The Climate-Environment-Society Nexus in the Sahara from ... · PDF fileLERNIA, NICK DRAKE,...

The Climate-Environment-Society Nexus in theSahara from Prehistoric Times to the Present Day

NICK BROOKS, ISABELLE CHIAPELLO, SAVINO DILERNIA, NICK DRAKE, MICHEL LEGRAND, CYRIL

MOULIN AND JOSEPH PROSPERO

The Sahara is a key region for studies of archaeology, human-environment interaction, globalbiogeochemical cycles, and global climate change. With a few notable exceptions, the region isthe subject of very little international scientific research, a fact that is remarkable given theSahara’s proximity to Europe, the developmental issues facing its growing population, theregion’s sensitivity to climate change and the Sahara’s potential for influencing globalclimate through the export of airborne mineral dust. This article seeks to address human-environment interaction in the Sahara from an interdisciplinary perspective, focusing on theimplications of Saharan environmental variability and change for human populations bothwithin and outside of the region on timescales ranging from decades to millennia. Thearticle starts by addressing past climatic changes and their impacts on human populations,before moving on to consider present day water resources and rainfall variability in theirlonger-term context; the possibility of a ‘greening’ of the southern Sahara as suggested bysome climate models is also discussed. The role of the Sahara as the world’s largest sourceof airborne mineral dust is addressed in some detail, as are the impacts of dust on climate,ecosystems and human health, as well as the implications of future changes in climate fordust production and the role of the Sahara in the Earth system. The article ends with a discus-sion and synthesis that explores the lessons that may be learnt from a study of the physical andsocial sciences in the Sahara, in particular focusing on what the signature of past environmentaland socio-cultural changes can tell us about human responses and adaptations to climatic andenvironmental change – a matter of great relevance to researchers and policy makers alike inthe context of anthropogenic climate change or ‘global warming’.

Nick Brooks is Assistant Director of the Saharan Studies Programme at the University of East Anglia, and asenior Research Associate at the Tyndall Centre for Climate Change Research. He has a background in thephysical sciences and climatology, and has been working in the field of Saharan geoarchaeology since 1999.Isabelle Chiapello is a CNRS Research Fellow at the University of Lille, where her recent work hasincluded analysis of the factors controlling dust production and transport in the Sahara and Sahel.Savino di Lernia is Professor of Ethnoarchaeology at the University of Rome ‘La Sapienza’, and Directorof the Italian-Libyan Archaeological Mission in the Acacus and Messak (Libyan Sahara). Nick Drake is aReader in the Department of Geography at King’s College, London, specialising in remote sensing and aridzone geomorphology. Michel Legrand is Professor of Atmospheric Sciences at the Laboratoire d’OptiqueAtmospherique at the Universite des Sciences et Technologies de Lille, specialising in the remote sensing ofatmospheric aerosols. Cyril Moulin is a researcher at the CEA/CNRS Laboratoire des Sciences du Climat etde l’Environnement in Gif-sur-Yvette, specialising in African climate and dust production and transport.Joseph Prospero is a Distinguished Faculty Scholar, Professor of Marine and Atmospheric Chemistry,and Director of the Cooperative Institute for Marine and Atmospheric Studies at the Rosenstiel Schoolof Marine and Atmospheric Science, University of Miami. He has worked extensively in the field of atmos-pheric chemistry, and on the transport of African dust to the Caribbean and North America.

The Journal of North African Studies, Vol.10, No.3–4 (September–December 2005) pp.253–292ISSN 1362-9387 print=1743-9345 onlineDOI: 10.1080=13629380500336680 # 2005 Taylor & Francis

Introduction

The purpose of this paper is to review the state of the environmental sciences in the

Sahara and to consider their wider relevance in a variety of contexts. In particular we

focus on the links between the physical and social sciences, and interdisciplinary links

between different research areas. A major aim of the paper is to demonstrate the

relevance of Saharan research in a number of fields for wider global issues such as

climate change and human adaptation. The paper is partly structured as a “narrative

history” of the Sahara, on the one hand in order to address a range of issues relating to

prehistoric, historic, contemporary and future climatic and environmental change, and

on the other to illustrate the relevance of palaeoenvironmental, archaeological and

historical information for studies of twenty first century human-environment

interaction.

We define the Sahara approximately as the region of northern African where

average annual rainfall is currently below 100 mm (Figure 1). This definition

should be interpreted loosely given the large fluctuations of rainfall at the fringes

of the desert on multiple timescales; the Sahara may be viewed as expanding and con-

tracting as a response to global and regional changes in climate that modulate rainfall

at its periphery. In the recent geological past the Sahara thus defined has extended

much further south than at present, and has shrunk to such an extent that it effectively

disappeared. Any consideration of the past or future evolution of the Sahara must

therefore also consider its periphery, particularly at its southernmost edge where

FIGURE 1

ISOHYETS REPRESENTING MEAN ANNUAL RAINFALL IN MM OVER NORTHERN AFRICA

FOR THE TWENTIETH CENTURY

Source: From the dataset of New et al. (2000) and obtained from the Climatic Research Unit. The Sahara isdefined here loosely as the region where mean annual rainfall is below 100 mm.

254 THE JOURNAL OF NORTH AFRICAN STUDIES

the largest variations occur in response to fluctuations in the strength and position of

the monsoon.1

The paper begins with a brief review of our knowledge of how the Saharan

environment has changed in the past, with a particular focus on the early Holocene

pluvial (approximately six to ten thousand years before present) and the subsequent

desiccation of the region. These changes are viewed within the context of global

climate change, a topic currently of serious concern to scientists and policy makers

tasked with developing strategies to mitigate the severity of, and promote adaptation

to, human-induced climate change in the twenty-first century. This discussion of past

changes in the Saharan environment frames the remainder of the paper, which deals

with the legacy of the past, and what we can learn from it.

The discussion of the physical aspects of past environmental change is followed

by a review of how these changes affected human societies. In particular, processes

of adaptation and their consequences are considered.

Moving on from discussions of the past, the present-day environment of the

Sahara is considered in terms of groundwater resources, and the mobilisation and

transport of mineral dust from Saharan sources. Water resources are considered

within the context of settlement and development. Mineral dust is discussed at

some length due to the importance of the Sahara as a global dust source, and the

impact of dust on human health, ecological productivity and global climate. Ground-

water and dust sources are considered within the context of past changes in the

Saharan environment, particularly fluvial activity, which have determined their

nature and distribution.

Current and future climate variability are also considered within the context of

existing rainfall variability and potential future changes in rainfall associated with

changes in the African monsoon.

Finally, issues of environmental change and variability, adaptation and develop-

ment are synthesised. The lack of international research in the Sahara when compared

with other parts of the world is addressed, and the case for cross-disciplinary research

in the Sahara is presented.

Past Environments of the Sahara

The history of the northern African land mass is one of dramatic changes in the

physical environment, of oscillations between arid phases during which much of

the sub-continent is effectively uninhabitable, and humid episodes that transform

the desert regions of the Sahara into a fertile landscape of lakes and savannah. On

timescales of tens of thousands of years these changes are driven by glacial cycles,

with glacial conditions in the northern hemisphere being associated with cold, arid

conditions over northern Africa.2,3,4,5,6 During the last glacial maximum (LGM)

some 21 thousand calendar years before present (21 ka), the Sahara desert covered

a much larger area than today, as apparent from the dating of fossil dunes some 58south of the present extent of mobile dunes.7 A combination of factors leads to

increased aridity during periods of glaciation, including reduced atmospheric

THE CLIMATE-ENVIRONMENT-SOCIETY NEXUS IN THE SAHARA 255

moisture availability, decreased solar heating of the land surface, and large-scale

changes in atmospheric and oceanic circulation.8,9

The Pleistocene

Over the past 1.65 million years, approximately corresponding to the Quaternary

period, there have been some 17 glacial cycles, each lasting approximately 100

ka10 and it is the generally held view that glacials are arid and interglacials humid.

However, determining the effects of Pleistocene11 climate change on the Sahara is

problematic due to the paucity of organic deposits. Organic remains, such as palaeo-

lake sediments, provide important information regarding arid-humid transitions.

However, these deposits are rapidly deflated by the wind during arid periods and

few ancient deposits have managed to endure the multiple episodes of Pleistocene

aridity. Because of these factors the number of lacustrine deposits12 in the Sahara

that have been dated to the last interglacial or older is no more than five. However,

it is possible to glean some useful information from these studies. Sediments of

Lake Megafezzan13 provide one of the longest records, with evidence for lacustrine

episodes at 380, 240, 128, 118(þ272 20), 74(þ232 16), 47(þ172 13), 30 ka

and from 14+ 1.7 to about 3 ka.14 This record of humidity conforms with the

view that interglacials are associated with humid periods but also indicates that epi-

sodic humidity occurs during glacial cycles. A similar picture emerges if we compare

these lacustrine episodes to the few others identified in the Sahara. A large lake

existed in the Basin of the Tunisian Chotts at 150, 92, 75(þ72 6) and

42.4+ 2ka,15 there is evidence for lacustrine conditions at 75+ 7 ka in southern

Algeria,16 and there are indications of humidity between 71 and 87 ka in southern

Egypt and northern Sudan.17 Some of these humid episodes are synchronous in

two or more regions of the Sahara (e.g. �128, �74 and 422 47 ka), while others

are not (e.g.�155 ka), although this is not unexpected due to the paucity of the lacus-

trine record and uncertainties in dating. Encouragingly, studies of alluviation on the

northern fringes of the Sahara agree with the data from the central Sahara. Northern

Libya was subjected to fluvial activity indicative of a wetter climate at 125+ 15,

76+ 4, 42+ 5.1, 23.2+ 2.8 and 12.5+ 1.5 ka,18 while in southern Tunisia incision

and subsequent alluviation occurred at 47+ 12 and from about 8+2 ka.19 Thus in the

north and central Sahara there is clear evidence for brief humid episodes occurring

during glacials, as well as in interglacials.

The Early Holocene Humid Episode

Summer insolation over northern Africa reached a maximum at the beginning of the

Holocene period, around 10 ka,20 by which time humid conditions had been

established in the Sahara. This so-called Holocene Climatic Optimum saw a greening

of the Sahara as the monsoon rain belt shifted hundreds of kilometres to the north,

leading to the formation of numerous lakes in areas that are now hyper-arid, and

the development of a mosaic of savannah and woodland throughout much of the

Sahara.21,22,23,24 It is during this period that the Sahara was reoccupied by human

populations, who initially survived through hunting and gathering, exploiting the


abundant humid climate fauna and flora of the then-moist Sahara, as discussed in

more detail below.

The Holocene Climatic Optimum was, however, interrupted by a number of arid

episodes, some of which appear to have been associated with transient changes in

climate at the global or hemispheric scale. The most important such events were

episodes of cooling in the North Atlantic, recurring on millennial timescales. Bond

et al.25 identify a number of such Atlantic cooling events dated at 11.1, 10.3, 9.4,

8.1, 5.9, 4.2, 2.8 and 1.4 ka. The 8.1 ka event coincides, within the envelope of

uncertainty associated with the dating of such episodes, with a period of aridity

lasting the order of centuries in the Sahara.26 The 5.9 ka event occurs around the

time of a dry episode evident in many, although not all, Saharan lake records,

marking the beginning of a shift towards more permanent aridity.27,28

The Desiccation of the Sahara

A number of data indicate that aridity increased in the Sahara after an arid episode

occurring around 6 ka.29 The subsequent desiccation was not a smooth, continuous

process characterised by a linear response of the monsoon to steadily declining

solar heating or insolation. Instead, it appears that desiccation occurred in a stepwise

fashion, with one or more episodes of abrupt drying.30,31,32 It is most likely that the

drying of the Sahara was the result of a complex interaction between declining solar

insolation, transient climatic perturbations such as the cold arid episode at around

6 ka, and dynamic vegetation-atmosphere feedbacks involving the retention and

recycling of moisture.33

The earlier �8 ka cold arid episode occurred at a time when solar insolation was

sufficiently strong to drive a recovery in the monsoon and in vegetation cover, once

this transient climatic perturbation had passed. However, insolation was considerably

weaker by �6 ka, and may have been insufficient to drive a full recovery of the

coupled vegetation-monsoon system after it had been disrupted by the Atlantic

cold episode identified around this time34. This transient climatic perturbation may

well have been the trigger for the final desiccation of the Sahara, occurring at a

time when the monsoon system was at least partially sustained through moisture

retention and recycling by vegetation systems that originally developed as a result

of strong insolation driving a vigorous monsoon. Evidence for abrupt desiccation

towards 5 ka in a number of Saharan regions35,36,37,38,39 suggests the final collapse

of remaining vegetation systems around this time, perhaps as solar insolation

crossed a threshold below which vegetation-atmosphere interactions could no

longer sustain rainfall, or possibly as a result of further transient perturbations from

which vegetation could not recover in a low-insolation regime.

By around 5 ka desiccation associated with the southward retreat of the monsoon

system was well established; this process culminated in the formation of the present-

day Sahara Desert. The timing and speed of this process of desertification varied from

place to place, and was mediated by geography, topography, hydrogeology and the

nature of regional climatic systems. For example, in the Wadi Tanezzuft, in the

Libyan Fezzan, the depletion of soil water reserves was not completed until about

3.5 ka, and fluvial activity persisted until around 2.7 ka40 apparently, this was due


to the combined effects of the large rainfall catchment area (the Tassili Mountains.)

and of the dune systems bordering the fluvial valley, which acted as water reservoirs.

The climatic and associated environmental changes described above were not

restricted to the Sahara. Abundant evidence indicates that the broad pattern of wet

conditions in the early to middle Holocene, interrupted by arid episodes and followed

by a process of desiccation starting around 6 ka and accelerating around 5 ka,

prevailed throughout many subtropical and extra-tropical northern hemisphere

regions.41 Our understanding of past environmental change in the Sahara thus

represents but one component in our wider understanding of global climatic and

environmental change which, when coupled with results from other regions,

enables us to develop a deeper understanding of the Earth System and of human

responses to environmental change. Such an understanding is necessary if we are

to confront issues such as long-term climate variability and anthropogenically-

driven climate change, processes which can have a profound impact on

socio-economic development.

Linking Past Environmental and Cultural Change

The Sahara occupies a privileged position in studies of human-environment inter-

action, as a consequence of the clear climatic signals in Saharan palaeoenvironmental

records, and the associated evidence of cultural change during key transition periods

between arid and humid conditions. In many other geographical regions archaeology

has been bedevilled by arguments over environmental determinism, leading to a

reluctance to consider the role of environmental change in mediating cultural trajec-

tories. In contrast, archaeologists working in the Sahara have by and large appreciated

the central role of environmental change, without falling into the trap of neglecting

other important factors in the development of human societies. In one sense, we

may view the Sahara as a laboratory of human response to environmental change,

given the overwhelming nature of the climatic changes that have affected the

region throughout the relatively archaeologically accessible Holocene. As such, the

Sahara can tell us much about processes of adaptation, the understanding of which

is crucial for the formulation of policies designed to address the impacts of future

climate change, particularly in the developing world.

Human Occupation of the Sahara in the Pleistocene

Human occupation of the Sahara has been limited by the availability of water, with

little or no occupation away from the Mediterranean coast and the Nile Valley

during arid phases associated with glacial epochs. A notable hiatus in human

occupation, at least in the central Saharan regions, appears to be related to the Late

Quaternary glacial: evidence for Upper Palaeolithic sites is very rare, and the latest

clear evidence of human occupation during the Pleistocene (the section of the

Quaternary predating the beginning of the Holocene some 10,000 years ago) dates

back to the Aterian, from 90 to 60 ka.42 A human presence around 30 ka in the

Wadi al-Shati region in Libya has also been hinted at, although the evidence is

equivocal.43 Nevertheless, the spread of modern humans during the Pleistocene


was facilitated by the existence of migration corridors, in particular the Nile Valley, in

which water was available even during the least hospitable phases of this epoch.

Evidence of Pleistocene occupation in the central Sahara is almost invariably

composed of scatters of lithic tools, most of them deprived of any faunal association

or stratigraphic context. Even in the exceptional contexts of the Acacus mountains,

well preserved Pleistocene sites have not been identified, although Uan Tabu44 and

Uan Afuda45 have yielded assemblages of worked stone, dated via optically stimu-

lated luminescence (OSL) and thermoluminescence (TL) methods. The archaeologi-

cal situation is much more fortunate in the Eastern Sahara, thanks to research by

Wendorf and associates, for example in Wadi Kubbanya, Bir Tarfawi and Bir

Sahara.46,47 Leaving aside technological developments and site organisation, one of

the most important issues to be placed on the archaeological agenda is the spreading

of modern humans in Africa north of the equator. The Sahara, together with the

Nubian corridor, is believed to be a major area for the dispersal of anatomically

modern humans, but coordinated international research in this field is lacking.

Furthermore, it is essential to build a detailed scheme of Late Quaternary climatic

trends, as a prerequisite to analyse and track possible Homo sapiens migration

paths. Recent archaeological research emphasises the role of the Aterian complex

as a possible technological indicator of the presence of modern humans in the

Sahara.48,49 ‘Continuistic’ views, based on (single and isolated) radiocarbon determi-

nations pointed in the past to a very late presence of Aterian sites in Morocco and

other North African contexts, suggesting a direct relation between Aterian peoples

and Upper Palaeolithic North African contexts. Recent research, specifically in

Egypt and Libya, pushes back this technological phase to 140–60 ka ago, and

underlines, at least in the central Sahara, a break in human occupation related to

the expansion of the ‘Ogolian’ desert.

Human-Environment Interaction in the Sahara in the Early-Mid Holocene

The archaeological evidence indicates that the latest phase of human occupation in

the central Sahara commenced at the beginning of the Holocene, around or soon

after 10 ka.50,51 These early Holocene communities consisted of groups of hunter-

gatherers, which in many cases practiced a fairly sedentary lifestyle, exploiting abun-

dant locally available food resources in the form of wild fauna and flora.52,53,54

It is worth noting that the scant palaeo-anthropological evidence (from Uan Afuda

and Uan Muhuggiag in the central Sahara of Libya) points to sub-Saharan affinities.55

This fits with more recent human remains from the Egyptian oasis, which indicate a

similar affinity on the basis of dental analysis.56 These findings support the hypothesis

of a northwards movement of human populations as they followed the monsoon rains,

which strengthened and penetrated further north into the Sahara at the beginning of

the Holocene. The gap between the beginning of the humid period in the Sahara

after the last glacial maximum (ca. 15–13 ka) and the appearance of the first

Holocene occupation sites might be interpreted as a consequence of the time taken

for vegetation and fauna to recolonise hyperarid environments.57

More cautiously, the first genetic data on Saharan palaeo-populations also indicate

a sub-Saharan affinity.58,59 Evidence for a southern provenance of the first Holocene


Saharans might also be seen in from rock art, although the subjective nature of such

interpretations must be recognised: in the Tassili60 and Acacus massifs61 depictions of

figures with what appear to be black African features have been interpreted as

indicating the possible presence of populations originating in sub-Saharan regions.

Dating of this material is controversial, but an Early Holocene attribution for the

so-called “Round Head” style of painting, as usually claimed in the literature,62,63

may reinforce the archaeological and palaeo-anthropological evidence.

It is during the early Holocene that we find the earliest (although still somewhat

controversial) evidence for cattle domestication in the eastern Sahara, the driest

part of North Africa at this time.64,65 Reviewing the literature on the development

of cattle domestication in Africa, Marshall and Hildebrand argue that this was a

local development based on the exploitation of indigenous species, and driven by

the desire of hunter-gatherers to increase the predictability of their food supply,

perhaps initially in order to ensure the availability of cattle for slaughter during

ritual festivals.66 They suggest that this development happened in the eastern

Sahara as a result of the greater environmental variability in this region, and hence

the greater need for intervention to ensure predictability of food supply, when

compared with other Saharan regions. Citing other work,67 Marshall and Hildebrand

conclude that during the frequent periods of drought affecting the eastern Sahara,

cattle would have represented ‘a more reliable resource than plants because their

populations are maintained through movements that exploit local differences in

topography, vegetation, and rainfall’.

According to Hassan68 and Marshall and Hildebrand, citing numerous sources,

cattle pastoralism spread westwards through the central Sahara in an uneven

fashion from around 7 ka, where it coexisted with hunting and gathering. Di Lernia

suggests that the diffusion of cattle-based culture in the Sahara occurred in response

to short, abrupt dry events during the 8th to 6th millennia before present (BP).69 The

�8 ka arid episode described above may have played a key role in the spread of

pastoralism. Di Lernia and Palombini suggest that the dry interval at approximately

8 ka ‘probably favoured the integration of cattle herding within foraging commu-

nities’ in the Libyan Sahara.70 If this episode also marked a shift from year-round

to seasonal rainfall as has been suggested by some authors,71 cattle herding would

have represented an appropriate adaptation in order to enhance food security in a

more seasonally variable environment.

In the Acacus mountains of Libya, semi-permanent settlements in lowland areas

gave way to seasonal migration during the late 7th millennium BP, when climatic

conditions deteriorated during a ‘severe, abrupt dry interval lasting several centu-

ries’.72 This new pattern of subsistence involved increased use of mountainous

areas in the dry season. Di Lernia and Palombini73 suggest that sheep and goats

(most probably introduced from the western Asia in the early 7th millennium

BP) were tended in highland regions from late winter to early spring in order to

reduce pressure on lowland pasture land, which was set aside for cattle in the

dry season; in the study areas in the Libyan Fezzan on which they focus,

cattle remains are found predominantly around the interdune palaeo-lakes in the

lowland ergs.


The nature of the spread of cattle herding through the Sahara is still a controversial

issue. According to Schild and Wendorf, this occurred during wet periods, allowing

people to move with their livestock westward.74 Conversely, Hassan75 and di

Lernia76,77 argue that migration by cattle-herding groups was stimulated by aridity.

They argue that this explanation is more compatible with the archaeological

record, which consists of scattered and isolated contexts throughout the Sahara.

Dates associated with these contexts also suggest discrete episodes of migration

spanning short periods,78 implying rapid, intermittent movements of small groups

of herders, colonising new and unfamiliar environments as they were forced to

move in search of water and pasture during arid crises.

Social and Cultural Responses to Mid-Late Holocene Desiccation

The process of desertification at around 5 ka appears to have been rapid in at least

some parts of the Sahara,79 and was associated with profound changes in human

societies. Increasingly harsh conditions would have had a profound effect on

culture and belief systems; it has been suggested that an increased diversity in the

subject matter of rock engravings and paintings, with a greater emphasis on images

relating to sexuality and fertility, and on ‘enigmatic’ or fantastical beings, dates to

the period of climatic deterioration, ‘when concerns over human and animal fertility

may have become acute’.80 In reference to the Fezzan region of Libya, Mattingly and

associates81 write that ‘Human activity was not cut off at a stroke, however, but may

have become more focussed on specific locations, with the rock art representing a

more sophisticated dialogue between people and a powerful spirit world, bordering

on formalised religion’. It has been suggested that this kind of relationship dates

back to the emergence of the Pastoral cattle cult at 6.4–6.0 ka, when non-utilitarian

slaughtering of precious livestock and the flourishing of an artistic tradition focused

on cattle appear to represent the first evidence of ritual relationships between human

populations, the physical environment, rainfall and ‘divinities’.82,83,84

Human responses to the desiccation of the Sahara were spatially heterogeneous

and mediated by geography, resource availability and local hydrogeological

responses. In the Libyan Fezzan two types of response may be identified, according

to di Lernia and Palombini.85 In higher elevation regions cattle herding almost com-

pletely disappeared after 5 ka. This was replaced by highly mobile pastoralism based

on sheep and goats and involving large-scale year round movement in order to exploit

remnant water and pasture, the origins of a nomadic lifestyle that persists to this day.

In contrast, lower elevation regions were characterised by increasing settlement in

relict oases, associated with sedentism and more intensive exploitation of local

resources. Di Lernia and Palombini characterise the former as a ‘light’ approach to

landscape exploitation, associated with a relative egalitarianism and sustainable use

of resources, and the latter as a ‘hard’ and ultimately unsustainable approach to the

landscape leading to possible degradation of the landscape, conflict (possibly over

land rights) and increased social stratification.

Behavioural adaptations focusing on resource extraction were associated with

profound changes in social organisation, as indicated by archaeological studies of

funerary monuments and burial practices. Di Lernia and Manzi86 describe the


period centred around 5 ka as the ‘hinge’ between different pastoral cultures in the

Wadi Tanezzuft in the Libyan Fezzan, which they term Middle and Late Pastoral. It

is around this time that stone funerary monuments are associated with human

burials in the Wadi Tanezzuft; previously these had been reserved for ritual animal

burials. This process is reflected in other central Saharan regions. A review of the

archaeological work by Sivilli87 indicates that in southwestern Libya and northeastern

Niger, the period 6.4–5.9 ka is represented exclusively by faunal burials; 5.8–4.9 ka

by faunal, human and ‘empty’ burials (i.e., cenotaphs); and 4.8 ka onwards exclusively

by human burials. Di Lernia andManzi88 interpret this innovation in funerary practices

as being associated with ‘emerging figures within the pastoral group. . . [representing]the first evidence in the area of a process of increasing social stratification’.

Di Lernia et al. (2002) interpret the evolution of funerary monuments and

practices and settlement patterns as evidence of major changes in population in the

Wadi Tanezzuft, and by implication in the greater central Sahara region. Funerary

monuments would have served a dual purpose in a landscape occupied by increasing

numbers of pastoralists, on the one hand acting as foci for gatherings of related social

groups, and on the other serving as markers of boundaries, territories or zones of

influence, asserting relationships between clan groups and the landscape.

In the Fezzan, it appears that climatic desiccation was associated with inward

migration, increased population density, changes in religious beliefs and practices,

social stratification and a more territorial approach to the landscape, as well as diver-

ging adaptations to facilitate resource extraction in a more hostile landscape. In

certain localities these changes appear to have provided the preconditions for the

emergence of complex urban societies and the formation of entities resembling

states, catalysed by the final desiccation of most of the landscape soon after

3 ka.89,90 The onset of conditions equivalent to present aridity might be viewed as

a threshold that stimulated a step change in social organisation as population

groups adopted new techniques to access water and utilise scarce productive land.

In the Wadi Tanezzuft, the depletion of soil water reserves was not completed until

about 3.5 ka, and fluvial activity persisted until around 2.7 ka.91 Further to the east

in the Wadi al-Hayat, there is evidence that springs dried up around or before 3 ka.92

These late dates for fluvial and spring activity correspond approximately to the

early stages of the Garamantian civilisation, which dominated the Fezzan between

about 1,000 BC and AD 700, a period ‘notable . . . for the local evolution of urbanism,

irrigated agriculture and writing’.93 The Garamantes are described by the Greek

historian Herodotus, writing in the fifth century BC, and later represented a regional

challenge to Roman aspirations in the central Sahara and North Africa. The Garaman-

tian ‘capital’ of Garama (or Old Germa, situated within 2 km of the modern town of

Germa/Jarma) was located in the Wadi al-Hayat (also known as the Wadi al-Ajal),

with Garamantian settlements also located in the nearby Wadis, al-Shati, Barjuj

and Tanezzuft.94,95,96,97,98

By 3.1+ 0.125 ka the springs had dried up and surface water was either very

scarce or absent at the base of the escarpment forming the southern boundary of

the Wadi al-Hayat.99,100 However, the water table was probably very near the

surface at the base of the wadi, as is evidenced by the fact that recent archaeological


excavations uncovered a well in the early Garamantian phase of Old Germa sunk to a

depth of 70 cm, indicating water at very shallow depths.101 Archaeobotanical

evidence indicates that irrigated agriculture was introduced soon after the desiccation

of the springs, in the early part of the first millennium BC,102 presumably in response

to a lack of water caused by the fall in the water table. There is no direct evidence for

irrigation systems at this time, but the water table was near the surface in Germa and it

is possible that wells were used to tap this resource. The earliest irrigation systems are

foggara which were used to tap the elevated water table at the base of the escarpment.

Archaeological evidence suggests that they were probably introduced by the final few

centuries BC, and definitely before the fourth century AD.103 Interestingly this roughly

corresponds with archaeobotanical evidence for the intensification and diversification

of agriculture involving the introduction of a farming system that utilised both winter

and summer crops.104 Foggara would also have allowed the extensification of

agriculture at a similar time. In combination these developments could have lead to

an increase in agricultural production, and there are likely to be strong connections

between this and the rise of the Garamantes as a major political power in the

central Sahara.105

The Garamantian culture appears to have been the result of local innovation, the

outcome of a process of increasing social complexity among the pastoral groups of the

Fezzan. Referring to the Wadi Tanezzuft, di Lernia et al.106 write that ‘In a certain

sense, Late Pastoral people became the Garamantes’. This conclusion is supported

by the work of Mattingly et al.,107 who find some of the latest Pastoral lithics and

pottery in early Garamantian forts along the southern edge of the Wadi al-Hayat.

In the Fezzan we thus see the ultimate expression of the human response to deserti-

fication in the development of a significant urban civilisation.

The emergence of the Garamantian polity, largely driven by changes in water

availability and geographic serendipity, is not the only example of increased social

complexity leading to the emergence of what we might call a ‘state-level society’

in a time of increasing aridity. The earlier development of Dynastic Egypt has also

been interpreted at least in part as a result of social responses to environmental desic-

cation.108,109,110 Palaeoenvironmental evidence indicates increasing aridity to the east

and west of the Nile Valley during the 6th millennium BP, with a final desiccation of

most of this region in the late sixth millennium BP, when the early Dynastic state

emerged.111,112,113,114 Excavations at Hierakonpolis are indicative of attempts to inte-

grate animal herds with a sedentary lifestyle at the beginning of the Dynastic period,

suggesting that mobile groups were forced or encouraged to settle permanently in the

Nile Valley.115 Midant-Reynes116 explicitly links local increases in population

density with desertification at the end of the Predynastic period. Archaeological

evidence provides abundant support for models of cultural evolution involving

increased social stratification and the concentration of political power as a response

to increased population densities resulting from a decrease in available productive

land, and associated migration/increased sedentism as a consequence of environ-

mental desiccation.117

To summarise, linked environmental and cultural change in the Sahara is not

simply a matter of poverty and famine resulting from aridity. During the Pastoral


Neolithic, in the 7th and 6th millennia BP, the Sahara gave birth to one of the great

cultures of antiquity, the Saharan cattle-based culture. Material cultural, occupation

sites, food security, rock art, ideology and funerary practices are all part of an extra-

ordinary culture, the nature of which was driven by a profound relationship with the

physical environment. This culture had to cope with numerous fluctuations in

resource availability associated with climatic variability on a range of timescale,

some of which were extremely severe. This cultural tradition only came to an end

with the final desiccation of the Sahara around 5 ka. However, the pastoral societies

of the Holocene Sahara did not vanish into the sand: the migration of pastoral groups

to refugia such as the Nile Valley, the Sahel, and the Saharan highlands and relict

oases made a vital contribution to many subsequent African cultures, including

Pharaonic Egypt, the Garamantes, the present-day cultures of the Sahel, and probably

the modern day Berber and Tuareg populations.118,119,120,121 In a more general sense,

we may claim that the essence of this cattle-based culture spread from the Sahara to

much of sub-Saharan Africa, shaping the history of the entire African continent.

Water Resources and Human Settlement

As has been the case throughout the Quaternary period, water resources are today the

limiting factor in human settlement and development. Increasing populations and

associated urbanisation and economic development are placing greater demands on

water resources throughout the Sahara. These resources may also be affected by

climate change in the near future, for example with elevated surface temperatures

further enhancing evaporation, and changes in meteorological patterns resulting in

changes in the distribution of rainfall.

Modern permanent settlements in the Sahara are situated where water is available

either from rainfall generated as a result of topography or the intrusion of extra-

Saharan weather systems (such as in highland and coastal regions), or at oases

where groundwater occurs at or near the surface in local topographic minima.

Water resources may therefore be affected by rainfall variability and changes in

groundwater levels.

Groundwater and Human Settlement

Most inland Saharan settlements rely almost exclusively on groundwater. Modern

pumping and irrigation technology has enabled the expansion of agriculture in

many locations, such as the Wadi al-Hayat in the Libyan Fezzan, which provides

an object lesson in the interaction of water resources, human populations and techno-

logical innovation. Here a downward trend in population density started in the first

millennium AD with the decline of the Garamantian culture. The cause of the

demise of the Garamantes is not known with any certainty, although declining

water levels either as a lagged response to the climatic desiccation of the region, or

as a result of over exploitation by local populations, has been suggested.122

However, other explanations, such as a decline in trade in the later years of the

Roman empire, should also be considered.123 Whatever the cause of the demise of

the Garamantes, it appears that groundwater levels declined between Garamantian


times and the end of the second millennium AD.124 In recent decades the socio-

economic decline of the Wadi al-Hayat has been reversed through the use of irriga-

tion, dependent on the pumping of groundwater from increasingly greater depths.

However, increases in population, the expansion of agriculture, and the process of

urbanisation have resulted in a lowering of groundwater levels since the 1970s,

exceeding 20 m in some locations according to local informants. White et al.125

used satellite imagery and aerial photography to examine changes in vegetation

distributions between the 1950s and the end of the twentieth century in the Wadi

al-Hayat, and found a shift in the vegetated zone resulting from the expansion of

agriculture at the southern edge of the wadi, and the die-back of vegetation in unirri-

gated northern areas. The most obvious adverse impacts are on stands of date palms,

many of which no longer survive without human intervention, and have given way to

more drought or salt-tolerant species.

Such studies raise questions of sustainability: can settlements in central Saharan

regions continue to expand given their reliance on fossil water reserves that are no

longer replenished by rainfall? Furthermore, how sustainable are schemes such as

the Great Manmade River project in Libya? Water use in Libya has been estimated

at some eight times higher than its renewable water resources, an extreme example

of a situation faced by all the countries of the southern Mediterranean coast which

means they are likely to become much more dependent on food imports (essentially

a means of importing water) in the future.126 Pressure on water use is likely to be

exacerbated by climate change; in the southern Mediterranean reductions in rainfall

of some 20–25 per cent and an increase in average annual temperature of 2–

2.758 C by the 2050s have been forecast by global climate models.127 Increases in

temperature will lead to greater evaporation, compounding water scarcity resulting

from reductions in rainfall. Such developments are likely to increase pressure on

fossil water reserves from the Saharan aquifers. Any expansions in the tourism

sector will also increase water demand. Given the likely increased use of fossil

groundwater, a better understanding of groundwater reserves and dynamics, and

the nature, capacity and behaviour of the Sahara’s large subterranean aquifers is

highly desirable. Such aquifers will not reach equilibrium to keep pace with extrac-

tion, which will result in the formation of ‘cones of depression’ near populated

areas or areas of intense agricultural activity. Ebraheem et al.128 conclude that the

Nubian Sandstone Aquifer under southwest Egypt does not exist in a steady state

and is still responding to past humid conditions, and estimate that the planned extrac-

tion of 1,200 million m3/year in the East Oweinat area could result in a drawdown ofup to 200 m relative to 1960s levels within 100 years, with the cone of depression

extending to Dakhla and Kharga oases.

Rainfall Variability

Given the extreme variability of precipitation even in the wetter parts of the Sahara,

the concept of ‘mean annual rainfall’ has little practical meaning. However, it is worth

pointing out that precipitation in some parts of the Sahara is sufficiently high for local

populations to rely at least partially on rainfall, and to be adversely affected during

periods when rainfall is anomalously low. Keenan129 describes how the Kel


Ahaggar Tuareg have developed strategies to cope with drought, and how drought

coupled with the social upheavals of Algerian independence led to hardship in the

1960s.

Drought is a concept that is also familiar to the inhabitants of the far west of the

Sahara.130 In the inland regions of the disputed territory of Western Sahara vegetation

is relatively abundant, sustained by occasional summer rains representing the extreme

northerly penetration of the African Monsoon beyond 208 N, and similarly scarce

rainfall associated with the Atlantic Westerlies. Despite the relative abundance of

rainfall and vegetation in this region, the lack of surface water (e.g. in the form

of oases), as well as the ongoing political conflict with Morocco and the possibility

of a resumption of hostilities, mitigates against permanent settlement. Nonetheless,

the region is used by nomads whose principal source of water takes the form of

milk from their animals.

The Sahelian region, situated at the southern margin of the Sahara, has experi-

enced one of the most persistent and severe changes in climate during the period of

meteorological records.131 This consisted of a multi-decadal scale drying trend

commencing in the 1960s and persisting into the 1990s, with severe droughts in

the early 1970s and 1980s, and some amelioration in recent years132,133 (Figure 2).

It is now well established that rainfall in the Sahel is driven largely by patterns of

global surface temperature, with the temperature of the Indian Ocean playing a key

role.134 While rainfall data for the Sahara are sparse, there is no evidence for any

comparable trend north of the Sahel-Sahara transition zone in those regions of the

Sahara where topography results in non-negligible rainfall. For example, both

annual and summer rainfall totals for Tamanrasset in the highlands of the Algerian

Sahara (home of the Kel Ahaggar Tuareg mentioned above) reveal a wet period in

the 1950s, as in the Sahel, but no long-term drying trend in subsequent decades

FIGURE 2

SPATIALLY AGGREGATED ANNUAL RAINFALL ANOMALIES (IN STANDARD DEVIATIONS)

REPRESENTING THE REGION 108 – 208 N; 258 W – 308 E, ROUGHLY CORRESPONDING TO

THE SAHELIAN ZONE. ANOMALIES ARE CALCULATED WITH RESPECT TO THE MEAN

FOR THE ENTIRE SERIES (1901 – 1998)

Source: From the dataset of New et al. (2000).


(Figure 3). Instead, the Tamanrasset rainfall record consists of quasi-periodic

variations of some 10–15 years, with extreme variations in annual rainfall from 0

to 160 mm. On average, approximately half of the rainfall at Tamanrasset occurs in

the summer, although variations between individual years are great. For example,

the period from July to September saw no recorded rainfall in 1926 and 1973,

whereas all of the recorded rainfall during 1999 occurred in these months.

The near to medium term future of the Sahara is uncertain in climatic terms. While

there are no indications that the region will experience a shift to humid conditions

comparable to those existing in the early Holocene, a number of climate modelling

studies suggest that anthropogenic climate change may be associated with a strength-

ening of the African summer monsoon and an intensification and northerly displace-

ment of monsoon rains, leading to wetter conditions in the northern Sahel and

southern Sahara.135,136,137 Other modelling work suggests that the Sahara may shift

northwards,138 implying that even if rainfall increases in the south of the Sahara, it

may decline in regions near the Mediterranean coast. The suggestion that the northern

Sahel and the southern margins off the Sahara may become wetter is consistent with

recent observations indicating a greening of the Sahel explained partly by increased

rainfall and partly by increased vegetation cover due to human activity.139,140

However, rainfall still remains below the high levels of the mid-twentieth

century,141 and interannual rainfall variability and drought still pose considerable

problems for Sahelian societies, as evident from the famine that is unfolding in

FIGURE 3A

TWENTIETH CENTURY ANNUAL RAINFALL TOTALS FOR TAMANRASSET, OVER WHICH IS

SUPERIMPOSED A 5-YEAR MOVING AVERAGE (SOLID LINE)

Source: The Climatic Research Unit.


Niger at the time of writing (July 2005). The impact of climate change on Sahelian

climate is likely to remain unclear for some time, until additional observational

and modelled data are available.

A shift in the monsoon rain belt would have profound implications for Saharan

societies. Areas that are not currently viable for human populations would become

available for pastoralism, and existing marginal areas might become viable for

rainfed agriculture. However, an intensified monsoon would also be associated

with more frequent flash floods and the likely spread of water borne diseases.

Recently high rainfall in the Sahel has enabled locusts to thrive, resulting in the

devastation of crops and food insecurity in a number of countries in 2004 and

2005.142 Grolle143 describes how heavy rainfall in the Sahel in 1953 triggered

famine. While an intensification of the monsoon would undoubtedly bring benefits

in terms of pastoralism and agriculture, the consequences would not necessarily be

wholly benign.

Extreme rainfall variability on different timescales is a fact of life in on the

margins of the Sahara; a failure to fully appreciate this fact was a factor in the

expansion of agriculture into historically marginal areas along the Sahel-Sahara

boundary in the 1950s and 1960s, which were anomalously wet when compared

with the twentieth century rainfall record as a whole. As sedentary agriculturalists

expanded northwards, mobile pastoralists were pushed into more marginal desert

areas. The vulnerability of both populations to drought was significantly increased,

FIGURE 3B

TWENTIETH CENTURY RAINFALL TOTALS FOR TAMANRASSET FOR THE MONTHS OF

JULY, AUGUST AND SEPTEMBER (THE WETTEST MONTHS IN THE SAHEL), WITH 5-YEAR

MOVING AVERAGE

Source: The Climatic Research Unit.


and the combination of elevated vulnerability and severe rainfall deficits in the early

1970s resulted in widespread famine and the collapse of livelihoods and social

systems, representing a significant discontinuity in Sahelian developmental trajec-

tories from which the region is still recovering.144

Keita145 describes the droughts of the 1970s and 1980s as a contributory factor in

the development of internal conflict in Mali, where it undermined the pastoral

livelihoods of the semi-nomadic Tuareg, causing large numbers of them to find

refuge in camps or urban areas where they experienced social and economic

marginalisation.146 Many Tuareg migrated to neighbouring countries, and some

young men become involved in conflicts throughout North Africa and the Middle

East, in which they acquired considerable military experience, eventually returning

to Mali having to face unemployment and marginalisation, creating the conditions

for the ‘Second Tuareg Rebellion’ in 1990. These conditions were exacerbated by

a history of mistrust between the Tuareg and post-independence governments, the

lack of available livelihoods and social support networks for returning migrants (as

a result of previous drought and conflict), continuing drought and associated

competition for resources between nomadic and settles peoples, and the flooding of

the region with small arms as a result of conflicts in neighbouring countries, particu-

larly Western Sahara. Keita147 describes the conflict between the nomadic Tuareg and

the settled communities as ‘not so much an ethnic or racial issue as an economic one’

highlighting ‘the economic dimensions of the problem in the north: communities

were fighting for scarce resources and jealously insisting that others were not

preferred.’ This pattern of marginalisation, drought and conflict has been repeated

throughout the Sahel, and demonstrates that climate variability and change can

combine with other factors to cause conflict, particularly in marginal environments

where populations are facing a number of different environmental, social, economic

and political stresses.

Any future increase in rainfall in the northern Sahel and southern Sahara will

be associated with the risk of unsustainable agricultural expansion if longer-term

climatic variability is not considered in development policies. A modelling

study by Maynard et al.148 suggests that the Sahel may be less prone to

drought in a world characterised by low to moderate levels of anthropogenic

greenhouse warming, associated with atmospheric greenhouse gas concentrations

near current values. However, other studies suggest that the greater levels of

greenhouse warming likely to result from the continued intensive use of fossil

fuels into the latter half of the twenty-first century may result in global

temperature patterns associated with drought conditions in northern

Africa.149,150 Any expansion of agriculture and settlement into areas made

newly productive by a strengthened monsoon may ultimately result in an exacer-

bation of vulnerability as in the 1950s and 1960s, followed by drought and

famine as conditions deteriorate, as they did after the 1960s. The interaction of

naturally occurring drought hazard and socially constructed vulnerability during

the twentieth century holds important lessons for economic development and

agricultural activity in a region where climatic variability on a variety of time-

scales is the norm.


Non-hydrological Hazards and Urbanisation

While the availability of water must remain the principal consideration in economic

development and settlement expansion, human health and comfort are also affected

by other environmental factors. In the Sahara, strategies for coping with extreme

heat, and also with dust storms, are also of great importance. As mentioned above,

climate change is likely to result in higher average surface temperatures, which in

turn will lead to more frequent heat extremes.151,152

A further instance of what may be termed “maladaptation” associated with

population expansion and the associated processes of economic development and

urbanisation is the move away from traditional architecture. Traditional building

materials and architectural styles are well adapted to the extreme desert environment,

particularly in terms of temperature regulation. Historically, settlements have been

built of mud brick, an inexpensive and readily available material that is an excellent

insulator, and have incorporated convection chimneys to encourage the circulation of

cool air, and covered walkways to shelter their inhabitants from the sun. The enclosed

nature of these settlements also affords some protection from dust. However, many of

these settlements are now being abandoned in favour of generic modern towns whose

construction pays little or no attention to the particular hazards associated with a

desert environment. Streets are now open to sun and dust (partly to accommodate

vehicular traffic), and poorly insulated modern buildings fitted with air conditioners

have replaced the subtle architecture of the traditional towns. In the town of

Ghadames in western Libya, the inhabitants of a traditional town (also a UNESCO

World Heritage Site) moved to an adjacent, purpose built modern town in the

1980s. In the summer many return to the old town, where they maintain their old

houses and gardens, as it is more comfortable than the modern settlement.

The challenge of urbanisation in the Sahara is to blend the most desirable elements

of traditional architecture with modern technology, providing comfort in an extreme

environment while ensuring access to modern amenities. Greater use of mud brick

instead of concrete and cement (the production of which is energy intensive) would

reduce both construction costs and greenhouse gas emissions, assisting both

adaptation to and mitigation of anthropogenic climate change. Fathy,153 working in

the context of development in rural Egypt, has advocated a greater emphasis on

such traditional building materials and styles in order to enhance development and

reduce poverty through the use of more affordable materials and the inclusion of

local people in the design and construction process.

The Sahara in the Earth System: Airborne Mineral Dust

Water, or rather the lack of it, is not the only ubiquitous feature of the Saharan

environment with which its inhabitants have had to contend since the onset of desic-

cation many millennia ago. The geomorphological processes associated with past

environmental changes in northern Africa have also resulted in the Sahara becoming

the world’s largest source of airborne mineral dust. The mobilisation, transportation

and deposition of dust has major impacts not only within the Sahara, but in many


regions outside of northern Africa. Dust mobilised in the Sahara is transported large

distances and deposited not only within Africa itself but also over the Atlantic, the

Mediterranean, the Middle East, Europe and the Americas.154,155,156,157 In many of

these regions dust transport and deposition has significant implications for climate,

ecology and human health. Given the potential for human impacts on the land

surface and anthropogenically driven climate change to affect dust production, the

generation of dust represents another key interface between people and the physical

environment in North Africa, with global implications. Here we consider the role of

dust in the coupled earth-ocean-atmosphere system, before moving on to address the

nature of Saharan dust sources and their evolution over time. Finally the impacts on

human systems (principally on human health) of Saharan dust, and of changes in its

mobilisation, transport and deposition, are discussed.

Dust Impacts on Climate

Dust directly affects climate at local and regional scales by modifying the temperature

structure of the atmosphere. On the one hand dust reduces the amount of shortwave

solar radiation reaching the Earth’s surface, causing cooling in the lower atmosphere,

On the other hand dust absorbs outgoing longwave radiation, trapping heat in the

atmosphere and causing warming in much the same way as greenhouse

gases.158,159 These two effects act in opposition to one another, and the overall

effect of dust on temperature depends on a variety of factors including the vertical

distribution of the dust, the size distribution of the dust particles, their composition,

the nature of the underlying surface, and the time of day. Where dust exists in an

elevated layer overlying less dusty air, the overall result will be a cooling of the

Earth’s surface and the lower levels of the atmosphere during the daytime. At

night, only the longwave effect acts, resulting in warmer surface temperatures.

Dust thus acts to reduce the daily temperature range at the Earth’s surface.

Dust exists in a elevated layer – the Saharan Air Layer (SAL) – overlying a

humid oceanic air mass over the Atlantic Ocean and over the Sahel during the

monsoon season. Warming within the dust layer, reinforced by near-surface

cooling during the daytime, acts to reinforce the temperature inversion at the base

of the SAL and stabilise the atmosphere, inhibiting convective activity. This phenom-

enon has been detected over the Sahel,160 where it has been proposed as a mechanism

of drought reinforcement161. Atmospheric stabilisation by dust has also been

observed to inhibit the development of hurricanes over the Atlantic.162 The reduction

of sea surface temperatures by dust over the northern Atlantic may also play a role in

reinforcing or sustaining the dipolar temperature anomaly (cooling over the northern

hemisphere relative to the southern hemisphere and northern Indian Ocean) associ-

ated with drought conditions in the Sahel.163,164,165,166 Low altitude cooling will

also reduce evaporation over the ocean or over moist land surfaces, further reducing

the likelihood of rainfall.167 Simulations using a general circulation model168 indicate

that evaporation and precipitation are reduced globally by dust, although they suggest

that the local presence of dust increases rainfall over deserts. However, this latter

conclusion is not supported by any observational evidence.


Dust can also affect global climate through its influence on marine and terrestrial

ecosystems. The most widely known such impact is the fertilisation of the ocean by

iron carried in aeolian dust originating in the world’s deserts. Jickells et al.169

estimate average global dust production at around 1.7 billion tonnes per year, with

almost two-thirds of this material originating from North Africa and 26 per cent of

the dust reaching the oceans. Once deposited in the ocean, iron from terrestrial

dust is believed to enhance biological productivity, which leads to the sequestering

of atmospheric carbon dioxide that is taken up by phytoplankton and exported to

deep water and ultimately to the ocean floor to be incorporated in marine sedi-

ments.170 Thus oceanic dust deposition contributes to a reduction in the greenhouse

gas content of the atmosphere, acting to modulate the rate of increase of anthropo-

genically driven greenhouse warming. The importance of this effect is a matter of

debate, however, and further research is necessary in order to quantify the impact

of dust on atmospheric greenhouse gas concentrations. Dust also acts to stimulate

the production of dimethyl sulphide (DMS) by marine organisms; when emitted to

the atmosphere, DMS is oxidized to sulfate aerosol, which scatters solar radiation

back to space, thereby acting to cool the Earth’s surface, further offsetting global

or regional surface warming.171

Nutrients from Saharan dust, such as phosphorous, are also important for certain

terrestrial ecosystems, and are believed to play a vital role in sustaining the Amazon

rainforest.172 Modelling studies suggest that the forests of Amazonia are vulnerable to

large-scale die-back caused by higher temperatures and reduced rainfall resulting

from anthropogenic climate change;173,174,175 any reduction in the supply of nutrients

to the forest canopy may conceivably accelerate this process. Die-back of the Amazon

forests would release large quantities of carbon dioxide into the atmosphere, acceler-

ating global greenhouse warming.

Finally, dust may affect climate by modifying the reflectance or albedo of the land

surface. Where dust is deposited over snow fields or ice sheets it reduces the reflec-

tance of the surface, increasing the amount of solar radiation absorbed by the surface

which results in heating and melting of the snow and ice. While such a process is

unlikely to be significant at the global scale today, this mechanism is believed to

have been important in regulating past glacial cycles, in periods when global dust

transport was much greater than at present.176 Dust deposition over mountainous

areas may accelerate the melting of glaciers and snow and ice fields, which are

already shrinking in many areas as a result of climate change. In certain mountain

regions this could impact on water resources, flood hazards and tourism, with

significant economic consequences.

Dust is clearly an important component of the atmosphere, and represents a

medium via which the Sahara influences global climate through the modulation of

global biogeochemical cycles. The Sahara, in turn, is sensitive to global climate

change, as demonstrated by its dramatic history of transitions between arid and

humid conditions. Any change in the Saharan dust cycle is likely to have significant

repercussions for other regions, and perhaps for global climate as a whole. Greater

efforts at understanding the dust cycle and the sensitivity of the Sahara to changes

in climate are thus required if we are to improve our capability to model and


predict future global climate change. Any such efforts must start with an assessment

of the nature of the dust sources themselves.

The Nature of Saharan Dust Sources

Dust is not emitted from the Sahara in a uniform or random fashion; the vast majority

of aeolian material is generated from a variety of specific source regions that exhibit

particular seasonal patterns of activity.177 These regions are prescribed by a combi-

nation of geomorphological and meteorological factors, principally a supply of free

particles available for erosion, little or no vegetation cover, and wind speeds that

regularly exceed the threshold value for emissions for the surface in question.

Wind erosion may be suppressed by the presence of surface crusts (which are

common in arid regions) and the presence of obstacles such as bushes, rocks and

pebbles. The process of emission, once the threshold wind speed has been exceeded,

begins with the mobilisation of large particles (in the size range 40–500 mm) which

are free at the soil surface. These are typically sand particles derived from quartz,

often with small clay particles adhering to their surface, and aggregates of clay.

The impact of these particles on the ground surface results in their own disaggregation

and the breaking up of aggregate particles within the surface. This sandblasting results

in the release of particles of dimension less than about 20 mm which are available for

uplift into the atmosphere and subsequent long-range transport.178,179

In recent years the major dust sources in northern Africa have been identified by

satellite remote sensing, principally using the Total Ozone Mapping Spectrometer

(TOMS) Aerosol Index (AI)180 and the Infra-red Difference Dust Index (IDDI)

derived from data acquired by METEOSAT.181,182 Global monitoring of aerosols

using the TOMS AI indicates that the Sahara is by far the worlds largest dust

source, and that the most active source region is the Bodele Depression, variously

described as the worlds largest single dust source, or the ‘dustiest place in the

world’.183,184 The Bodele Depression contains the exposed lake sediments of the

now desiccated Lake Megachad, and it is these sediments that provide the material

mobilised as airborne dust. However, the importance of the Bodele Depression is

also a result of its geographical situation to the southwest of the gap between the

Tibesti and Ennedi Mountains, which funnels and amplifies the prevailing northeast-

erly winds, resulting in a high frequency of events during which surface winds exceed

the threshold velocity for sediment mobilisation.185 In addition, field and satellite

observations reveal that the lake sediments are partially covered by mobile sand

dunes, which provide an abundant supply of coarse material with which the

palaeolake surface is sandblasted.

The western Saharan region straddling the borders of Mali, Mauritania and

Algeria also appears to be particularly active; in the IDDI imagery this region is as

prominent as the Bodele Depression, and is flanked by additional prominent

sources in south-central Algeria and the northern regions of Western Sahara

(Figure 4). Satellite imagery and field observations indicate that these source

regions are heterogeneous in nature, consisting of a number of different sources

including but not restricted to dry lake beds. Field observations by two of

the authors (Drake and Brooks) in the north of Western Sahara, coupled with


observations from the Moderate Resolution Imaging Spectrometer (MODIS) have

identified clay lake beds as sources of airborne dust. An extensive gravel plain strad-

dling the border of Western Sahara and Mauritania also appears to be a significant

source of dust; this plain is cut by dry channels and it is likely that the sediments

mobilised by the wind in this location derive from both past wind and water action.

The coarse resolution of the TOMS and IDDI imagery (in the region of 40 km)

means that many of the dust sources indicated by long term averages of both types

of data remain an enigma. The situation is complicated by disagreement between

the TOMS and IDDI datasets, which is due at least in part to the lack of sensitivity

of the TOMS AI to low-level dust.186 The main western source region is more promi-

nent, and located further north, in the IDDI than in the TOMS dataset. In the IDDI, the

Bodele region is linked to the western source area by a zone of intermediate IDDI

values extending through central Niger and northern Mali (including the location

of the western maximum in the TOMS data); similar magnitude values are apparent

over much of south-central Mauritania near the northern limit of the monsoon rains. A

number of other regions appear as sources in the IDDI but not in the TOMS

data.187,188 These include a region extending from south to north through the

centre of northern Sudan and southern Egypt, approximately following the course

FIGURE 4

MEAN ANNUAL IDDI VALUES OVER AFRICA, REPRESENTING AVERAGE REDUCTION IN

THE BRIGHTNESS TEMPERATURE OF THE EARTH IN KELVIN, AS MEASURED BY THE

INFRA-RED CHANNEL OF METEOSAT. HIGHER VALUES INDICATE GREATER ABSORPTION

OF INFRA-RED RADIATION BY ATMOSPHERIC AEROSOLS; OVER THE SAHARA MINERAL

DUST IS THE DOMINANT AEROSOL, WHILE HIGH VALUES IN OTHER REGIONS MAY

RESULT FROM BIOMASS BURNING. SAHARAN REGIONS WHERE IDDI VALUES ARE

CONSISTENTLY HIGH (REPRESENTED BY THE ORANGE AND RED AREAS) ARE

INTERPRETED AS REPRESENTING MAJOR DUST SOURCE REGIONS; NOTE THE BODELE

DEPRESSION IN THE CENTRAL-SOUTHERN SAHARA AND THE REGION OF HIGH VALUES

IN THE WESTERN SAHARAN REGION


of the Nile and extending several hundred kilometres either side of it, and a region to

the northwest of the Tibesti mountains where the borders of Libya, Niger and Chad

meet, extending northwards into the central regions of southern Libya.

Despite their disagreement, the TOMS and IDDI data clearly indicate that past

fluvial activity plays an important role in determining the distribution and nature of

the main Saharan dust sources. Sediments from palaeolake Chad provide the material

for wind erosion that makes the Bodele Depression such an active dust source, and

smaller dry lake beds are evident as dust sources in Western Sahara. However,

while topographic lows containing fluvial sediments from past humid episodes are

clearly important in the global dust cycle, a straightforward mapping of dust

sources onto such features appears overly simplistic. For example, the second

largest enclosed basin, the recently identified Lake MegaFezzan in southwestern

Libya,189 appears to generate little or no dust. This palaeolake exhibits very different

geomorphology to palaeolake Chad, as its sediments are capped by hard calcrete and

gypsum crusts and large dunes. Clearly, not all palaeolake surfaces act as dust

sources. Neither are all dust sources associated with palaeolakes, as illustrated by

the emission of dust from the gravel plains of Western Sahara. Instead, dust

sources appear to be heterogeneous in nature, with different sources being activated

at different surface wind speeds.190

Climatic Versus Human Influences on Dust Source Activity

During the 1990s it was suggested by a number of authors that apparent increases in

the mobilisation and transport of northern African dust over the latter half of the twen-

tieth century were the result of land degradation in the Sahel, combined with climatic

desiccation.191,192,193 However, there is little or no evidence for widespread systema-

tic land degradation resulting from human activity in this region, and a number of

studies have convincingly argued that what has often been interpreted as systematic,

regional-scale desertification (the ‘southwards march of the Sahara’) is nothing more

than the physical expression of an oscillation of the ‘desert boundary’ as the activity

and maximum northerly position of the Inter-Tropical Convergence Zone and associ-

ated monsoon air mass varies on interannual and interdecadal timescales.194,195,196,197

Historically, the controversy over the role of land degradation in dust production

comes from the paucity of measurements that would permit the identification of any

long-term trends in dust mobilisation in the Sahel. Indeed, for many years the only

location at which dust was measured on a regular basis was Barbados, some

5,000 km from the western coast of Africa.198 Available since 1965, i.e., around

the beginning of both dramatic population growth in the Sahel and the decline in rain-

fall in this region, this dataset would have been suitable for the search for trends in

production (rather than transport) if its ability to represent processes occurring in

the African continent were not so questionable. In contrast to this highly localised

record, satellites offer a unique opportunity to monitor atmospheric dust at the

global scale. However, suitable satellite imagery for dust monitoring has only been

available since the very end of the 1970s, and thus covers too short a period for it

to be useful in discriminating between an anthropogenic trend and the ‘natural’

impact of drought in the 1980s and 1990s. Combining satellite imagery and data


from the Barbados record may provide a means of addressing this problem in the near

future.

However, as discussed above, assessments of the major dust source regions

indicate that they are overwhelmingly are located in Saharan regions that are unlikely

to have experienced significant human impacts.199,200,201 It appears that climatic

variations have been the principal driver of observed changes in dust production.

On the one hand variations in rainfall have affected vegetation cover and thus

influenced the susceptibility of certain surfaces to wind erosion, while on the other

changes in atmospheric processes are likely to have resulted in changes in the

balance between dust mobilisation and deposition within the Sahel and southern

Sahara.202,203 Climate change may play a key role in modulating future dust export

from the Sahara. Furthermore, Chiapello and Moulin204 have shown that the North

Atlantic Oscillation (NAO), a large-scale meteorological system that essentially

controls the strength of the Trade-Winds during the winter, explains a large part of

the year-to-year variability in dust transport from the Sahara to the tropical North

Atlantic. Its influence spreads further since a correlation of the NAO index with

dust activity in the Bodele depression suggests that the NAO influences dust

emissions as well as transport.205 It is thus likely that future changes in atmospheric

circulation will strongly affect the dust cycle, a problem that numerical models will

have to answer in the forthcoming years.

Future Dust Emissions

On long timescales, dust emissions from northern Africa are modulated by large-scale

changes in climate, as demonstrated by the abrupt increase in wind-blown dust depos-

ited off West Africa at the end of the last Saharan humid phase around 5.5 ka.206

While dust sources do not map directly onto palaeolake surfaces, they are strongly

associated with fluvial activity, begging the question of how long they may remain

active before their reserves of erodible material are depleted once fluvial activity

has effectively ceased.

Wind action is limited to the soil surface, so emission can be sustained in the long

term only if the superficial soil layer is constantly supplied with fine material. Such a

supply can be explained in terms of the settling of dust from the atmosphere, the

disaggregation of larger particles by wind action, and the disturbance of the surface

due to the motion of sand dunes (anthropogenic disturbance, e.g. by vehicles, may

play a role, but the contribution of such processes has not been quantified and there

is no evidence for large ‘new’ dust sources, as discussed above). The occasional

action of rainfall and associated fluvial activity also represents an efficient means

of disturbing the land surface and of carrying fine material to basins via networks

of wadis. While such events are rare in the Sahara, the lack of vegetation means

that single events can disturb and mobilise large amounts of material. Furthermore,

rainfall does occur fairly regularly over some of the Saharan highlands, providing a

means of replenishing erodible material in the adjacent lowlands. Rajot et al.207

conclude that the vertical flux of emitted dust is not supply limited, even with a

fraction of silt and clay as low as 1.5 per cent near the surface. This finding is


supported by evidence that areas not likely to be replenished by recent fluvial activity,

such as flat gravel plains as described above, are also sources of dust.

There thus appears to be little reason to expect the supply of material available for

wind erosion in the Sahara to be exhausted in the foreseeable future. However,

climate change has the potential to alter the amount of dust mobilised in and exported

from the Sahara via its impacts on the land surface. In particular, the possible ‘green-

ing’ of the southern Sahara as a consequence of anthropogenic greenhouse warming

as discussed above, may serve to reduce the supply of dust from certain sources

through the stabilisation of the land surface. A strengthening of the African

Monsoon and resulting expansion of vegetation cover as modelled by Brovkin208

and Claussen et al.209 could conceivably cut off the dust supply from sources close

to the current monsoon belt, including the Bodele Depression. Given the role of

dust in the climatic, meteorological and biogeochemical processes discussed earlier

in this section, such a partial greening of the Sahara could have significant global

and regional implications.

Impacts of Dust on Human Systems

As a component of the Earth system which contributes to the regulation of biogeo-

chemical cycles, ecological productivity and climate, variations in dust mobilisation,

transport and deposition have implications for human activities. Impacts on ocean

productivity may affect fish stocks, with associated economic implications, while

individual dust events affect transport and day-to-day human activities. Large-scale

influences of dust on climate and the possible role of dust in modulating rainfall

may potentially be implicated in food insecurity. The role of dust in the removal of

carbon dioxide from the atmosphere via its influence on marine and terrestrial biologi-

cal productivity is likely to play a role in the sensitivity of the global climate to

anthropogenic greenhouse gas emissions. In order to model and predict future

climate changes that will require adaptive responses in human societies, dust must

be better represented in global climate simulations. The representation of dust in

regional climate models and the incorporation of its effects in regional scenarios of

future climate is necessary if certain key processes are to be represented accurately;

for example, changes in atmospheric dust content over the northern tropical Atlantic

might have profound implications for future trends in hurricane activity. Anticipation

of such trends will be crucial in the development of policies, strategies and measures

to cope with future changes in climate and their impacts.

Regardless of its impact on future changes in climate, the present-day impact of

dust on human health is sufficient grounds for further research into its characteristics,

mobilisation, transport and deposition. Mineral aerosols transported large distances

are generally less than 20 mm in dimension, and typically contain large numbers of

particles of sub-micron dimension. They therefore span the size range for particles

associated with adverse impacts the human respiratory system.210 Saharan dust

events regularly contribute to air pollution limits being exceeded in the Mediterranean

region,211 and African dust frequently reaches the Americas and north-western

Europe, where it can contribute to high atmospheric pollutant levels when combined

with aerosol particles from other sources.212 Variations in concentrations of particles


whose size is the order of 10 mm (PM10s) in Trinidad are associated predominantly

with the transport of dust from northern Africa.213

Very little internationally published research has been carried out into the health

impacts of airborne dust within northern Africa (although see Laval, 1967214 and

Fossati, 1969.215) Nonetheless, research in other parts of the world suggests that

these impacts are likely to be great. Lung conditions in populations living in arid

or semi-arid regions have been linked with exposure to silica in the Himalayas,216

the Thar Desert of India217 and the Southwest USA,218 and the deposition of silica

in the lungs is associated with a widely recognised condition known as ‘desert lung

syndrome’.219,220 Dust events originating in the interior desert of Australia have

been associated with increased incidences of asthma in Brisbane.221 The evidence

for links between dust and asthma in the Caribbean is more equivocal. It has been

suggested that dust clouds originating in the Sahara are associated with increased

paediatric asthma accident and emergency emissions in Trinidad, where there is a

widespread belief that the passage of Saharan dust exacerbates rhinitis and

asthma.222 However, emerging findings from an ongoing study by one of the

authors (Prospero) and colleagues appear to challenge this conclusion. There is

evidence that dust from northern Africa is associated with the long-range transport

of micro-organisms; daily aerosol samples collected throughout 1996–1997 from

the trade winds reaching Barbados yielded significant concentrations of viable (i.e.

culture-forming) bacteria and fungi only when African dust was present.223 In parts

of the Sahel, dust storms are believed to be responsible for certain potentially fatal

sicknesses.224

There is thus a strong public health case for research into the health effects of dust

in the Sahara, and into ways to ameliorate any such effects. A greater understanding of

the activity and nature of Saharan dust sources, and of their potential future evolution,

is therefore relevant for health policy both inside and outside northern Africa, as well

as for activities such as climate forecasting. Such an understanding requires compre-

hensive campaigns of field work to investigate the geomorphological, geochemical

and mineralogical nature of the key dust sources and the surface conditions associated

with emission, as well as remote sensing studies of dust sources and transport. Such

fieldwork is currently underway in the Bodele Depression225 source region and is at

an early stage inWestern Sahara, but there is little such activity in other parts of north-

ern Africa beyond the monitoring of dust event frequencies and visibility.

Discussion and Conclusions

Research focusing on regions rather than processes is currently somewhat unfashion-

able. However, a focus on specific regions such as the Sahara provides an excellent

opportunity to investigate linkages between different physical and social systems

and to conduct truly interdisciplinary research. The Sahara in particular has been

neglected by the research community for a number of reasons, including (i) perceived

difficulty of access and security issues, which discourages researchers from conduct-

ing field work, (ii) the economic and political marginalisation of the region and the

associated lack of participation of Saharan countries in international research


programmes, (iii) the difficulty faced by nationals of Saharan countries in travelling

outside of the region, (iv) the isolation of individual researchers working on Saharan

issues within countries outside of the region, and (v) the fact that what little research is

conducted in the Sahara is often seen as either rather esoteric in nature (e.g. research

into archaeology and rock art) or only of relevance to a tiny minority of specialists

(e.g. ethnographic and anthropological research).

The lack of interest in the Sahara among the wider research community is remark-

able given the proximity of the region to Europe and the strong historical ties between

Europe and North Africa, not to mention the importance of a number of Saharan

nations as oil producers. Indeed, these factors have drawn the attention of the security

community, particularly in the United States, although this attention is not always

matched by detailed knowledge of the political reality on the ground.226 The

combination of drought with other economic, social and political issues to precipitate

political instability and conflict at the margins of the Sahara has been described above.

Developmental and security issues should be sufficient to foster a greater interest in

the Sahara in Europe and North America; however, so should the broader issues

related to the Sahara’s role in the Earth system and human well-being as discussed

in this paper.

The Sahara as a Component of the Earth System

The neglect of the physical sciences in the Sahara is particularly notable. While a

number of teams are conducting archaeological and palaeoenvironmental research

in a variety of Saharan countries, there has been very little work on contemporary

environmental issues in the Sahara. This fact is thrown into sharp relief by the

recent investment of millions of US Dollars in research into the transport of

mineral dust from Chinese sources.227 Material from these sources transported over

densely populated parts of East Asia, notably the Korean peninsula, has relevance

for regional climate, human health and ecological systems, and also reaches North

America. However, there is very little funding for, and no concerted programme

of, research into the transport of dust from the Sahara, which produces more dust

than the Chinese sources, providing up to two thirds of the global atmospheric dust

burden228 and contributing to air pollution in Africa, Europe, the Caribbean and

North America. Just as in China, trends in dust storm activity and in the export of

dust over adjacent regions have been detected for the Sahara.229 Furthermore, dust

has been implicated in the suppression of rainfall and the possible exacerbation of

drought in the Sahel, and has been demonstrated to suppress tropical cyclone and hur-

ricane formation over the Atlantic. Better understanding and monitoring of Saharan

dust events thus has the potential to improve seasonal climate forecasts for a key

marginal region of sub-Saharan Africa, and enhance hurricane forecasting capabili-

ties. Long-range forecasts of dust mobilisation and transport based on climate

change scenarios coupled with dynamic representations of geomorphological

processes and vegetation cover may help us forecast climate variability and extremes

(specifically droughts and hurricanes), and assist in the development of policies to

deal with the potential impacts of climate change.


On longer timescales, a better understanding of how the Saharan land surface

responds to climate change in terms of dust production and vegetation feedbacks

should enable us to reduce uncertainties in the representation of the carbon cycle

and atmospheric aerosols in global climate models. The incorporation of mineral

aerosols into such models will improve the representation of the Earth System and

should lead to more realistic climate change scenarios, with improved estimates of

the sensitivity of global climate and of ecosystems to increases in atmospheric green-

house gas concentrations and global mean surface temperatures. This in turn will help

policy makers to assess climate change risks and design climate change mitigation

and adaptation policies. Current computer models suggest that anticipated changes

in global climate may lead to a greening of the southern Sahara and northern Sahel

in the near to medium term, a process that may reduce the supply of erodible material

for aeolian transport. Such a reduction in dust supply could exacerbate global

warming via adverse impacts on carbon-storing marine and terrestrial ecosystems,

representing a positive feedback in the climate system (although these effects

might be offset to a certain extent by increased carbon sequestration in a more

densely vegetated Sahara). Reduced dust emissions from northern Africa may also

increase the frequency and/or intensity of Atlantic hurricanes, exacerbating hurricanerisk in the Caribbean, the southeastern United States, Central America and even

northern South America. On the other hand reduced atmospheric dust content may

have beneficial impacts in terms of human health and rainfall in certain regions.

While the mechanisms associated with all these processes have been clearly ident-

ified, the processes themselves are not understood in a quantitative manner; the

importance of the associated impacts could be significant, but will remain a matter

of speculation pending further research.

Climate Change and Human Adaptation: Lessons from the Past

The past provides us with no exact analogues for the anthropogenic greenhouse

warming that is anticipated over the coming centuries. Nonetheless, studies of the

Saharan past can teach us many lessons about physical and social systems and

their interaction. The Holocene climatic optimum occurring between about 10 and

5 ka, which was associated with humid conditions in the Sahara, might provide us

with a very approximate model for future climate change, provided such comparisons

are treated with caution – the mechanisms behind the early Holocene warming and

monsoon intensification are different from those driving present day anthropogenic

climate change. Furthermore, it must be appreciated that global mean temperatures

are likely to exceed those of the Holocene climatic optimum before the end of the

twenty-first century.230

Palaeoenvironmental data can help us develop a better understanding of past

changes in climate and the connections between changes at high and low latitudes.

Such connections are exemplified by the association between cooling in the North

Atlantic, apparently originating at high latitudes, and arid crises in the Sahara and

the wider northern hemisphere subtropical and extra-tropical region during otherwise

warm humid periods (see Brooks231 for a review of these changes and their impacts

on human societies). It is plausible that such cooling episodes will result from inputs


of freshwater from melting ice and increased river runoff in the Atlantic associated

with anthropogenic climate warming.232 The likelihood of such events being precipi-

tated by human-induced climate change is currently a matter of great debate, and

studies of past climates can yield information about the conditions under which

such events occur. Alley argues that past events have occurred with little or no exter-

nal forcing and that the difficulty current climate models have in reproducing such

events indicates that they may be more likely to occur in the future than is generally

accepted. The occurrence of such events in the warm early Holocene suggests that

anthropogenic greenhouse warming may not necessarily mean that such transient

cold, arid episodes will be absent from the foreseeable future. Studies from regions

such as the Sahara can tell us how severe such events must be in order for their

effects to be felt in low latitudes with potentially adverse consequences for human

populations. Again, such information can help policy makers develop strategies for

avoiding ‘dangerous’ climate change associated with the crossing of critical

thresholds in the climate system beyond which we are likely to experience abrupt

changes in climate.233,234,235

As well as yielding information on the workings of the climate system and links

between climate change and environmental impacts, studies of the past can tell us

much about how human societies respond to environmental change. While patterns

of vulnerability and the nature of human responses to climatic and environmental

change are context-specific, it is possible to make some very general (although not uni-

versal) observations regarding human responses to such changes.236,237 The Sahara

reinforces the lesson, evident from more recent experiences in the Sahel, that mobility

is the key to the long-term sustainability of livelihoods in highly variable environ-

ments. The long records of environmental change in northern Africa caution us to

beware of complacency during periods of abundance associated with increased rain-

fall, as these are invariably followed by episodes of scarcity. Social systems in such

environments must be flexible and responsive; livestock-based pastoralism is a

more appropriate strategy than large-scale rain-fed agriculture where rainfall is

scarce and unpredictable. While this observation might be a moot point in the

hyper-arid regions of the Sahara, it is of great relevance for marginal regions such

as the Sahel, where developmental pressures and aspirations of economic growth

led to agricultural expansion in the anomalously wet 1950s and early 1960s, a time

of optimism driven by independence from colonial rule and the prevailing ideology

of technological progress. This expansion only served to exacerbate the impacts of

the subsequent period of drought and desiccation.238 Although model studies (to a

certain extent supported by observations) suggest that the northern Sahel and southern

Sahara may become wetter in the near future, the real possibility of occurrence of cold

arid episodes as described above, or of a reversal of the greening trend at high atmos-

pheric greenhouse gas concentrations, means that extreme caution should be exercised

in the exploitation of climatic amelioration along the Sahel-Sahara boundary. Model-

ling studies and palaeoclimatic data caution us that social systems in northern Africa

must be prepared to confront extreme environmental variability if they are to survive.

Studies of past responses to climatic and environmental change in the Sahara

paint a picture of largely reactive, ad hoc ‘adaptation’ occurring during times of


environmental crisis. The rapid spread of cattle herding suggests migration as a last

resort to environmental deterioration, rather than a smooth and painless process in

which human populations respond to changes as they occur without suffering signifi-

cant hardship.239 While this may not appear to be a very controversial conclusion, it

stands in sharp contrast to current paradigms of adaptation, which promote the devel-

opment of resilience and ‘adaptive capacity’ as a means of coping with climate

change240 – the implication being that adaptation is a means of ‘neutralising’ the

impacts of climate change and thus of avoiding adverse consequences. In fact

the current view of adaptation as something wholly benign is challenged by the

archaeological record, which demonstrates that adaptation to environmental change

has generally involved compromises, and is often what we may term sub-optimal’

in that, while it enables human populations to survive in the face of change, it

often has negative aspects. The archaeological record illustrates that Saharan societies

became more territorial and stratified as they responded to climatic desiccation, in

response to a reduction in the productivity of the physical environment. Where

these processes culminated in the emergence of urbanisation based on technological

solutions to adaptation, as in the Nile Valley and the Libyan Fezzan, they proved to be

vulnerable to collapse in the face of later climatic crises or simply unsustainable in the

longer term.241,242 While the factors behind the demise of the Garamantian culture in

this region are not known, it is possible that declining groundwater levels played a

role, and that the Garamantes were confronted with limits to their technological

ability to adapt to environmental change.

Future Economic Development in the Sahara

Modern technology enables settlement in previously uninhabitable regions, and

permits the rehabilitation of areas where settlement and agriculture have been in

decline, such as the Garamantian heartland in the Fezzan. This transformation (in

terms of both reality and aspiration) is evident in the local toponymy, with the

Wadi al-Ajal (usually translated as ‘Valley of Death’) having been renamed the

Wadi al-Hayat (‘Valley of Life’). Another example of technology enabling greater

access to water is the Great Manmade River, which transports fossil water from the

aquifers of the central Sahara in southern Libya to the Mediterranean coast, providing

an expanding population and a developing economy with a vital resource. However,

water resources in the central Sahara are finite, and estimates of the amount of water

available, and the time until it runs out, vary considerably. A better understanding of

the origins of this fossil water, of the dynamics of the aquifers, and of the conditions

necessary for their recharge, may contribute to a more sustainable approach with

water management. An appreciation of the origins and finite nature of Saharan

water resources, coupled with policies designed to encourage water conservation,

might go some way towards extending the lifetime of fossil reserves; policies

designed to foster more ‘responsible’ use of resources are more likely to succeed

when underpinned by campaigns of awareness raising, which must ultimately be

based on a sound scientific understanding of the region’s natural resource base.

Water is not the only precious commodity in the Sahara. Oil and tourism underpin

much of the economic development in the region. Tourism in the Sahara ultimately


depends on the preservation of natural and cultural heritage, being predominantly

based on tourists’ interest in landscapes, history and archaeology. This heritage can

be compromised by a number of factors, including insensitive development and oil

exploration, which threaten a number of sites of scientific and archaeological interest,

and have already damaged or destroyed some.243 Of course tourism itself can cause

significant damage when it is not carefully managed, through souvenir hunting and

practices such as the wetting of rock paintings in order to temporarily enhance

their appearance for photographic purposes.244,245,246 Nonetheless, tourism represents

a potentially lucrative, and sustainable, source of income for many Saharan

communities and nations. Whereas oil reserves will eventually run out, or may

become significantly less profitable if concerns about climate change lead to a shift

to alternative energy sources, natural and cultural heritage represent an unlimited

resource if carefully managed. While it may be expedient in the short term to sacrifice

heritage for the sake of oil exploration and production, it is neither necessary nor sen-

sible in the long-term. Oil-based development can be complemented by low-impact

tourism that exploits the great interest in natural history and the human past among

the educated populations of affluent nations, who are constantly seeking novel

destinations. Such tourism can be underpinned by research in the physical and

social sciences, which furthers our understanding of the natural and cultural heritage.

In some parts of the Sahara, local people are educating themselves in subjects such as

archaeology in order to participate more fully in heritage tourism and assist in the

management of valuable archaeological sites.247

Conclusions

In conclusion, we stress that the Sahara is not merely the ‘empty space’ of the popular

imagination. Like other regions, the Sahara is faced with the challenge of ‘sustainable

development’ in the face of a growing population and an uncertain climatic future. In

the Sahara, this challenge is compounded by a reliance on non-renewable water

resources, extreme environmental variability, and a high sensitivity to global

climate change. The challenge of coping with variability and change is particularly

acute at the southern margins of the desert, along the oscillating Sahara-Sahel bound-

ary. Here, future development policies must be built around this variability, and

founded on a profound appreciation of it. There is much that can be learnt from

traditional resource management practices, based on flexibility and mobility, and

tried and tested livelihood models and coping strategies should not simply be

discarded in the name of progress and modernisation. Technological innovation has

a role to play in development, for example through the development of methods to

access and deploy scarce water resources, provided the consequences of such

activities are carefully considered. Economic diversification also has its part to

play, for example in the form of locally-run and sensitively managed archaeology-

based tourism.

The physical and social sciences have a key role to play in the future development

of the greater Saharan region. However, in order that this role be fulfilled, there must

be more international support for Saharan research, and much more collaboration


between members of the international scientific community and their counterparts in

the Saharan nations. Finally, the role of the Sahara in influencing global climate

requires much more investigation; the potential of Saharan dust to affect the global

carbon cycle and perhaps accelerate the process of anthropogenic climate change

means that to ignore the Sahara is to neglect a key component of the Earth system.

NOTES

1. N. Brooks, Drought in the African Sahel: Long-Term Perspectives and Future prospects, (Norwich:Tyndall Centre Working Paper No. 61. 2004); available at www.tyndall.ac.uk.

2. A. Goudie, Environmental Change, Third Edition, (Oxford: Oxford University Press 1992).3. F. Gasse and E. Van Campo, ‘Abrupt post-glacial climate events in West Asia and North Africa

monsoon domains’, Earth and Planetary Science Letters 126 (1994) pp. 435–456.4. D. Jolly, S. P. Harrison, B. Damnati and R. Bonnefille, ‘Simulated climate and biomes of Africa

during the late Quaternary: Comparison with pollen and lake status data’, Quaternary ScienceReviews 17 (1998) pp. 629–657.

5. R. B. Alley, P. A. Mayewski, T. Sowers, M. Stuiver, K. C. Taylor and P. U. Clark, ‘Holocene climaticinstability: A prominent, widespread event 8200 years ago’, Geology 25 (1997) pp. 483–486.

6. U. von Grafenstein, H. Erlenkeuser, J. Muller, J. Jouzel and S. Johnsen, ‘The cold event 8200 yearsago documented in oxygen isotope records of precipitation in Europe and Greenland’, ClimateDynamics 14 (1998) pp. 73–81.

7. M. R. Talbot, ‘Late Pleistocene rainfall and dune building in the Sahel’, In A. A. Balkema (Ed.),Palaeoecology of Africa 16, (Rotterdam: Balkema 1983), pp. 203–213.

8. J. Maley, ‘Last Glacial Maximum lacustrine and fluviatile Formations in the Tibesti and other Saharanmountains, and large-scale climatic teleconnections linked to the activity of the Subtropical JetStream’, Global and Planetary Change 26 (2000) pp. 121–136.

9. Goudie (note 2).10. Ibid.11. The Pleistocene corresponds to the majority of the Quaternary period, excluding the last 10,000 years,

which constitutes the Holocene.12. Deposits laid down in the beds of ancient, now dry lakes, or palaeolakes.13. A recently identified large palaoelake in southwestern Libya.14. N. A. Drake, K. H. White and S. McLaren, ‘The Palaeohydrology and Palaeoclimate of Lake

Megafezzan: A Giant Palaeolake in the Fezzan Basin, Libyan Sahara’, Quaternary ScienceReviews (in press).

15. C. Causse, C. Coque, J. Ch. Fontes, F. Gasse, E. Gibert, H. Ben Ouezdou, and K. Zouari, ‘Two highlevels of continental waters in the southern Tunisian chotts at about 90 and 150 ka’, Geology 17(1989) pp. 922–925.

16. J. C. Fontes and F. Gasse, ‘PALHYDAF (Palaeohydrology in Africa) program: objectives, methods,major results’, Palaeogeography, Palaeoclimatology, Palaeoecology 84 (1991) pp. 191–215.

17. B. J. Szabo, C. V. Haynes and T. A. Maxwell, ‘Ages of Quaternary pluvial episodes determined byuranium-series and radiocarbon dating of lacustrine deposits of Eastern Sahara’, Palaeogeography,Palaeoclimatology, Palaeoecology 113 (1995) pp. 227–242.

18. J. S. Rowan, S. Black, M. G. Macklin, B. J. Tabner, and J. Dore, ‘Quaternary environmental change inCyrenaica evidenced by U-Th, ESR and OSL dating of coastal alluvial fan sequences’, Libyan Studies31 (2000) pp. 5–16.

19. K. White, N. A. Drake, A. C. Millington and S. Stokes, ‘Constraining the timing of alluvial fanresponse to late Quaternary climatic changes, southern Tunisia’, Geomorphology 17 (1996)pp. 295–304.

20. E. Tuenter, S. L.Weber, F. J. Hilgen and L. J. Lourens, ‘The response of the African summer monsoonto remote and local forcing due to precession and obliquity’, Global and Planetary Change 36 (2003)pp. 219–235.

21. N. Petit-Maire (ed.) Le Shati, lac pleistocene du Fezzan (Libye) (Paris/Marseille: CNRS 1982).22. Jolly et al. (note 4).23. M. Cremaschi, M. Pelfini, L. Arzuffi, V. Di Mauro, M. Santilli and A. Zerboni, ‘A palaeoclimatic

record for the late Holocene in the central Sahara: tree rings of Cypressus dupretiana from theWadi Tanezzuft area (SW Fezzan, Libya)’, in M. Kaennel Dobbertin and U. Braker (eds.) Abstracts


of the International Conference “Tree Rings and People”, Davos, 22nd26th September 2001,(Birmensdorf, Switzerland 2001).

24. F. Wendorf, R. Schild and Associates, Holocene Settlement of the Egyptian Sahara, Vol. 1: TheArchaeology of Nabta Playa. (New York: Plenum 2001).

25. G. Bond, W. Showers, M. Cheseby, R. Lotti, P. Almasi, P. deMenocal, P. Priore, H. Cullen, I. Hajdasand G. Bonani, ‘A pervasive millennial-scale cycle in North Atlantic Holocene and glacial cycles’,Science 278 (1997) pp. 1257–1266.

26. Gasse and van Campo (note 3) & Alley et al. (note 5).27. Gasse and van Campo (note 3).28. B. Damnati, ‘Holocene lake records in the Northern Hemisphere of Africa’, Journal of African Earth

Sciences 31 (2000) pp. 253–262.29. N. Brooks, ‘Cultural responses to aridity in the Middle Holocene and increased social complexity’,

Quaternary International (in press).30. Ibid.31. F. A. Hassan, ‘Palaeoclimate, food and cultural change in Africa: an overview’, in F. A. Hassan, (ed.)

Droughts, Food and Culture. (New York: Kluwer Academic/Plenum Publishers 2002) pp. 11–26.32. S. di Lernia, ‘Dry climatic events and cultural trajectories: adjusting Middle Holocene Pastoral

economy of the Libyan Sahara’, in F. A. Hassan (ed.) Droughts, Food and Culture. (New York:Kluwer Academic/Plenum Publishers 2002), pp. 225–250.

33. M. Claussen, V. Brovkin and A. Ganopolski, ‘Climate change in northern Africa: The past is not thefuture’, Climatic Change 57 (2003) pp. 99–118.

34. Bond et al. (note 25).35. R. Vernet, Les paleoenvironments du nord de l’Afrique depuis 600.000 ans. Dossiers et Recherches

sur l’Afrique 3. (Paris: CNRS 1994).36. H. F. Lamb, F. Gasse, A. Benkaddour, N. el-Hamouti, S. van der Kaars, W. T. Perkins, N. J. Pearce,

and C. N. Roberts, ‘Relation between century-scale Holocene arid intervals in tropical and temperatezones’, Nature 373 (1995) pp. 134–137.

37. M. Cremaschi, ‘Late Quaternary geological evidence for environmental changes in south-westernFezzan (Libyan Sahara)’, in M. Cremaschi and S. di Lernia (eds.) Wadi Teshuinat: Palaeoenviron-ment and prehistory in south-western Fezzan (Libyan Sahara) (Rome: Centro Interuniversitario diRicerca per le Civilta e l’Ambiente del Sahara Antico 1998) pp. 13–47.

38. di Lernia (note 32).39. Hassan (note 31).40. M. Cremaschi and S. di Lernia, ‘Environment and settlements in the Mid-Holocene palaeo-oasis of

Wadi Tanezzuft (Libyan Sahara)’, Antiquity 75 (2001) pp. 815–825.41. Brooks (note 29).42. M. Cremaschi, S. di Lernia and E. A. A. Garcea, ‘Some Insights on the Aterian in the Libyan

Sahara: Chronology, Environment, and Archaeology’, African Archaeological Review 15 (1998)261–286.

43. N. Petit-Maire (ed.) Le Shati, lac pleistocene du Fezzan (Libye) (Paris/Marseille: CNRS 1982).44. E. A. A. Garcea (ed.) Uan Tabu in the settlement history of Libyan Sahara. Arid Zone Archaeology,

Monographs 2 (Firenze: All’Insegna del Giglio 2001).45. S. di Lernia (ed.) The Uan Afuda cave. Hunter-gatherer societies of Central Sahara. Arid Zone

Archaeology, Monograph 1 (Firenze: Edizioni All’Insegna del Giglio 1999), pp. 223–237.46. F. Wendorf, R, Schild, A. E. Close, The Prehistory of Wadi Kubbaniya, Vol. 2: Stratigraphy,

Palaeoeconomy and Environment (Dallas: Southern Methodist University Press 1989).47. F. Wendorf, R. Schild, A. E. Close, Egypt during the Last Interglacial: the Middle Paleolithic of Bir

Tarfawi and Bir Sahara (New York, London: Plenum Press 1993).48. F. Wendorf and R. Schild, ‘The Middle Palaeolithic of North Africa—A status Report’, in E. Klees

and R. Kuper, (eds.), New Light on the Northeast African Past (Cologne, 1992), pp. 39–78.49. Cremaschi et al. (note 42).50. B. Barich (ed.) Archaeology and Environment in the Libyan Sahara: the excavations in the

Tadrart Acacus, 1978–1983. BAR International Series #368. (Oxford: British ArchaeologicalReports 1987).

51. di Lernia (note 32).52. K. Wasylikowa, K. Plant, ‘Macrofossils from the archaeological sites of Uan Muhuggiag and

Ti-n-Torha, southwestern Libya’, in L. Krzyzaniak, M. Kobusiewicz, and J. Alexander (eds.),Environmental Change and Human Culture in the Nile Basin and Northern Africa Until theSecond Millennium B.C. (Poznan Archaeological Museum, Poznan 1993), pp. 25–41.


53. K. Wasylikowa, J. Mitka, F. Wendorf and R. Schild, ‘Exploitation of wild plants by the earlyNeolithic hunter-gatherers of the Western Desert, Egypt: Nabta Playa as a case-study’, Antiquity71 (1997) pp. 932–941.

54. A. M. Mercuri, G. Trevisan Grandi, M. Mariotti, M. Lippi and M. Cremaschi, ‘New pollen data fromthe Uan Muhuggiag rockshelter. Libyan Sahara, VII-IV millennia bp’, In M. Cremaschi andS. di Lernia (eds.), Wadi Teshuinat. Palaeoenvironment and prehistory in south-western Fezzan(Libyan Sahara). (Milano: Quaderni di Geodinamica Alpina e del Quaternario, 7 1998), pp. 107–124.

55. S. di Lernia and G. Manzi, ‘Funerary practices and anthropological features at 8000–5000 BP. Someevidence from central-southern Acacus (Libyan Sahara)’, in M. Cremaschi and S. di Lernia (Eds.),Wadi Teshuinat: Palaeoenvironment and prehistory in south-western Fezzan (Libyan Sahara).(Milano: Centro Interuniversitario di Ricerca per le Civilta e l’Ambiente del Sahara Antico. 1998),pp. 217–242.

56. J. D. Irish, ‘Human skeletal remains from three Nabta Playa sites’, In F. Wendorf, R. Schild andassociates (eds.) Holocene settlement of the Egyptian Sahara 1. The archaeology of Nabta Playa.(New York: Kluwer Academic/Plenum Publishers 2001), pp. 521–528.

57. M. Cremaschi and S. di Lernia, ‘The geoarchaeological survey in central Tadrart Acacus andsurroundings (Libyan Sahara): Environment and cultures’, in M. Cremaschi and S. di Lernia(Eds), Wadi Teshuinat: Palaeoenvironment and prehistory in south-western Fezzan (LibyanSahara), (Milano: Centro Interuniversitario di Ricerca per le Civilta e l’Ambiente del SaharaAntico 1998), pp. 234–296.

58. C. Babalini, C. Martınez-Labarga, J. A. Lorente, M. Lorente and O. Rickards, ‘Ancient DNA studies:first results’, in S. di Lernia and G. Manzi (eds.) Sand, Stones, and Bones. The Archaeology of death inthe Wadi Tanezzuft Valley (5000–2000 bp). The Archaeology of Libyan Sahara – AZA MonographsVolume I. (Firenze: All’Insegna del Giglio: Firenze, 2002) pp. 269–280.

59. S. di Lernia and O. Rickards, ‘Genetics in African archaeology. First insights from the Acacus Mts.(Libyan Sahara)’, Manuscript in preparation.

60. H. Lhote, Vers d’autres Tassili. (Paris: Arthaud 1976).61. F. Mori, Tadrart Acacus. Arte Rupestre e Culture del Sahara Preistorico. (Torino: Einaudi 1965).62. Ibid.63. J-L. Le Quellec, Art Rupestre et Prehistoire du Sahara. (Paris: Bibliotheque scientifique Payot 1998).64. F. Wendorf and R. Schild, ‘Nabta Playa and its role in northeastern African prehistory’, Journal of

Anthropological Archaeology 17 (1998) pp. 97–123.65. Wendorf et al. (note 24).66. F. Marshall and E. Hildebrand, ‘Cattle before crops: the beginnings of food production in Africa’,

Journal of World Prehistory 16 (2002) pp. 99–143.67. R. H. Behnke Jr., I. Scoones and C. Kerven, Range Ecology at Disequilibrium: NewModels of Natural

Variability and Pastoral Adaptation in African Savannas. (London: Overseas Development Institute1993).

68. Hassan (note 31).69. S. di Lernia. ‘Building monuments, creating identity. Cattle cult as a social response to rapid environ-

mental changes in the Holocene Sahara’, Quaternary International (in press).70. S. di Lernia and A. Palombini. ‘Desertification, sustainability, and archaeology: indications from the

past for an African future’, Origini XXIV (2002) pp. 303334.71. H. Flohn and S. Nicholson. ‘Climatic fluctuations in the arid belt of the “Old World” since the last

glacial maximum; possible causes and future implications’, Palaeoecology of Africa 12 (1980)pp. 3–21.

72. di Lernia and Palombini (note 70).73. Ibid.74. R. Schild and F. Wendorf, ‘The megaliths of Nabta Playa.’ Focus on Archaeology 1, 2204, pp. 10–15.75. Hassan (note 31).76. di Lernia (note 69).77. di Lernia (note 45).78. di Lernia (note 69).79. Cremaschi et al. (note 23).80. D. J. Mattingly, C. M. Daniels, J. N. Dore, D. Edwards, J. Hawthorne et al., The Archaeology of

Fazzan: Volume 1, Synthesis. (Tripoli and London: Department of Antiquities, Tripoli, and Societyfor Libyan Studies, London 2003).

81. Ibid.82. Wendorf and Schild (note 65).


83. F. Paris, ‘African livestock remains from Saharan mortuary contexts’, in R. M. Blench andK. C. MacDonald (eds.), The Origins and Development of African Livestock: Archaeology, Genetics,Linguistics, and Ethnography, (London: UCL Press 2000), pp. 111–126.

84. di Lernia (note 69).85. di Lernia and Palombini (note 70).86. di Lernia and Manzi (note 55).87. S. Sivili, ‘A historical background: mortuary archaeology in the Sahara between colonialism and

modern research’, in S. di Lernia and G. Manzi (Eds.), Sand, Stones and Bones: The Archaeologyof Death in the Wadi Tannezzuft Valley (5000–2000 bp), The Archaeology of Libyan Sahara,Volume I, AZA Monographs 3. (Firenze: All’Insegna del Giglio 2002).

88. S. di Lernia and G. Manzi (eds.), Sand, Stones, and Bones. The archaeology of death in the WadiTanezzuft Valley (5000–2000 bp). The Archaeology of Libyan Sahara, Volume I, AZA Monographs3. (Firenze: All’Insegna del Giglio 2002).

89. N. Brooks, N. Drake, S. McLaren and K.White, K. ‘Studies in geography, geomorphology,environment and climate’, in D. J. Mattingly (ed.) The Archaeology of Fazzan. (Tripoli andLondon: Department of Antiquities, Tripoli, and Society for Libyan Studies, London 2003).

90. N. A. Drake, A. Wilson, R. Pelling, K. H. White, D. J. Mattingly and S. Black. Water Table Decline,Springline Desiccation and the Early Development of Irrigated Agriculture in the Wadi al-Ajal,Libyan Fazzan. Libyan Studies 35 (2004) pp. 95–112.

91. M. Cremaschi and S. di Lernia. ‘Environment and settlements in the Mid-Holocene palaeo-oasis ofWadi Tanezzuft (Libyan Sahara)’, Antiquity 75 (2001) pp. 815–825.

92. Brooks et al. (note 89).93. Mattingly et al. (note 80).94. S. di Lernia, G. Manzi and F. Merighi. ‘Cultural variability and human trajectories in later prehistory

of theWadi Tenezzuft’, In S. di Lernia and G. Manzi (Eds.) Sand, Stones and Bones: The Archaeologyof Death in the Wadi Tannezzuft Valley (5000–2000 bp), The Archaeology of Libyan Sahara, VolumeI, AZA Monographs 3. (Firenze: All’Insegna del Giglio 2002).

95. M. Liverani, ‘The Garamantes: a fresh approach’, Libyan Studies 31 (2000) pp. 17–28.96. M. Liverani, ‘Looking for the southern frontier of Garamantes’, Sahara 12 (2000) pp. 31–33.97. M. Liverani, ‘The Libyan Caravan Road In Herodotus IV.181–185’, Journal of the Economic and

Social History of the Orient 43 (2000) pp. 496–520.98. Mattingly et al. (note 80).99. Brooks et al. (note 89).100. Drake et al. (note 90).101. Mattingly et al. (note 80).102. M. van der Veen, ‘Ancient Agriculture in Libya: A Review of the Evidence’, Acta Palaeobotanica 35

(1995) pp. 85–98.103. A. I. Wilson and D. J. Mattingly, ‘Irrigation technologies: foggaras, wells and field systems’, in

Mattingly (ed.) The Archaeology of Fazzan vol. 1, Synthesis. (Tripoli and London: Department ofAntiquities, Tripoli, and Society for Libyan Studies, London 2003), pp. 235–78.

104. Drake et al. (note 90).105. Ibid.106. di Lernia et al. (note 94).107. Mattingly et al. (note 80).108. B. Midant-Reynes, The Prehistory of Egypt. (Oxford: Blackwell 1992).109. B. Adams and K. M. Cialowicz, Protodynastic Egypt. (Buckinghamshire, UK: Shire Publications Ltd.

1997).110. Brooks (note 29).111. Midant-Reynes (note 108).112. P. Hoelzmann, B. Keding, H. Berke, S. Kropelin and H-J. Kruse, ‘Environmental change and archae-

ology: lake evolution and human occupation in the Eastern Sahara during the Holocene’, Palaeogeo-graphy, Palaeoclimatology, Palaeoecology 169 (2001) pp. 193–217.

113. T. Wilkinson, Genesis of the Pharaohs. (London: Thames and Hudson 2003).114. J. Linstadter and S. Kropelin, ‘Wadi Bakht revisited: Holocene climate change and prehistoric

occupation in the Gilf Kebir region of the Eastern Sahara, SW Egypt’, Geoarchaeology 19 (2004)pp. 735–778.

115. D. Wengrow, ‘Rethinking “cattle cults” in early Egypt: Towards a prehistoric perspective on theNarmer Palette’, Cambridge Archaeological Journal 11 (2001) pp. 91–104.

116. Midant-Reynes (note 108).


117. Brooks (note 29).118. M. Brett and E. Fentress, The Berbers. (Oxford: Blackwell Publishers 1996).119. M. Hachid, Les Premieres Berberes: Entre Mediterranee, Tassili et Nil. (Algiers: INA-YAS, Edisud

2001).120. Mattingly et al. (note 80).121. Di Lernia et al. (note 94).122. Brooks et al. (note 89).123. Liverani (notes 95 and 97).124. Mattingly et al. (note 80).125. K. White, N. Brooks, N. Drake, M. Charlton and S. MacLaren, ‘Monitoring vegetation change in

desert oases by remote sensing; a case study in the Libyan Fazzan’, Libyan Studies 34 (2003)pp. 153–166.

126. H. Yang, H. and A. J. B. Zehnder, ‘Water scarcity and food import: A case study for southernMediterranean countries’, World Development 30 (2002) pp. 1413–1414.

127. R. Ragab and C. Prudhomme, ‘Climate change and water resources management in arid and semi-aridregions: prospective and challenges for the 21st century’, Biosystems Engineering 81 (2002) pp. 3–34.

128. A. M. Ebraheem, S. Riad, P. Wycisk, A.M. Seif El-Nasr, ‘Simulation of impact of present and futuregroundwater extraction from the non-replenished Nubian Sandstone Aquifer in southwest Egypt’,Environmental Geology 43 (2002) pp. 188–196.

129. J. Keenan, The Tuareg: People of Ahaggar. 2nd Edition (London: Sickle Moon Books 2002).130. N. Brooks, S. di Lernia, N. Drake, M. Raffin and T. Savage, ‘The geoarchaeology of Western Sahara:

Preliminary results of the first Anglo-Italian expedition in the “free zone”’, Sahara 14 (2003)pp. 63–80.

131. M. Hulme, ‘Climatic perspectives on Sahelian desiccation: 1973–1998’, Global EnvironmentalChange 11 (2001) pp. 19–29.

132. Ibid.133. Brooks (note 1).134. A. Giannini, R. Saravanan and P. Chang, ‘Oceanic forcing of Sahel Rainfall on interannual to inter-

decadal timescales’, Science 302 (2003) pp. 1027–1030.135. V. Brovkin, ‘Climate-vegetation interaction’, Journal de Physique IV (2002) pp. 57–72.136. M. Claussen, V. Brovkin, A. Ganopolski et al., ‘Climate change in northern Africa: The past is not the

future’, Climatic Change 57 (2003) pp. 99–118.137. K. Maynard, J. F. Royer and F. Chauvin, ‘Impact of greenhouse warming on theWest African summer

monsoon’, Climate Dynamics 19 (2002) pp. 499–514.138. P. Liu, G. A. Meehl and G. X. Wu, ‘Multi-model trends in the Sahara induced by increasing CO2’,

Geophysical Research Letters 4 (2002) pp. 1881–1884.139. A. Anyamba and C. J. Tucker, ‘Analysis of Sahelian vegetation dynamics using NOAA-AVHRR

NDVI data from 1981–2003’, Journal of Arid Environments 63 (2005) pp. 596–614.140. L. Olsson, L. Eklundh and J. Ardo, ‘A recent greening of the Sahel—trends, patterns and potential

causes,’ Journal of Arid Environments 63 (2005) pp. 556–566.141. Anyamba and Tucker (note 139).142. See www.fao.org143. J. Grolle, ‘Heavy rainfall, famine, and cultural response in the West African Sahel: the Muda of

1953–54’, Geojournal 43 (1997) pp. 205–214.144. B. Thebaud and S. Batterby, ‘Sahel pastoralists: opportunism, struggle, conflict and negotiation. A

case study from eastern Niger’, Global Environmental Change 11 (2001) pp. 69–78.145. K. Keita. Conflict and Conflict Resolution in the Sahel: The Tuareg Insurgency in Mali. Strategic

Studies Institute Monograph (1998) 46 pp. Downloaded on 13.10.2004 from http://www.carlisle.army.mil/ssi/pubs/pubResult.cfm/hurl/PubID¼200/.

146. Ibid.147. Ibid, p. 20.148. K. Maynard, J. F. Royer and F. Chauvin, ‘Impact of greenhouse warming on theWest African summer

monsoon’, Climate Dynamics 19 (2002) pp. 499–514.149. J. F. B. Mitchell, T. C. Johns, W. Ingram and J. A. Lowe, ‘The effect of stabilising the atmospheric

carbon dioxide concentrations on global and regional climate change’, Geophysical Research Letters27 (2000) pp. 2977–2980.

150. Brooks (note 1).151. A. McMichael, A. Githeko, R. Akhtar, R. Carcavallo, D. Gubler, A. Haines, S. Kovats, P. Martens,

J. Patz and A. Sasaki, ‘Human health’, In IPCC [McCarthy, J., Canziani, O., Leary, N., Dokken,


D. and White, K., eds.] Climate Change 2001: Impacts, Adaptation, and Vulnerability. (Cambridge:Cambridge University Press 2001).

152. R. Ragab and C. Prudhomme, ‘Climate Change and Water Resources Management in Arid and Semi-arid Regions: Prospective and Challenges for the 21st Century’, Biosystems Engineering 81 (2002),pp. 3–34.

153. H. Fathy. Architecture for the Poor (Chicago: University of Chicago Press 2000).154. S. Rodrıguez, X. Querol, A. Alastuey, M-M. Viana, M. Alarcon, E. Mantilla, C. R. Ruiz, ‘Compara-

tive PM10–PM2.5 source contribution study at rural, urban and industrial sites during PM episodes inEastern Spain’, Science of the Total Environment 328 (2004) pp. 95–113.

155. J. M. Prospero and P. J. Lamb, ‘African Droughts and Dust Transport to the Caribbean: ClimateChange Implications’, Science 304 (2003) pp. 1024–1027.

156. D. W. Griffin and C. A. Kellogg, ‘Dust Storms and Their Impact on Ocean and Human Health: Dust inEarth’s Atmosphere’, EcoHealth 1 (2005) pp. 284–295.

157. T. D. Jickells, Z. S. An, K. K. Andersen, A. R. Baker, G. Bergametti, N. Brooks, J. J. Cao, P. W. Boyd,R. A. Duce, K. A. Hunter, H. Kawahata, N. Kubilay, J. La Roche, P. S. Liss, N. Mahowald,J. M. Prospero, A. J. Ridgwell, I. Tegen, R. Torres ‘Global Iron Connections: Desert Dust, OceanBiogeochemistry and Climate’, Science 307 (2005) pp. 67–71.

158. M. O. Andreae, ‘Raising Dust in the Greenhouse’, Nature 380 (1996) pp. 389–390.159. I. Tegen, A. A. Lacis and I. Fung, ‘The Influence on climate forcing of mineral dust from disturbed

soils’, Nature 380 (1996) pp. 419–422.160. N. Brooks, Dust-climate interactions in the Sahel-Sahara zone of northern Africa, with particular

reference to late twentieth century Sahelian drought. PhD thesis (Norwich: Climatic ResearchUnit/University of East Anglia 2000).

161. N. Brooks and M. Legrand, ‘Dust variability over Northern Africa and rainfall in theSahel’, In Linking Climate Change to Land Surface Change (Dordrecht/Boston/London: KluwerAcademic Publishers 2000) pp. 1–25.

162. J. P. Dunion and C. S. Velden, ‘The Impact of the Saharan Air Layer on Atlantic Tropical CycloneActivity’, Bulletin of the American Meteorological Society 85 (2003) pp. 353–365.

163. C. K. Folland, T. N. Palmer and D. E. Parker, ‘Sahel rainfall variability and worldwide sea tempera-tures, 1901–85’, Nature 320 (1986) pp. 602–606.

164. P. Alpert, Y. J. Kaufman, Y. Shay-El, D. Tanre, A. da Silva, S. Schubert and J. H. Joseph, ‘Quanti-fication of dust-forced heating of the lower troposphere’, Nature 395 (1998) pp. 367–370.

165. S. E. Schollaert and J. T. Merrill, ‘Cooler seas surface west of the Sahara Desert correlated to dustevents’, Geophysical Research Letters 25 (1998) pp. 3529–3532.

166. Prospero and Lamb (note 155).167. R. L. Miller, I. Tegen and J. Perlwitz, ‘Surface radiative forcing by soil dust aerosols and the

hydrologic cycle’, Journal of Geophysical Research D Atmospheres, 109 (2004) D04203,doi:10.1029/2003JD004085.

168. Ibid.169. Jickells et al. (note 157).170. Ibid.171. Ibid.172. R. Swap, M. Garstang, S. Greco, R. Talbot and P. Kallberg, ‘Saharan dust in the Amazon basin’,

Tellus 44 (1992) pp. 133–149.173. P. M. Cox, R. A. Betts, C. D. Jones, S. A. Spall and I. J. Totterdell, ‘Acceleration of global warming

due to carbon-cycle feedback in a coupled climate model’, Nature 408 (2000) pp. 184–187.174. R. A. Betts P. M. Cox, M. Collins, P. P. Harris C. Huntingford and C. D. Jones, ‘The role of eco-

system-atmosphere interactions in simulated Amazonian precipitation decrease and forest diebackunder global climate warning,’ Theoretical and Applied Climatology 78 (2004) pp. 157–175.

175. C. Huntingford, P. P. Harris, N. Gedney, P. M. Cox, R. A. Betts, J. Marengo and J. H. C. Gash, ‘Usinga GCM analogue model to investigate the potential for Amazonian forest dieback’, Theoretical andApplied Climatology 78 (2004) pp. 177–185.

176. S. P. Harrison, K. E. Kohfeld, C. Roelandt and T. Claquin, ‘The role of dust in climate changes today,at the last glacial maximum and in the future’, Earth-Science Reviews 54 (2001) pp. 43–80.

177. Brooks and Legrand (note 161).178. B. Marticorena and G. Bergametti, ‘Modeling the atmospheric dust cycle: 1. Design of a soil-derived

dust emission scheme’, Journal of Geophysical Research 100 (1995) pp. 16415–16430.179. S. C. Alfaro, A. Gaudichet, L. Gomes and M. Maille, ‘Modeling the size distribution of a soil aerosol

produced by sandblasting’, Journal of Geophysical Research 102 (1997) pp. 11239–11249.


180. R. Washington, M. Todd, N. J. Middleton and A. S. Goudie, ‘Dust-Storm Source Areas Determinedby the Total Ozone Monitoring Spectrometer and Surface Observations,’ Annals of the Association ofAmerican Geographers 93 (2003) pp. 297–313.

181. Brooks and Legrand (note 161).182. TOMS measures the amount of ultraviolet radiation reflected by the Earth system (i.e. the surface

and overlying atmosphere combined) at different wavelengths; dust is detected using a method thataccounts for the different reflective properties of different substances at these wavelengths(Washington et al., 2003 – see endnote 179). The IDDI detects dust using a different techniquebased on the reduction of infra-red radiation escaping the Earth’s atmosphere due to the coolingeffect of atmospheric aerosols resulting from their blocking of incoming solar radiation, coupledwith their absorption of infra-red radiation emitted by the Earth’s surface (Brooks and Legrand,2000 – see endnote 157). The TOMS AI and the IDDI data are therefore derived from verydifferent principles to detect dust, so it should not be surprising that there are differences in theresulting data.

183. Prospero and Lamb (note 155).184. Washington et al. (note 180).185. Ibid.186. N. M. Mahowald and J-L. Dufresne, ‘Sensitivity of TOMS aerosol index to boundary layer height:

Implications for detection of mineral aerosol sources’, Geophysical Research Letters 31 (2004)L03103.

187. Brooks and Legrand (note 161).188. A. S. Goudie and N. J. Middleton, ‘Saharan dust storms: nature and consequences’, Earth-Science

Reviews 56 (2001) pp. 179–204.189. Drake et al. (Note 14).190. Marticorena and Bergametti (note 178).191. G. M. N’Tchayi, J. Bertrand, M. Legrand and J. Baudet, ‘Temporal and spatial variations of the atmos-

pheric dust loading throughout West Africa over the last thirty years’, Annales Geophysicae 12 (1994)pp. 265–273.

192. I. Tegen and I. Fung, ‘Contribution to the atmospheric mineral aerosol load from land surface modi-fication’, Journal of Geophysical Research 100, no. D9 (1995) pp. 18707–18726

193. I. Tegen, A. A. Lacis and I. Fung, ‘The Influence on climate forcing of mineral dust from disturbedsoils’, Nature 380 (1996) pp. 419–422.

194. C. J. Tucker, H. E. Dregne and W. W. Newcomb, ‘Expansion and contraction of the Sahara Desertfrom 1980 to 1990’, Science 253 (1991) pp. 299–301.

195. C. J. Tucker, W. W. Newcomb and H. E. Dregne, ‘AVHRR data sets for determination of desertspatial extent’, International Journal of Remote Sensing 15 (1994) pp. 3547–3565.

196. S. E. Nicholson and C. J. Tucker, ‘Desertification, drought, and surface vegetation: an examplefrom the West African Sahel’, Bulletin of the American Meteorological Society 79 (1998)pp. 815–829.

197. S. D. Prince, E. B. DeColstoun, and L. L. Kravitz, ‘Evidence from rain-use efficiencies does notindicate extensive Sahelian desertification’, Global Change Biology 4 (1998) pp. 359–274.

198. Prospero and Lamb (note 155).199. Brooks and Legrand (note 161).200. Goudie and Middleton (note 188).201. Washington et al. (note 180).202. Brooks and Legrand (note 161).203. Prospero and Lamb (note 155).204. I. Chiapello and C. Moulin, ‘TOMS and METEOSAT satellite records of the variability of Saharan

dust transport over the Atlantic during the last two decades (1979–1997)’, Geophysical ResearchLetters 29 (2002) article no. 1176.

205. P. Ginoux, J. M. Prospero, O. Torres and M. Chin, ‘Long-term simulation of global dust distributionwith the GOCART model: Correlation with North Atlantic Oscillation’. Environmental Modellingand Software 19 (2004) pp. 113–128.

206. P. de Menocal, ‘Abrupt onset and termination of the African Humid Period: rapid climate responses togradual insolation forcing’, Quaternary Science Reviews 19 (2000) pp. 347–361.

207. J.-L. Rajot, S. C. Alfaro, L. Gomes and A. Gaudichet, ‘Soil crusting on sandy soils and its influence onwind erosion’, Catena 53 (2003) pp. 1–16.

208. Brovkin (note 135).209. Claussen et al. (note 136).


210. D. W. Griffin and C. A. Kellogg, ‘Dust Storms and Their Impact on Ocean and Human Health: Dust inEarth’s Atmosphere’, EcoHealth 1 (2004) pp. 284–295.

211. Rodriguez et al. (note 154).212. J. Prospero, ‘Long-range transport of mineral dust in the global atmosphere: Impact of African dust on

the environment of the southeastern United States’, Proceedings of the National Academy of Sciences96 (1999) pp. 3396–3403.

213. W. S. Rajkumar and A. S. Chang, ‘Suspended particulate matter concentrations along the East-WestCorridor, Trinidad, West Indies’, Atmospheric Environment 34 (2000) pp. 1181–1187.

214. P. Laval, ‘Desert Silicosis’, Maroc Med. 47 (1967) pp. 357–360.215. C. Fossati, ‘Possibilities and frequency of pulmonary silicosis in inhabitants of the Libyan desert’,

Med. Lav. 60 (1969) pp. 144–149.216. R. K. Patial, ‘Mountain desert silicosis’, Journal of the Association of Physicians of India 47 (1999)

pp. 503–504.217. M. L. Mathur and R. C. Choudhary, ‘Desert Lung Syndrome in rural dwellers of the Thar Desert,

India’, Journal of Arid Environments 35 (1997) pp. 559–562.218. C. Brambilla, J. Abraham, E. Brambilla, K. Benirschke and C. Bloor, ‘Comparative pathology of

silicate pneumoconiosis’, American Journal of Pathology 96 (1979) pp. 149–169.219. M. S. Nouh, ‘Is the desert lung syndrome (nonoccupational dust pneumoconiosis) a variant of pulmonary

alveolarmicrolithiasis?Report of 4 caseswith reviewof the literature’,Respiration55 (1989) pp. 122–126.220. Mathur and Choudhary (note 217).221. S. Rutherford, E. Clark, G. McTainsh, R. Simpson and C. Mitchell, ‘Characteristics of rural dust

events shown to impact on asthma severity in Brisbane, Australia’, International Journal of Bio-meteorology 42 (1999) pp. 217–225.

222. K. Gyan, W. Henry, S. Lacaille, A. Laloo, C. Lamsee-Ebanks, S. McKay, R. M. Antoine and M. A.Monteil, ‘African dust clouds are associated with increased paediatric asthma accident and emergencyadmissions on the Caribbean island of Trinidad’, International Journal of Biometeorology 49 (2005)pp. 371–376.

223. J. M. Prospero, E. Blades, G. Mathison and R. Naidu, ‘Interhemispheric transport of viable fungi andbacteria from Africa to the 2 Caribbean with soil dust’, Aerobiologia 21 (2005) pp. 1–19.

224. H. Osbahr, personal communication.225. J. Giles, ‘The dustiest place on Earth’, Nature 434 (2005) pp. 816–819.226. J. Keenan, ‘Terror in the Sahara: the implications of US imperialism for North & West Africa’,

Review of African Political Economy 31 (2004) pp. 475–496.227. D. Cyranoski, ‘China plans clean sweep on dust storms,’ Nature 421 (2003) pp. 101–101.228. Jickells et al. (note 157).229. G. M. N’Tchayi, J. Bertrand, M. Legrand and J. Baudet, ‘Temporal and spatial variations of the

atmospheric dust loading throughout West Africa over the last thirty years’, Annales Geophysicae12 (1994) pp. 265–273.

230. N. Brooks, C. Huntingford, J. H. C. Gash, B. Kjellen, J. Kohler, R. Starkey and R. Warren, ‘Climatestabilisation and “dangerous” climate change: A review of the issues’, Climatic Change (submitted).

231. Brooks (note 29).232. R. B. Alley, ‘Palaeoclimatic insights into future climate challenges’, Philosophical Transactions of

the Royal Society of London A 361 (2003) pp. 1831–1848.233. S. H. Schneider, ‘What is ‘dangerous’ climate change?’, Nature 411 (2001) pp. 17–19.234. S. Dessai and M. Hulme, Does climate change need probabilities? Tyndall Centre Working Paper 34

(Norwich: Tyndall Centre for Climate Change Research/University of East Anglia 2003).235. Brooks (note 29).236. N. Brooks, W. N. Adger and P. M. Kelly, ‘The determinants of vulnerability and adaptive capacity at

the national level and the implications for adaptation’, in W. N. Adger, N. Arnell and E. L. Tompkins(eds.) Adaptation to Climate Change: Perspectives Across Scales. Global EnvironmentalChange PartA 15 (2005) pp. 151–162.

237. Brooks (note 29).238. Thebaud and Batterby (note 145).239. di Lernia (note 69).240. B. Lim and E. Spanger-Siegfried, Adaptation Policy Frameworks for Climate Change: Developing

Strategies, Policies and Measures. (Cambridge: UNDP/Cambridge University Press 2005).241. F. A. Hassan, ‘Nile floods and political disorder in early Egypt’, in H. N. Dalfes, G. Kukla, and

H. Weiss (eds.) Third Millennium BC Climate Change and Old World Collapse, NATO ASI SeriesVol. I 49 (Berlin Heidelberg: Springer-Verlag 1997) pp. 711–723.


242. di Lernia and Palombini (note 70).243. N. Brooks, S. di Lernia, N. Drake, M. Raffin and T. Savage, ‘The geoarchaeology of Western

Sahara: Preliminary results of the first Anglo-Italian expedition in the “free zone”’, Sahara 14(2003) pp. 63–80.

244. J. Keenan, ‘The theft of Saharan rock art’, Antiquity 74 (2000) pp. 287–288.245. J. Keenan, ‘The Lesser Gods of the Sahara’, Public Archaeology 2 (2002) pp. 131–150.246. Brooks et al. (note 177).247. J. Keenan, personal communication.


The Climate-Environment-Society Nexus in the Sahara from ... · PDF fileLERNIA, NICK DRAKE,...

Documents

Transcript of The Climate-Environment-Society Nexus in the Sahara from ... · PDF fileLERNIA, NICK DRAKE,...