last twelve individual centuries.
A similar variability may have occurred in earlier millennia, and indeed we shall notice evidence of it within the last three thousand years. Indications derived from the more rapidly changing beetle faunas have been used to fix the timing and magnitude of the sharpest climatic changes at the end of the last ice age.
Many fascinating studies by G.R.Coope at Birmingham have demonstrated the value of the well-preserved remains of the abundant beetle faunas and their rapid responses to climatic change in this field of research.
The main features of the curves in fig. 37 are indeed the rapidity of the changeover to post-glacial temperatures, particularly the early peak of the summer temperatures, and the drastic reversion to a glacial climate for a few centuries in the eleventh millennium before Christ.
The warmest post-glacial times appear as a broad hump of the curves between eight thousand and four thousand years ago, followed by an undulating decline to our present climate.
The variations which we can determine within the last millennium or more include some sharp fluctuations of shorter time-scale.
TO BE CONTINUED ...
The smooth curves in fig. 37, which outline the history of the prevailing temperatures in England since the depths of the last ice age about twenty thousand years ago, are an example of what may be deduced about the climate by way of pollen analysis used to reconstruct the vegetation history.
These curves are thought to be a reasonable approximation to the 500-or 1000-year mean temperatures of the high summer and winter months.
Points added about the right-hand end of each curve indicate the amount of deviation apparently distinguishing the last twelve individual centuries.
A similar variability may have occurred in earlier millennia, and indeed we shall notice evidence of it within the last three thousand years. Indications derived from the more rapidly changing beetle faunas have been used to fix the timing and magnitude of the sharpest climatic changes at the end of the last ice age.
Many fascinating studies by G.R.Coope at Birmingham have demonstrated the value of the well-preserved remains of the abundant beetle faunas and their rapid responses to climatic change in this field of research.
The main features of the curves in fig. 37 are indeed the rapidity of the changeover to post-glacial temperatures, particularly the early peak of the summer temperatures, and the drastic reversion to a glacial climate for a few centuries in the eleventh millennium before Christ.
The warmest post-glacial times appear as a broad hump of the curves between eight thousand and four thousand years ago, followed by an undulating decline to our present climate.
The variations which we can determine within the last millennium or more include some sharp fluctuations of shorter time-scale.
TO BE CONTINUED ...
Re: CLIMATE, HISTORY AND THE MODERN WORLD
Archaeology
The archaeology of man and the larger mammals is on the whole less informative about past climates than the record of the insects and pollens and micro-organisms in the sea that are at the mercy of their immediate surroundings.
The mobility and adaptability of man and the larger animals allow them to roam widely, experimenting with unfamiliar foods when necessity drives, and straying at least temporarily into environments which might in the long run prove hostile.
There are exceptions where deductions can be made about past climate, as for instance where routes of travel were established across terrain that at other times was closed by snow and ice, desert or marsh.
But one type of find deserves mention here, that of the men of about two thousand years ago whose whole bodies have been found preserved in the peat-bogs of Denmark (and comparable finds in the peat of the Hebrides and of bodies released by melting of the Greenland ice): in some of these cases the preservation has been so good that analysis of the stomach contents tells us about the food they lived on.
In the Danish cases these results have been tied in with near-by pollen analysis studies to confirm the cultivation of Iron Age fields in the neighbourhood and the weeds that were eaten along with the primitive barley and linseed.
We also know of a Danish alcoholic drink of two thousand years ago, made from barley, cranberry and bog myrtle: this from analysis of residues in pottery of the period.
TO BE CONTINUED ...
The archaeology of man and the larger mammals is on the whole less informative about past climates than the record of the insects and pollens and micro-organisms in the sea that are at the mercy of their immediate surroundings.
The mobility and adaptability of man and the larger animals allow them to roam widely, experimenting with unfamiliar foods when necessity drives, and straying at least temporarily into environments which might in the long run prove hostile.
There are exceptions where deductions can be made about past climate, as for instance where routes of travel were established across terrain that at other times was closed by snow and ice, desert or marsh.
But one type of find deserves mention here, that of the men of about two thousand years ago whose whole bodies have been found preserved in the peat-bogs of Denmark (and comparable finds in the peat of the Hebrides and of bodies released by melting of the Greenland ice): in some of these cases the preservation has been so good that analysis of the stomach contents tells us about the food they lived on.
In the Danish cases these results have been tied in with near-by pollen analysis studies to confirm the cultivation of Iron Age fields in the neighbourhood and the weeds that were eaten along with the primitive barley and linseed.
We also know of a Danish alcoholic drink of two thousand years ago, made from barley, cranberry and bog myrtle: this from analysis of residues in pottery of the period.
TO BE CONTINUED ...
Re: CLIMATE, HISTORY AND THE MODERN WORLD
Tree rings
The potential of the yearly growth rings of trees to indicate a detailed record of climate was for long little exploited outside North America.
Ring widths in trees occurring at the upper altitude limit and at the poleward limit are relatively straightforward to interpret climatically.
(Good growth years are warm years, poor growth years are years with cold summers in these positions.)
The same applies, though apparently to some less extent, to trees near the warm-arid margin of their occurrence, where it is failure of the rain that explains narrow growth rings.
The situation is much more complicated in the middle of the habitat of the tree species, though useful possibilities of interpretation may exist where the tree is known to have grown on a dry site where moisture stress might often be important or on a permanently damp site where temperature fluctuations might be the only important variable.
H.C.Fritts at the Laboratory of Tree Ring Research at the University of Arizona in Tucson has studied in some detail how the width of each ring is related to the weather of the preceding fifteen months.
His colleague V.C.La Marche has built up a chronology of ring widths in the very long-lived bristlecone pine trees on the heights of the White Mountains on the California-Nevada border.
This record extends back to 3431 BC and its dating has been demonstrated as sound by repetition.
This long series at the upper tree line essentially registers summer temperature.
It is of interest that from AD 800 to the present century its hundred-year averages are correlated, in a statistically significant degree, with the temperatures derived for central England.
They tend therefore to provide independent support for the English temperature history presented in fig. 30.
The tree ring method, or ‘dendroclimatology’, is progressively coming into much more extended use in climatic research.
Finds of well-preserved tree trunks buried in peat in Ireland, in eastern England and in river gravels in central Germany promise to produce an ultimately unbroken chronology extending back at least four thousand years and possibly much longer.
The large size of the oaks which grew in the fenlands of eastern England in the warmest post-glacial times attracted the attention of H. Godwin already many years ago.
Estimates of the temperatures prevailing in those times may be derived from observation of the greater height of the upper tree limit on the mountains of Europe (and other parts of the world) and the northward displacement of the northern limits of various species.
Curves representing the post-glacial history of the upper tree line in various parts of the world tend to parallel the last ten thousand years of the temperature curves shown in fig. 37.
In the closer detail which we can survey in the last thousand years, Dr J. Fletcher of the Research Laboratory for Archaeology and the History of Art, Oxford, has compared the tree ring sequences shown by English and German oaks, using for the former a chronology built up from English oak chests in Westminster Abbey.
He finds that for some centuries in the Middle Ages prior to AD 1250 there was 70–75 per cent agreement between the sequences of broader and narrower years, growth rings in the English and German trees, but after 1400 the agreement fell to only 50–55 per cent.
Presumably westerly winds sweeping across both countries were much commoner in the former period than in the latter.
This is not altogether surprising, since the latter period was when the colder climate of the Little Ice Age was setting in and winds from the north and east are known to have increased in frequency.
Other new possibilities in tree ring work are offered by measuring isotope ratios and by X-ray techniques for measuring wood density and examining cell sizes.
The ratios of the stable isotopes of oxygen, and similarly of the stable isotopes of carbon, hydrogen and nitrogen, in the substance of the wood may each lead to estimates of the temperatures prevailing when the wood was formed.
Examination of the cell structure of the wood, as is being pioneered by F.Schweingruber and his colleagues at the Forest Research Institute at Birmensdorf near Zürich, may be even more promising.
The large cells of the early season’s growth are followed by a darker pack of dense, smaller cells in the late summer wood; but sometimes even variations within the season can be established.
Furthermore, better correlations have been obtained across wide areas of Europe between the density measurements than in the case of ring widths.
TO BE CONTINUED ...
The potential of the yearly growth rings of trees to indicate a detailed record of climate was for long little exploited outside North America.
Ring widths in trees occurring at the upper altitude limit and at the poleward limit are relatively straightforward to interpret climatically.
(Good growth years are warm years, poor growth years are years with cold summers in these positions.)
The same applies, though apparently to some less extent, to trees near the warm-arid margin of their occurrence, where it is failure of the rain that explains narrow growth rings.
The situation is much more complicated in the middle of the habitat of the tree species, though useful possibilities of interpretation may exist where the tree is known to have grown on a dry site where moisture stress might often be important or on a permanently damp site where temperature fluctuations might be the only important variable.
H.C.Fritts at the Laboratory of Tree Ring Research at the University of Arizona in Tucson has studied in some detail how the width of each ring is related to the weather of the preceding fifteen months.
His colleague V.C.La Marche has built up a chronology of ring widths in the very long-lived bristlecone pine trees on the heights of the White Mountains on the California-Nevada border.
This record extends back to 3431 BC and its dating has been demonstrated as sound by repetition.
This long series at the upper tree line essentially registers summer temperature.
It is of interest that from AD 800 to the present century its hundred-year averages are correlated, in a statistically significant degree, with the temperatures derived for central England.
They tend therefore to provide independent support for the English temperature history presented in fig. 30.
The tree ring method, or ‘dendroclimatology’, is progressively coming into much more extended use in climatic research.
Finds of well-preserved tree trunks buried in peat in Ireland, in eastern England and in river gravels in central Germany promise to produce an ultimately unbroken chronology extending back at least four thousand years and possibly much longer.
The large size of the oaks which grew in the fenlands of eastern England in the warmest post-glacial times attracted the attention of H. Godwin already many years ago.
Estimates of the temperatures prevailing in those times may be derived from observation of the greater height of the upper tree limit on the mountains of Europe (and other parts of the world) and the northward displacement of the northern limits of various species.
Curves representing the post-glacial history of the upper tree line in various parts of the world tend to parallel the last ten thousand years of the temperature curves shown in fig. 37.
In the closer detail which we can survey in the last thousand years, Dr J. Fletcher of the Research Laboratory for Archaeology and the History of Art, Oxford, has compared the tree ring sequences shown by English and German oaks, using for the former a chronology built up from English oak chests in Westminster Abbey.
He finds that for some centuries in the Middle Ages prior to AD 1250 there was 70–75 per cent agreement between the sequences of broader and narrower years, growth rings in the English and German trees, but after 1400 the agreement fell to only 50–55 per cent.
Presumably westerly winds sweeping across both countries were much commoner in the former period than in the latter.
This is not altogether surprising, since the latter period was when the colder climate of the Little Ice Age was setting in and winds from the north and east are known to have increased in frequency.
Other new possibilities in tree ring work are offered by measuring isotope ratios and by X-ray techniques for measuring wood density and examining cell sizes.
The ratios of the stable isotopes of oxygen, and similarly of the stable isotopes of carbon, hydrogen and nitrogen, in the substance of the wood may each lead to estimates of the temperatures prevailing when the wood was formed.
Examination of the cell structure of the wood, as is being pioneered by F.Schweingruber and his colleagues at the Forest Research Institute at Birmensdorf near Zürich, may be even more promising.
The large cells of the early season’s growth are followed by a darker pack of dense, smaller cells in the late summer wood; but sometimes even variations within the season can be established.
Furthermore, better correlations have been obtained across wide areas of Europe between the density measurements than in the case of ring widths.
TO BE CONTINUED ...
Re: CLIMATE, HISTORY AND THE MODERN WORLD
Ocean bed deposits
There is one valuable source of information about past climates which we have not so far mentioned.
Rather as pollen counts at different levels — effectively different ages — in cores taken from peat and lake bed deposits on land (and on former lands now submerged by the sea) can give us a very long record of vegetation and climate history, but with limited time resolution, so have counts of the representatives of various marine biological (micro-) species in the deposits on the ocean bed yielded extremely long records from which sea-surface temperature histories can be derived.
And such conclusions can be verified by oxygen isotope measurements applied to the calcium carbonate (CaCO3) in the skeletal remains concerned.
Most ocean bed deposits, which also include inorganic mineral dust blown by the winds, accumulate extremely slowly: 1–4 cm per thousand years seems to be typical, compared with 1–4 cm per century for peat in west European peatbogs.
This means that the deposits sampled on undisturbed parts of the ocean bed may yield extremely long records.
The longest so far, from the equatorial Pacific, analysed by N.J.Shackleton of Cambridge, goes back over more than two million years and indicates that ice ages have occurred approximately every hundred thousand years during the last million years or more.
Again comparisons of the course of climatic development in each interglacial and its timing are possible.
Such slow sedimentation means, however, that one cannot distinguish variations of shorter duration than some thousands of years.
In any case, the minute burrowing animals that live within the sediment blur the record, mixing at any given time the uppermost layer of the ooze laid down over about the last two thousand years.
This is a limitation on the time resolution of data from ocean sediments, which means that while they are of great importance to our knowledge of the long-term and gross-scale changes of climate between ice ages and interglacials, and to verify what we have derived from other evidence, they offer little of the shorter-term detail with which human history is largely concerned.
Oceanographic research in the realms which here interest us should therefore concern itself with places where the sediment is laid down ten or more times as rapidly as the world average — and remains undisturbed — and, at the other end of the scale, with periods that can be covered by fisheries and whaling records, etc.
TO BE CONTINUED ...
There is one valuable source of information about past climates which we have not so far mentioned.
Rather as pollen counts at different levels — effectively different ages — in cores taken from peat and lake bed deposits on land (and on former lands now submerged by the sea) can give us a very long record of vegetation and climate history, but with limited time resolution, so have counts of the representatives of various marine biological (micro-) species in the deposits on the ocean bed yielded extremely long records from which sea-surface temperature histories can be derived.
And such conclusions can be verified by oxygen isotope measurements applied to the calcium carbonate (CaCO3) in the skeletal remains concerned.
Most ocean bed deposits, which also include inorganic mineral dust blown by the winds, accumulate extremely slowly: 1–4 cm per thousand years seems to be typical, compared with 1–4 cm per century for peat in west European peatbogs.
This means that the deposits sampled on undisturbed parts of the ocean bed may yield extremely long records.
The longest so far, from the equatorial Pacific, analysed by N.J.Shackleton of Cambridge, goes back over more than two million years and indicates that ice ages have occurred approximately every hundred thousand years during the last million years or more.
Again comparisons of the course of climatic development in each interglacial and its timing are possible.
Such slow sedimentation means, however, that one cannot distinguish variations of shorter duration than some thousands of years.
In any case, the minute burrowing animals that live within the sediment blur the record, mixing at any given time the uppermost layer of the ooze laid down over about the last two thousand years.
This is a limitation on the time resolution of data from ocean sediments, which means that while they are of great importance to our knowledge of the long-term and gross-scale changes of climate between ice ages and interglacials, and to verify what we have derived from other evidence, they offer little of the shorter-term detail with which human history is largely concerned.
Oceanographic research in the realms which here interest us should therefore concern itself with places where the sediment is laid down ten or more times as rapidly as the world average — and remains undisturbed — and, at the other end of the scale, with periods that can be covered by fisheries and whaling records, etc.
TO BE CONTINUED ...
Re: CLIMATE, HISTORY AND THE MODERN WORLD
Part II
CLIMATE AND HISTORY
CLIMATE AT THE DAWN OF HISTORY
The last three chapters have indicated in outline what is now known of the past record of climate and what's more, given further research, we may hope to reconstruct.
Further details of present knowledge will be mentioned at appropriate points in this and later chapters.
Let us now look at the everchanging scene of climate and environment as the stage on which the history of mankind has been played.
THE ICE AGE WORLD AND THE PEOPLING OF THE AMERICAS AND AUSTRALIA
The earliest gleam of man’s own record of his story comes to us in the sketches and paintings on the walls of caves, the record left to us by the inhabitants of central France and northern Spain during the last ice age of the environment which they knew between forty thousand and fifteen thousand years ago (fig. 38).
It is a world of bison and other wild cattle, mammoths, rhinoceros, horses and deer, hunted down with arrows and spears, in a treeless landscape.
The areas concerned were, of course, always beyond the range of the great ice sheets.
But we see man living in caves in the rock walls of valleys that were still habitable in the Dordogne, the Pyrenees and Cantabria, adapted to and exploiting a landscape and a fauna that differed from today’s.
Similar cave paintings, all executed in red, found in the Kapovaia cave in the southern Urals show seven mammoths and two rhinoceroses as well is a number of horses.
A big encampment of mammoth hunters has been identified at Vyzovsk, at 65 °N on the Pechora river in the northeast of European Russia: 98 per cent of the bones dumped were mammoth, and the remains of a dwelling built of mammoth bones were found.
Another find, near Vladimir, east of Moscow, indicates the clothing worn by people of the once common European Cro-Magnon type thirty thousand years ago.
The articles had been decorated with ivory beads, which still traced the form of a pullover shirt with a round neck and trousers with boots, also a head covering.
Archaeological finds in North America point to human beings living in ice-free areas of Alaska, north of the ice-sheets, during the last ice age and probably therefore roaming to and fro across the dry plain which then linked Alaska to Siberia.
This dry land existed because world sea level was lowered about 100 m by the loss of the water constituting the expanded glaciers and ice-sheets.
The Mongoloid traits of the American Indians suggest that their ancestors came from Asia, most likely exploiting the dry land connection where the Bering Strait now exists.
The distribution of earliest dated archaeological traces of human occupation of the Americas suggests arrival during the ice age; and there is some probability that an ice-free corridor through Alberta, between the ‘Cordilleran’ ice-sheet over the Rocky Mountains and the huge ‘Laurentide’ ice-sheet centred where Hudson Bay is now, was used for the migration south.
This corridor existed for many thousands of years during relatively milder phases in the middle of the ice age and it re-appeared about twelve thousand years ago, towards the end of the glaciation, when the ice-sheets were dwindling.
The dating of finds associated with human activity hints at migrations by this route in both these periods.
Radiocarbon dates also indicate the arrival of the first human population in Australia during the ice age, perhaps forty thousand years ago, in the time when the lowered sea level created great stretches of dry land almost linking Australia to Asia.
But there, there do seem to have remained some open water straits which the people somehow managed to cross.
Thus, we see the early hunting and gathering communities of human beings living often, though not everywhere or in all cases, in sparsely distributed groups, restricted in their range by the barriers of ice and ocean and high mountains but also exploiting the opportunities that the ice age world offered.
An aspect of the ice age world that has not been much written about is the enormously greater extent of many lakes and inland seas in temperate and lower latitudes.
They were there because of shifts of the main rainfall belts and the reduced evaporation resulting from lower temperatures than now and increased cloudiness.
The Caspian Sea spread far to the north-west and north of its present shores into the central and eastern part of European Russia, and attained over twice its present size.
And in early post-glacial times the Arctic Ocean waters invaded much of northwest Siberia from the north, where the land had been depressed by the ice load.
Lake Chad, which is at present but a remnant in the southern fringe of the Sahara, became a great inland sea in ice age times as big as the present Caspian.
And in North America west of the main watershed, the continental divide, there were numerous lakes, the biggest of which was another great inland sea, Lake Bonneville: this spread out from the present Great Salt Lake of Utah to attain an area of over 50,000 km2 (as big as the present Aral Sea in central Asia) and a depth of over 300 m.
Other lakes in the same general region included Lake Lahontan with an area of about 25,000 km2 in northwestern Nevada and Searles Lake and the Salton Lake in southeastern California.
This watery landscape in the western mountain region of the present United States continued in existence into early post-glacial times, as long as the dwindling
Laurentide ice-sheet still covered much of Canada.
In Australia, too, there were lakes where there are none today.
TO BE CONTINUED ...
CLIMATE AND HISTORY
CLIMATE AT THE DAWN OF HISTORY
The last three chapters have indicated in outline what is now known of the past record of climate and what's more, given further research, we may hope to reconstruct.
Further details of present knowledge will be mentioned at appropriate points in this and later chapters.
Let us now look at the everchanging scene of climate and environment as the stage on which the history of mankind has been played.
THE ICE AGE WORLD AND THE PEOPLING OF THE AMERICAS AND AUSTRALIA
The earliest gleam of man’s own record of his story comes to us in the sketches and paintings on the walls of caves, the record left to us by the inhabitants of central France and northern Spain during the last ice age of the environment which they knew between forty thousand and fifteen thousand years ago (fig. 38).
It is a world of bison and other wild cattle, mammoths, rhinoceros, horses and deer, hunted down with arrows and spears, in a treeless landscape.
The areas concerned were, of course, always beyond the range of the great ice sheets.
But we see man living in caves in the rock walls of valleys that were still habitable in the Dordogne, the Pyrenees and Cantabria, adapted to and exploiting a landscape and a fauna that differed from today’s.
Similar cave paintings, all executed in red, found in the Kapovaia cave in the southern Urals show seven mammoths and two rhinoceroses as well is a number of horses.
A big encampment of mammoth hunters has been identified at Vyzovsk, at 65 °N on the Pechora river in the northeast of European Russia: 98 per cent of the bones dumped were mammoth, and the remains of a dwelling built of mammoth bones were found.
Another find, near Vladimir, east of Moscow, indicates the clothing worn by people of the once common European Cro-Magnon type thirty thousand years ago.
The articles had been decorated with ivory beads, which still traced the form of a pullover shirt with a round neck and trousers with boots, also a head covering.
Archaeological finds in North America point to human beings living in ice-free areas of Alaska, north of the ice-sheets, during the last ice age and probably therefore roaming to and fro across the dry plain which then linked Alaska to Siberia.
This dry land existed because world sea level was lowered about 100 m by the loss of the water constituting the expanded glaciers and ice-sheets.
The Mongoloid traits of the American Indians suggest that their ancestors came from Asia, most likely exploiting the dry land connection where the Bering Strait now exists.
The distribution of earliest dated archaeological traces of human occupation of the Americas suggests arrival during the ice age; and there is some probability that an ice-free corridor through Alberta, between the ‘Cordilleran’ ice-sheet over the Rocky Mountains and the huge ‘Laurentide’ ice-sheet centred where Hudson Bay is now, was used for the migration south.
This corridor existed for many thousands of years during relatively milder phases in the middle of the ice age and it re-appeared about twelve thousand years ago, towards the end of the glaciation, when the ice-sheets were dwindling.
The dating of finds associated with human activity hints at migrations by this route in both these periods.
Radiocarbon dates also indicate the arrival of the first human population in Australia during the ice age, perhaps forty thousand years ago, in the time when the lowered sea level created great stretches of dry land almost linking Australia to Asia.
But there, there do seem to have remained some open water straits which the people somehow managed to cross.
Thus, we see the early hunting and gathering communities of human beings living often, though not everywhere or in all cases, in sparsely distributed groups, restricted in their range by the barriers of ice and ocean and high mountains but also exploiting the opportunities that the ice age world offered.
An aspect of the ice age world that has not been much written about is the enormously greater extent of many lakes and inland seas in temperate and lower latitudes.
They were there because of shifts of the main rainfall belts and the reduced evaporation resulting from lower temperatures than now and increased cloudiness.
The Caspian Sea spread far to the north-west and north of its present shores into the central and eastern part of European Russia, and attained over twice its present size.
And in early post-glacial times the Arctic Ocean waters invaded much of northwest Siberia from the north, where the land had been depressed by the ice load.
Lake Chad, which is at present but a remnant in the southern fringe of the Sahara, became a great inland sea in ice age times as big as the present Caspian.
And in North America west of the main watershed, the continental divide, there were numerous lakes, the biggest of which was another great inland sea, Lake Bonneville: this spread out from the present Great Salt Lake of Utah to attain an area of over 50,000 km2 (as big as the present Aral Sea in central Asia) and a depth of over 300 m.
Other lakes in the same general region included Lake Lahontan with an area of about 25,000 km2 in northwestern Nevada and Searles Lake and the Salton Lake in southeastern California.
This watery landscape in the western mountain region of the present United States continued in existence into early post-glacial times, as long as the dwindling
Laurentide ice-sheet still covered much of Canada.
In Australia, too, there were lakes where there are none today.
TO BE CONTINUED ...