Introduction

CO PY RI GH TE D MA TE RI AL I Global contexts C HA PTE R 1 Introduction Sarah E. Metcalfe and David J. Nash 1.1 Why the tropics matter 1.1.1 Defining the tropics In its strictest sense, the term ‘tropics’ refers to those parts of the world that lie between the Tropic of Cancer (23.4378 °N) and the Tropic of Capricorn (23.4378 °S). These latitudinal boundaries mark, respectively, the most northerly and southerly position at which the Sun may appear directly overhead at its zenith. Indeed, the word ‘tropical’ comes from the Greek tropikos, meaning ‘turn’, since the tropics of Cancer and Capricorn mark the latitudes at which the Sun appears to turn in its annual motion across the sky. Unfortunately, the outer boundary of the tropics sensu lato cannot be defined in such rigid astronomical terms. Certainly latitude is a major factor determining the distribution of tropical climatic zones, through its control on solar radiation receipt (Fig. 1.1), but regions with distinctive climatological, physical or biological characteristics are not easily delimited by linear boundaries. The tropics include a diverse range of environments and climates (see Chapter 2). Rather than being uniformly hot and wet, the area between the tropics of Cancer and Capricorn encompasses some of the wettest regions on Earth (e.g. the rainforests of western Amazon and central Congo basins) as well as some of the driest (e.g. the Atacama Desert of northern Chile and Peru). The one feature common to all tropical climates is a relatively limited seasonal fluctuation in insolation and tem- perature. Instead, differences in the quantity and temporal distribution of available moisture account for regional and seasonal variability (Savage et al., 1982). Authors such as Reading et al. (1995) have provided useful overviews of the various attempts to define the climates of the tropics. Some of the most widely used classifications are based directly upon meteorological parameters such as rainfall and temperature. The classic Köppen–Geiger system (Fig. 1.2), for example, centres on the concept that natural vegetation is the best expression of climate, with climate zone boundaries positioned with vegetation distribution in mind. The Köppen– Geiger scheme combines average annual and monthly temperatures and precipitation, and the seasonality of precipitation. Köppen (1936) defined tropical climates as those exhibiting a constant high temperature (at sea level and low elevations), with all 12 months of the year having average temperatures of 18 °C or higher. This classification excludes cooler highland regions (defined as areas above 900 m elevation), which comprise around 25% of the total land area within the tropics (Reading et al., 1995). These regions still receive high amounts of solar radiation and do not have a pronounced winter season, but temperatures may be sufficiently depressed to affect biological activity. Rainfall levels and the seasonal distribution of precipitation are then used to subdivide tropical climates into tropical rainforest (Af), tropical monsoon (Am), and tropical savanna climates Quaternary Environmental Change in the Tropics, First Edition. Edited by Sarah E. Metcalfe and David J. Nash. © 2012 John Wiley & Sons, Ltd. Published 2012 by John Wiley & Sons, Ltd. 3 4 Chapter 1 20 20°S 30°N 20°S 20 Megajoules/day 10°N 15 10 0°N 15 0°S 10°N 20°S 10 30°N 5 0 5 30°N J 60°N F M 60°N 45°S A M J J A S O N D 0 Fig. 1.1 Solar radiation received at the Earth’s surface assuming an atmospheric transmission coefficient of 0.60 (after McGregor and Nieuwolt (1998) Tropical Climatology, John Wiley & Sons Ltd.). (Aw). Köppen (1936) includes a range of other climate types within the tropics sensu stricto, including tropical and subtropical steppe (BSh), tropical desert (BWh) and humid subtropical climates (Cfa, Cwa). Some highland areas within the tropics also exhibit a temperate climate with dry winters (Cwb). Working from an agricultural perspective, Jackson (1989) split the tropics into three zones (Humid, Wet and Dry, and Dry) according to the level and seasonal distribution of rainfall (Fig. 1.3). This classification recognises the importance of seasonality for agricultural productivity, and is less focused on natural vegetation zones than the Köppen–Geiger scheme. Other attempts to classify climates within the tropics are based around hydro-meteorology, with climate types defined according to the balance of precipitation inputs and evapotranspiration outputs. Garnier (1958), for example, differentiated humid tropical climates according to the number of months in which actual evapotranspiration equals potential evapotranspiration. The ratio of precipitation to potential evapotranspiration has also been used by Middleton et al. (1997), drawing upon Thornthwaite (1948) and Meigs (1953), to define an aridity index for categorising dry tropical climates. In this volume, the astronomical definition of the tropics is used to broadly demarcate the geographical scope of each of the substantive chapters. However, recognising that climate boundaries are fuzzy and mobile in the present day and that climate zones shifted by many degrees of latitude during the various glacials and stadials that characterise the Quaternary Period, coverage in many chapters extends polewards north and south of 23.4378° into the subtropics where appropriate. The Quaternary Period is defined here as encompassing the last 2.58 million years of the Earth’s history (Gibbard et al., 2010), the timescale ratified by the Executive Committee of the International Union of Geological Sciences in June 2009. 1.1.2 Importance of the tropics In comparison with the mid latitude regions of Europe and North America, our understanding of Quaternary palaeoenvironments in the tropics is, at best, patchy for some areas and extremely poor to non-existent in others. As a result, any attempt to expand our understanding of past environmental conditions in low latitude regions is likely to be a valuable contribution to knowledge. However, more significantly, understanding tropical palaeoenvironments may also be key to establishing the drivers of global environmental change. As discussed in section 1.5 of this chapter, the last 10–15 years have seen an increasing recognition of the significance of tropical regions in climate forcing (e.g. Kerr, 2001; Broecker, 2003). The tropical oceans and atmosphere play an important contemporary role in redistributing incoming solar radiation and would have been instrumental in transmitting past variations in radiation receipt to Fig. 1.2 The Köppen–Geiger climate classification system updated with CRU TS 2.1 temperature and VASClimO v1.1 precipitation data for 1951 to 2000 (after Kottek et al., 2006). (See Colour Plate 1) Introduction 5 6 Chapter 1 Fig. 1.3 Classification of the tropics based on the seasonal distribution of rainfall (after Jackson (1989) Climate, water and agriculture in the tropics, Longman; Reading et al. (1995) Humid tropical environments, John Wiley & Sons Ltd.) (See Colour Plate 2) Introduction other parts of the Earth system. Tropical oceans and landmasses also act as sources and sinks of greenhouse gases, with, for example, tropical forests acting as contemporary CO2 sinks (Cox et al., 2000) and tropical oceans (IPCC, 2007) and major river and wetland systems such as the Amazon (Richey et al., 2002) outgassing CO2 to the atmosphere. The decay of vegetation within tropical wetlands is a major source of contemporary biogenic CH4 (Loulergue et al., 2008). Indeed, much of the variation in CH4 concentration recorded in the Antarctic Vostok ice core coincides with fluctuations in the size and extent of tropical lakes and wetlands (cf. Raynaud et al., 1988; Chappellaz et al., 1990; Brook et al., 2000). Tropical forest ecosystems and soils are known to act as important contemporary sources for atmospheric N2O, with N2O emissions typically increasing during wet season conditions and falling during drier periods. Data from the Antarctic EPICA Dome C ice coring site suggest that biospheric changes in the low latitudes may have been instrumental in controlling emissions of N2O on glacial–interglacial timescales (Schilt et al., 2010). The precise mechanism through which this process operated is unknown, but deep water changes in the North Atlantic, and associated Dansgaard–Oeschger (D–O) events, may have had an influence on atmospheric N2O levels, either through indirect changes in low latitude ecosystems and soils or by a direct change in marine N2O production (Schmittner and Galbraith, 2008). Identifying changes in tropical environments over the past 2–3 million years may have considerable resonance for our understanding of the drivers of human evolution. Recent fossil discoveries and advances in the analysis of existing fossil collections, coupled with the emergence of high resolution palaeoclimatic records, have focused attention on the role that past shifts in climate variability may have had in the evolutionary history of African mammalian fauna, including early hominids (de Menocal, 2004). Although this topic is still hotly debated, the basic premise is that large-scale shifts in climate over the course of the last 5–6 million years altered the ecological composition of African landscapes, thereby generating specific faunal adaptation or speciation pressures which ultimately 7 led to genetic selection and innovation. Evidence from Atlantic and Indian Ocean cores suggests that climate in the African subtropics fluctuated between markedly wetter and drier conditions in time with orbital variations. De Menocal (1995) identifies progressive shifts in African climate variability and increasing aridity after 3.0–2.6 Myr, 1.8–1.6 Myr and 1.2–0.8 Myr, coincident with the onset and intensification of high-latitude glacial cycles. Analysis of well-dated mammal fossil databases suggests African faunal assemblage and, perhaps, speciation changes coincident with the appearance of more varied and open habitats at 2.9–2.4 Myr and after 1.8 Myr. These periods roughly coincide with key junctures in hominid evolution, including the emergence of the genus Homo around 2.5 Myr (de Menocal, 2004). Environmental changes, particularly during the late Pleistocene, may also have played a role in shaping pathways for the dispersal of early modern humans around the Earth. For example, corridors formed by pluvial ‘mega-lakes’ during Marine Isotope Stage (MIS) 5 may have provided transSaharan pathways for humans migrating ‘out of Africa’, offering an alternative route to the Nile Valley (Drake et al., 2011). Biogeographical and palaeohydrological evidence (ibid.) suggests that similar migration pathways across the Sahara, in the form of linked lakes, rivers and inland deltas, may have existed during the early Holocene (see Chapter 4). The migration of humans into Australia, either as a single or several successive waves, also appears to have been influenced by global environmental changes. There is still much debate about the timing of the earliest arrivals; the minimum widely-accepted timeframe places this at around 45 kyr BP (e.g. O’Connell and Allen, 2005) with an upper estimate of around 60 kyr BP (e.g. Roberts et al., 1990, 1993, 1994). Regardless, this migration was achieved during the closing stages of the Pleistocene, when sea levels were much lower than they are today (see section 1.2.2 and Fig. 1.5) and an extensive land bridge existed across the Arafura Sea, Gulf of Carpentaria and Torres Strait (Lourandos, 1997). The tropics are also highly important because they support very large numbers of species compared 8 Chapter 1 with other regions of the globe (Mace et al., 2005). This is especially true of the tropical moist forests which show the highest global levels of species and family richness and of endemism. There is increasing concern about the threat posed to tropical ecosystems by both direct human action and by future climate change (itself probably anthropogenic). Although we hear most about the tropical rainforest (e.g. Hubbell et al., 2008), it is the tropical dry forests that have been most affected to date, with about half being lost to cultivation. Mapping of species loss (mammals, birds and amphibians) since AD 1500 shows a significant concentration in tropical latitudes, especially in the tropical Americas and Australasia (Baillie et al., 2004). As well as direct loss of habitat and species, with their economic and cultural values, changes in tropical ecosystems have wider implications because of their role in the global biogeochemical and hydrological cycles. Some of these issues are discussed further in section 1.5 of this chapter. As the chapters within this volume highlight, tropical vegetation and landscape systems have shown considerable sensitivity to climatic changes during the Quaternary Period; by inference, tropical landscapes might be expected to show a similar scale of response to future human-induced and natural environmental changes. Couplings between vegetation cover and the susceptibility of the ground surface to water or wind erosion mean that shifts in vegetation density and type in response to anthropogenic and climatic changes may act to either stabilise or destabilise land surfaces. Tropical fluvial systems, for example, are highly sensitive to external forcings in the form of short and long term shifts in effective precipitation and vegetation cover. The nature of the response within individual fluvial systems reflects the antecedent conditions, the degree and duration of the environmental change, possibly the rate of change, and whether the change is sufficient to trigger in-channel threshold-crossing events (Thomas, 2008). In northeast Queensland, Australia, for example, a long-term deterioration of the rainforest vegetation cover after 78 kyr BP, steepening after 40 kyr BP with a shift toward dry sclerophyll forest, led to widespread soil erosion and the accumulation of fine alluvial fan deposits within fluvial systems fronting the eastern highlands of the Great Dividing Range (Nott et al., 2001; Thomas et al., 2001, 2007). Many tropical environments contain relict landforms (and their associated sediments) formed under previously wetter or drier conditions, which may be reactivated under future climatic change scenarios. Environmental modelling studies in the Kalahari Desert, for example, have suggested that large areas of presently stable and well-vegetated ‘fossil’ Pleistocene sand dunes could be reactivated if changes in wind regime and a reduction in vegetation cover (in response to warming and reduced available moisture) occur as a result of twentyfirst century climate warming (Thomas et al., 2005). 1.2 Development of ideas 1.2.1 Early ideas about tropical environmental change The possibility that high and mid latitude regions had undergone major environmental changes was recognised as early as 1779 when the Swiss aristocrat Horace-Bénédict de Saussure identified granite boulders on the limestone slopes of the Jura ranges that had been transported some 90 km from their source in the Mont Blanc massif (de Saussure, 1779). In keeping with contemporary ideas that the Earth’s features had been shaped by the biblical Great Flood, de Saussure suggested that these ‘erratics’ had been moved by water. Bernard Friedrich Kuhn was the first to propose that the boulders had, in fact, been transported by more extensive glaciers (Kuhn, 1787; de Beer, 1953), a conclusion reached independently some eight years later by James Hutton following a visit to the Jura (Hutton, 1795). John Playfair famously extended these ideas in 1802, and, by the time of the publication of Etudes sur les Glaciers by Louis Agassiz in 1840, the concept of Die Eiszeit or large scale Ice Age in Europe was well established. In contrast, for many years, the dominant view of the tropics was that they had seen very little climatic change, with core areas such as Amazonia remaining unaffected by the cycles of glaciation and deglaciation that drove massive environmental Introduction changes in higher latitudes (Richards, 1952). This was despite the suggestion made by Louis Agassiz, after mistaking deeply weathered bedrock for glacial diamicton during a visit to Brazil in 1865– 1866, that the western Amazon basin had been glaciated (Agassiz, 1868). As early as 1850, the Scottish missionary and explorer David Livingstone had recognised that salt accumulations in the Makgadikgadi Depression of Botswana were ‘the remains of the very slightly brackish lakes of antiquity’ (Livingstone, 1857: 67). However, some of the main advances in our understanding of low latitude palaeoenvironments were made in the USA (see Goudie, 1999). John Strong Newberry, for example, suggested that the landscapes of the Colorado Plateau were ‘formerly much better watered than they are now’ (1861: 47). In 1863, Thomas Francis Jamieson was the first to propose that wetter conditions and higher lake levels in the southwest USA were equated with high latitude glacial episodes (a concept often termed the ‘glacial = pluvial’ hypothesis). This idea was adopted by Israel Russell (1885) and Grove Karl Gilbert (1890) to explain the origins of strandlines within the Pleistocene ‘pluvial’ lakes Lahontan and Bonneville (Fig. 1.4). The notion that low latitude pluvials were synchronous with high latitude glacials was widely accepted and was ultimately assumed to apply across the tropics. The corollary of this view, that post-glacial times were characterised by desiccation, was also widely applied (Goudie, 1972), most notably in the Asian and African tropics and subtropics (Goudie, 1999). In southern Africa, for example, Schwarz (1923) proposed a grandiose scheme to divert rivers from the north to flood the Kalahari Basin as a means to ameliorate a supposed progressively drying climate. (a) (b) 1.2.2 The twentieth century revolution By the mid 1940s, challenges to the post-glacial desiccation and ‘glacial = pluvial’ hypotheses began to emerge. One of the most important conceptual advances was the recognition that some tropical areas that are now relatively moist, far from progressively desiccating had been significantly drier in the past. The main evidence for this came first from the identification of ancient dunefields in Fig. 1.4 (a) Sketch of Lake Bonneville shorelines and (b) Map of Lake Bonneville by G.K. Gilbert (from Gilbert, 1890, images courtesy of USGS). 9 10 Chapter 1 Texas (Price, 1944), and then vegetated ergs along the equatorward margins of the southern Sahara (Grove, 1958; Grove and Warren, 1968), northern Kalahari (Grove, 1969) and Indian deserts (Goudie et al., 1973). Prior to the advent of luminescence dating in the 1980s, the ages of these aeolian deposits could only be estimated relative to sediments that could be radiocarbon dated, but their mere existence was a nail in the coffin for progressive desiccation. The 1970s represented a major shift in our understanding of low latitude palaeoenvironments, as detailed records from lake basins in tropical Africa (e.g. Grove and Goudie, 1971; Grove et al., 1975; Street and Grove, 1976) and elsewhere (cf. Street-Perrott et al., 1979) started to be published. Views of the stability of the tropical rainforest also changed (Flenley, 1979). With these studies, it became apparent that the story of the tropics was much more complex than previously thought, with many areas exhibiting fluctuating rather than consistently high lake levels around the time of the Last Glacial Maximum (LGM). Compilations of global lake level fluctuations (e.g. Street-Perrott et al., 1979) served to demonstrate that, in many ways, what happened in the southwest USA, the home of the ‘glacial = pluvial’ hypothesis, was the exception rather than the norm. The picture that emerges today, as summarised by each of the regional chapters in this volume, is that the magnitude and timing of climate change in different parts of the tropics and subtropics is considerably more complex than pioneers such as Gilbert and Russell ever could have envisaged. In parallel with the growth in knowledge about terrestrial tropical environments, our understanding of Quaternary stratigraphy in tropical oceans has been revolutionised since the 1950s (Imbrie and Imbrie, 1979). This has been due primarily to the introduction of new equipment for coring offshore and deep-ocean sediments. Damuth and Fairbridge (1970), for example, used evidence from deep-sea piston cores taken in the Guiana Basin to suggest that an arid to semi-arid climate dominated large portions of equatorial South America during the Pleistocene glacial phases. Similarly, analyses of multiple cores off northwest Africa by Diester-Haas (1976) revealed fluctuations in the extent of the Sahara during the late Pleistocene. Some of the longest and highest resolution records available for the tropics now come from marine settings, and provide important insights into ocean palaeotemperatures, terrestrial chemical environments and variations in the offshore transport of dust, pollen and fluvial sediments (e.g. Larrasoaña et al., 2003; Peterson and Haug, 2006) (see Chapter 3, section 3.2.2). The technology used to extract marine cores has been adapted and utilised on land, such that a number of long terrestrial records are now also available for the tropics (e.g. Trauth et al., 2003). At the interface between land and the oceans, the Quaternary has seen major changes in relative sea level. A number of different factors may be involved, especially locally. However, at the global scale, glacio-eustatic change dominates, reflecting the volume of water locked up in ice sheets and glaciers (with glacial or stadial periods being marked by low sea levels). Although the association between changes in ice volume and sea level was put forward in the early twentieth century, major advances in reconstructing sea level were made during the 1960s and 1970s. Key sea level reconstructions (covering about the last 400 kyr) have come from tropical areas, primarily the Huon Peninsula of Papua New Guinea (Aharon and Chappell, 1986) (see Chapter 6) and Barbados (Fairbanks, 1989). Both these are based on dated sequences of coral reefs. An updated version of Fairbanks’ reconstruction for the period since the LGM is shown in Fig. 1.5. This indicates that sea level was 120–125 m lower than present at the LGM. Fairbanks identified two periods of very rapid rise (>20 m) associated with meltwater pulses 1A and B, which he dated to around 12 kyr and 9.5 kyr BP. These events were later re-dated to 14 and 11 kyr BP, following a reassessment of the record using U-Th dating (Bard et al., 1990). Bard et al. also reported sea level of +5 to +10–m in the last interglacial (MIS 5e). This and subsequent studies have confirmed the coincidence of periods of high sea level with insolation maxima, consistent with Milankovitch forcing (see section 1.4 of this chapter). The impact of these changes in sea level was particularly pronounced in areas with exten- Introduction 11 Fig. 1.5 Composite record of relative sea level change over the last 32 kyr, based on data from Barbados. Data from Peltier and Fairbanks (2008) IGBP PAGES/World Data Center for Paleoclimatology, Data series 2008-101. sive continental shelves affecting marine currents, regional groundwater levels and the ease of migration of terrestrial organisms including humans (see especially Chapters 6 and 7). Advances in our understanding of tropical palaeoenvironments have been prompted, in part, by the availability of new avenues for environmental reconstruction, but also reflect the development of new chronological techniques (Goudie, 1999). The introduction of radiocarbon dating in the 1950s, for example, meant that it was possible, for the first time, to obtain age estimates from late Quaternary sediments and landforms rather than having to rely on stratigraphic correlation. The radiocarbon revolution was followed in the 1960s by the development of potassium-argon and uranium-series dating, dendrochronology and palaeomagnetism. These chronological tools were refined in the 1970s and 1980s, with new approaches such as amino- stratigraphy, electron-spin resonance, luminescence and cosmogenic radionuclide exposure dating introduced in more recent decades. For many of these techniques, the availability of mass spectrometry has permitted high temporal resolution dating of materials, including the micro-sampling of cave deposits (e.g. Wang X et al., 2007; Wang Y et al., 2008) and geochemical sediments such as calcrete (e.g. Candy et al., 2004). Two examples serve to highlight the importance of the new dating tools for our understanding of tropical palaeoenvironments. First, the development of optically-stimulated luminescence (OSL) dating since the 1980s has allowed the age of deposition of a wide range of carbon-poor sediments to be determined, most notably those preserved within fossil dunes and other aeolian deposits (cf. Singhvi and Porat, 2008). This has led to the establishment of detailed chronological frameworks for 12 Chapter 1 many of the world’s desert regions (cf. Munyikwa, 2005) as well as major advances in our understanding of how aeolian dunes evolve over time (e.g. Telfer and Thomas, 2006). Second, the rapid advances in cosmogenic radionuclide analysis in the last decade have provided a basis for exposure ‘dating′ of landforms, the quantification of erosion rates and other geologic applications in areas where opportunities for any form of chronological investigation were once extremely limited. Cosmogenic radionuclide dating has been used, for example, to establish the timing of dunefield initiation in central Australia (Fujioka et al., 2005) and, alongside other techniques, to estimate residence times for groundwater in the Nubian Aquifer beneath the Western Desert in Egypt (Patterson et al., 2005). Alongside chronological developments, there have been a number of other methodological improvements, including the application of an increasingly sophisticated range of field and laboratory approaches. These include new techniques for sedimentological and geochemical analysis which have offered important insights into Quaternary depositional environments. Amongst the most significant of these was the advent of stable isotope analyses of sediments and biological remains in the 1950s. Oxygen isotope analysis, in particular, pioneered by Cesare Emiliani (1955), is now one of the most important tools in Quaternary stratigraphy and is routinely applied in a variety of terrestrial and marine contexts to reconstruct environmental signals such as palaeotemperature, water balance (P–E), precipitation source and amount (Leng, 2006). Stable carbon isotope analysis can be undertaken on either inorganic (authigenic calcite, biogenic carbonate), or organic C. In combination with measurements of C/N, δ13Corganic is widely used to determine the sources (C3 or C4 terrestrial vegetation, aquatic macrophytes, algae) of organic matter coming into lacustrine systems. The application of compound specific δ13C analysis is particularly effective in this regard (e.g. Street-Perrott et al., 2004). More recent studies (e.g. Chase et al., 2009), have utilised variations in stable nitrogen isotope analyses as a means of establishing past rainfall levels. A large (and growing) number of palaeoecological techniques are now available for environmental reconstruction. Most early attempts to reconstruct changes in tropical flora were heavily reliant upon pollen analyses. However, it is now possible to utilise other plant remains such as macrofossils (e.g. those preserved within rodent middens; Betancourt et al., 1990; Pearson and Dodson, 1993; Holmgren et al., 2007) and phytoliths (e.g. Parker et al., 2004), not only to reconstruct terrestrial vegetation changes but also to identify shifts in CO2 concentration (Beerling and Woodward, 1993). Changes in terrestrial aquatic environments can be identified through the analysis of molluscs, diatoms and ostracods (e.g. Fritz et al., 1999; Holmes and Engstrom, 2005), whilst our understanding of changes in marine environments has been revolutionised through the analysis of foraminifera and other microorganisms such as radiolaria and coccoliths (cf. Lowe and Walker, 1997). The development of transfer functions – essentially variants on multiple linear regression models employed to establish relationships between biological data and environmental variables – now permits palaeoenvironmental parameters to be reconstructed quantitatively from fossil floral and faunal assemblages (e.g. Birks and Birks, 1980; Birks, 2005). The need for such transfer functions to reflect biologically meaningful relationships has to be borne in mind, however. Finally, our understanding of tropical and subtropical environmental variability in recent centuries has greatly improved thanks to new efforts to tap the wealth of information contained within annually resolved proxies (e.g. corals, tree rings, speleothems). Climate chronologies derived from historical documentary materials are now available, for example, for large areas of Africa (e.g. Nicholson, 2000, 2001; Nash and Endfield, 2002, 2008; Grab and Nash, 2010; Nash and Grab, 2010) and show remarkable agreement with regional tree ring records (e.g. Therrell et al., 2006) and fossil coral (Zinke et al., 2004, 2005). 1.2.3 Advances in modelling The application of computer modelling to palaeoclimate studies is now central to efforts to synthesise and understand change in climate systems and environments. The role of factors such as insolation forcing, tectonism and vegetation feedbacks have Introduction all been explored in relation to tropical regions, with a particular emphasis on their impacts on monsoons. The application of modelling to tropical palaeoclimates is explored explicitly in Chapter 9, so this section will provide only a brief introduction. The reader is also referred to a number of reviews of climate modelling, including those of McGuffie and Henderson Sellers (2001), Cane et al. (2006) and the IPCC (2007). Climate models are derived from weather forecasting models, originally conceived by John von Neumann who founded the GFDL (Geophysical Fluid Dyamics Laboratory). The first comprehensive general circulation experiments were undertaken by Smagorinsky (1963) and by 1965 it was realised that computer models could also be used to explore past climates. There are a range of model types from 1-D energy balance models, to 3-D general circulation models (GCMs). Pioneering work on the application of modelling to palaeoclimate was carried out by Gates (1976a,b) and Manabe and Hahn (1977). This work brought climate modellers and palaeoclimatologists together, as palaeodata (e.g. CO2 concentrations, sea-surface temperatures (SSTs), ice sheet extents) were needed to set model boundary conditions. Through the 1980s, John Kutzbach and his co-authors led the way in exploring drivers of change in the monsoon using the NCAR CCM (Community Climate Model) (e.g. Kutzbach and Guetter, 1986; Prell and Kutzbach, 1987; Ruddiman and Kutzbach, 1989). This effort was complemented by significant developments in data-model comparisons through COHMAP with a particular focus on 18 k and 6 k 14 C yr BP (COHMAP Members, 1988). This tradition has been continued through the PMIP (Palaeoclimate Modelling Intercomparison Project). PMIP1 used CGMs with atmosphere only, or with slab ocean, while PMIP2 used coupled ocean– atmosphere (–vegetation) models (Braconnot et al., 2007). The results from PMIP are discussed in more detail in Chapter 9. The development of fully coupled ocean– atmosphere models (see Chapter 9) represented a major challenge due to the very different response times and resolutions of these two key elements of the climate system. Early coupled models such as the UK Met Office’s HadCM2 required flux adjust- 13 ments to keep the two elements together, but this wasn’t needed in later models. The advent of these coupled models allowed annual climatologies and seasonal cycles to be reproduced (IPCC, 2001). This has been vital in efforts to model the El Niño Southern Oscillation (ENSO; see Chapter 9). The most recent development is the use of fully coupled Earth System Models such as HadGEM2ES (dynamic vegetation response) and ECHAM5/ JSBACH-MPIOM (e.g. Dallmeyer et al., 2010). GCMs now dominate, but simpler models are still used where long time series are a key requirement (e.g. Crowley et al., 1992) and the run times of more comprehensive models would still be prohibitive even with the significant computing power now available. These models of intermediate complexity continue to play an important role in helping to understand long term climate change, including the role of Milankovitch cycles and transitions between different climate modes (interglacial/ glacial). Groot et al. (2011) use one of these models, CLIMBER (see also Chapter 9), to help interpret the arboreal pollen record from the Fuquene Basin in Colombia between 284 kyr and 27 kyr BP (see Chapter 8). Models play a very important part in helping our understanding of tropical climate change. They have also helped us to appreciate the importance of the tropics in driving climate change, especially the role of the tropical oceans (Hostetler et al., 2006) and feedbacks from greenhouse gases (particularly methane) (Loulergue et al., 2008). Unfortunately, there are still some parts of the world where climate models struggle to reproduce modern climate, and hence one can have little confidence in their use in palaeoclimatic studies. This is particularly the case in areas of complex terrain. The use of regional scale models and finer resolution GCMs can help to address this (e.g. Hostetler et al., 1994). 1.3 Establishment of the tropical climate system In the popular imagination, the tropics are both warm and wet, and it is the case that 56% of total global precipitation falls in the tropics (Wang and 14 Chapter 1 Ding, 2008). As noted above, in tropical climates it is the distribution of rainfall, rather than temperature, which determines the seasons, and the seasonality and overall amount of precipitation that distinguishes the major tropical environments: rainforest, savanna and desert (Bridgman and Oliver, 2006). The reader is referred to Chapter 2 for more on tropical climatology, but in this section some background is given on two key elements of the tropical climate system: the monsoon and ENSO. Although the dominant dynamic controls on the tropical climate are the location of the Intertropical Convergence Zone (ITCZ) and the subtropical high pressure systems (Hadley cells), perhaps the best known feature of the tropical climate is the monsoon. The name comes from the Arabic word ‘mausim’ for a seasonal reversal of winds recognised in the Arabian Sea and Indian Ocean and exploited by Greek and Arab traders. The importance of this seasonal change in winds and the resulting precipitation to trade (sailing ships) and livelihoods (crops etc.) was recognised early on. Failure of the monsoon rains in 1866 and 1871 led to the establishment of the India Meteorological Department in 1875 and the subsequent work of H. F. Blanford and Sir G.T. Walker to forecast and understand monsoon variability. In Walker’s case, his analysis of meteorological data from around the globe led to the recognition of the Southern Oscillation (identifying the importance of change in the eastern tropical Pacific) and its link to monsoon rainfall (Walker, 1924). The Southern Oscillation is discussed further below. Although there are various delineations of monsoon areas – Wang and Ding (2008) suggest that they cover 19.4% of the Earth’s surface – monsoon rain accounts for 30.8% of total global precipitation. Given that these areas are home to more than 55% of the world’s human population (McGregor and Niewolt, 1998) and support the world’s most biologically diverse and ecologically complex terrestrial ecosystems (tropical forests) (Wilson, 1986) their significance is evident and changes in monsoon climates both in the past and into the future are important to understand. As described in Chapter 2 (and Chapters 4 to 8 for regional details), the monsoon climate is characterised by a reversal of prevailing wind direction and a contrast between a wet summer and dry winter. This seasonal reversal in wind direction (conventionally a change of ≥120° between January and July) is driven by differential heating of oceans and continents. Evaporation and condensation processes add strength to the system and Coriolis results in the curved trajectories of monsoon winds. Traditionally monsoons were associated with Africa, Asia (India and East Asia) and Australia, being best developed in South and Southeast Asia. More recently monsoon-type systems have also been identified in the tropical Americas, although not fulfilling all the original criteria (McGregor and Nieuwolt, 1998). In this volume, we adopt this wider definition of monsoons. It is clear that monsoons are an enduring feature of the Earth’s climate system, with monsoon climates recognised in deposits from ancient super continents (e.g. Pangaea) (Clift and Plumb, 2008). It seems likely that the inception of the modern Asian monsoon dates to the construction of Asia, as it now exists, through the collision of the Indian and Asian blocks around 45 to 50 Myr. The elevation of the Tibetan Plateau/Himalayas also appears to be important, and early work on the effects of mountains on monsoons was carried out by Hahn and Manabe (1975). Prell and Kutzbach (1992) linked the modern elevation of the Tibetan Plateau to the strength of the monsoon. The date that Tibet reached its present height is not clear (and may be regionally variable). Around 8 Myr has been suggested, but estimates vary between 35 to less than 7 Myr and a high plateau may have existed before 8 Myr. There is evidence for stronger monsoons after 8 Myr from deep sea cores in the Arabian Sea (Kroon et al., 1991) and in Chinese loess sequences which date back to 7–8 Myr. Loess itself is a proxy for the winter monsoon, and the interbedded palaeosols for the summer monsoon. Loess–palaeosol sequences may date back to more than 7 Myr (An, 2000) (see Chapter 6, section 6.2), but with a significant increase in loess accumulation since about 2.7 Myr. Harris (2006) questions whether the shift around 8 Myr is actually due to the monsoon itself or wider oceanographic changes associated with Introduction increasing glaciations of Antarctica. Elsewhere, uplift in western North America is also seen as important; Tibet and the Rockies reach high enough elevations to disrupt the circulation in the upper atmosphere, affecting the mean location and amplitude of winter planetary waves and the location of the Siberian High. Harris (2006) suggests that monsoon intensification may actually date from the Miocene–Oligocene boundary (∼24–22 Myr) (based on data from the South China Sea), with the East Asian monsoon starting earlier than the Indian/Arabian monsoon. The former is more dependent on the evolution of the West Pacific Warm Pool and the latter on the uplift of the Himalayas/Tibet. Harris highlights the importance of tectonic influences on ocean currents, particularly the severance of Indonesian through flow and the closure of the Panama gateway helping to create the modern Pacific Ocean. The association between the development of the modern monsoon and the onset of the last glacial is a topic that has been widely debated. Ruddiman and Raymo (1988) suggest that uplift in Tibet and western North America played a role in the intensification of glaciations over the Pliocene that culminated in large scale glaciations about 2.4 Myr. The impact of uplift in these areas on CO2 drawdown (via the weathering effect) has also been the focus of considerable interest. Raymo and Ruddiman (1992) propose that uplift of the Tibetan plateau over the last 40 Myr and the associated increase in erosion, lowered global CO2, driving a positive feedback of global cooling. Mudelsee and Raymo (2005) provide a wider view of tectonic forcing and the development of Northern Hemisphere (NH) glaciations. The patterns of global climatic, tectonic and biotic events are summarised in Fig. 1.6. The theme of changes in the tropics driving environmental change is explored further below. At interannual timescales, ENSO is the dominant source of climatic variability in the tropics, where the Southern Oscillation Index (or SOI) is a measure of the strength of the Walker circulation (the east– west circulation across the Pacific which in ‘normal’ years gives high pressure and dry conditions in the east and low pressure and rain in the west). Major 15 weakening of the Walker circulation (low index conditions) results in the warm phase of the SOI, with warming of the eastern Pacific, weakening of the easterly trade winds and the ITCZ south of its usual position over South America. A deeper than normal thermocline develops on the east side of the Pacific, leading to a breakdown of the normal upwelling current and the area of high precipitation effectively moves east across the Pacific. The most obvious impacts occur along the west coast of tropical South America, with fisheries declining and major increases in precipitation (and erosion). The biggest impacts occur around Christmas, hence the name El Niño (the boy child). ENSO displays a cyclicity of 2–7 years and the strongest El Niño’s of twentieth century occurred in 1982–1983 and 1997–1998. Changes originating in the tropical Pacific have clear impacts on temperature and precipitation around the world, including mid latitudes (with seasonality) (Diaz and Markgraf, 2000). Significantly for tropical regions, El Niños (the warm phase of the SOI) are generally associated with weakened monsoons, and La Niñas (the cold phase of the SOI) with strong monsoons. As described above, it was Sir Gilbert Walker’s efforts (as Director General of Observatories, India Meteorological Department) to understand and forecast monsoons that led him to identify the Southern Oscillation, based on changes in the pressure gradient between Tahiti and Darwin. The global reach of the effects of El Niño have made understanding both its past and possible future a major research focus and the severe El Niño of 1982–1983 was a further stimulus to research. There have been a number of major syntheses of ENSO and its impacts including those of Diaz and Markgraf (1992, 2000) and Sarachik and Cane (2010). These draw on a range of sources including the (relatively short) instrumental record, historical records and proxy data. The first major publication based on historical records was that of Quinn et al. (1987), spanning 450 years, and focusing on Peru and southern Ecuador. Since then, there have been a number of syntheses of historical records including Ortlieb (2000) and Gergis and Fowler (2009). Records for El Niño per se extend back to the last interglacial, based on corals (Tudhope et al. 2001). 16 Chapter 1 r v Fig. 1.6 Summary of global environmental changes over the last 65 Myr. Redrawn from Zachos, J., Pagani, M., Sloan, L., et al. (2001) Trends, rhythms and aberrations in global climate 65Ma to present, Science 292, 686–93, copyright (2001), with permission from the American Association for the Advancement of Science. Other sources of palaeo-ENSO records include some lake and marine sediments, ice cores and tree rings (see Chapter 8, section 8.7 in particular). These provide evidence of changes in both frequency and intensity in ENSO (weaker in glacials, stronger in interglacials). There have evidently also been changes within the Holocene, with the ENSO being weaker during the early–mid Holocene and stronger after about 5 kyr BP, possibly with a major step up after 3 kyr BP. This variability in the Holocene has been linked to orbital changes (see also section 1.4), with early Holocene stronger summer insolation in the Northern Hemisphere (NH) resulting in stronger trade winds and mon- soons and a weaker ENSO cycle. One question is whether ENSO has ever stopped completely. Based on modelling, Clement et al. (2001) suggest that it did in the Younger Dryas (or YD; c. 13–11.5 kyr BP) and possibly before. In addition to ENSO itself there is increasing interest in persistent El Niño-type states, also called ENSO-type variability, over decades to millennia (see for example Chapter 7). Molnar and Cane (2002) suggest a virtually permanent El Niño like state in the Pliocene (5 to 2.7 Ma), with strong Hadley cell circulation driving more heat transport into mid and high latitudes. Onset of this state may have been driven by closure of the Indonesian Introduction seaway (see above). The authors suggest that this El Niño-type state might explain a lack of major NH ice sheets (winters warmer and drier). When this situation came to an end, major NH glaciations set in and cold upwelling zones became established off the west coasts of Africa and America. Fedorov et al. (2006) put this change into the context of gradual cooling over the Cenozoic which eventually reduced the depth of the ocean thermocline and allowed cold waters to come to the surface. Increasingly detailed marine records from both the normally cold eastern equatorial Pacific (Lawrence et al., 2006) and West Pacific Warm Pool (MedinaElizalde and Lea, 2010) have explored the Pliocene to Pleistocene transition and possible drivers, emphasising the importance of the tropics. Modelling of ENSO is discussed in some detail in Chapter 9, section 9.4.1, but a recent review by Sarachik and Cane (2010) sees this as still problematic, even with state of the art coupled models (e.g. IPCC, 2007). These shortcomings with current models cause problems for both ENSO forecasting (short term) and modelling changes in ENSO over the longer term (see Chapter 11, section 11.3.1). Simpler coupled models (e.g. Zebiak and Cane, 1987) are still widely used, for example to look at interactions between ENSO and other climatic forcings such as insolation change (see Chapter 9, section 9.5.3). 1.4 Drivers of tropical environmental change In the tropics, as elsewhere, the recognition of insolation as a driver of long term climatic change has been of profound importance. The variations in incoming radiation, which have come to be known as Milankovitch cycles, are referred to in many chapters, so a brief introduction to them and in particular their importance in tropical latitudes is given here. The theoretical framework for these cycles came from James Croll in the 1860s, but has come to be much more closely associated with Milutin Milankovitch (1941). The cycles of eccentricity, tilt and precession (100 kyr, 41 kyr and ∼23 kyr respectively) affect both the distribution 17 and/or amount of incoming solar radiation (insolation). The net effect of these cycles on insolation at 20 °N and °S is illustrated in Fig. 1.6. The changing amplitudes over the last 160 kyr and the clear evidence of the precession cycle are notable. Milankovitch proposed that changes in insolation due to the 41 kyr cycle and the 23 kyr cycle could drive NH ice growth (and hence explain ice ages). Work on marine core isotope records starting in the 1970s confirmed the presence of these periodicities (Hays et al., 1976). The ubiquity of the cycles in records of all kinds is now well known, but questions of how they are transmitted and amplified over different timescales remain (see Zachos et al., 2001 and Ruddiman, 2006, for overviews). In the context of this volume it is also notable that long cores from the tropical oceans played an important role in extending the marine timescale based on orbitally-driven insolation inputs, with the oxygen isotope (δ18O) signal (adjusted for lags in response to insolation changes), forming the basis for the SPECMAP timescale (Imbrie et al., 1984; Shackleton et al., 1990). The strongest direct effect on the tropics comes from precession (see Fig. 1.7), which alters the intensity of summer insolation and hence seasonality (although over a full year insolation anomalies sum to zero). Stronger seasonality in the tropics, associated with changes in monsoon strength, gives an antiphase signal between the NH and Southern Hemisphere (SH). The pattern of precession-driven changes in the position of the ITCZ and in monsoon strength is sometimes referred to as the ‘typical tropics’. Clement et al. (2004) suggest that the precession cycle is vital to understanding climatic change in the tropics, particularly the hydrological response. A dominance of precession (or half precession) cyclicity is reported from a number of tropical sites, including Lake Naivasha, Kenya (Trauth et al., 2003), Sanbao Cave, China (Wang, Y et al., 2008), Carnegie Ridge, Panama Basin (Pena et al., 2008) and Botuvera Cave, Brazil (Wang, X et al., 2007), although it should be noted that precession cycles are not clear in Australian records (see Chapter 7, section 7.8.2). Whether the coupling of monsoon strength with precession holds through time has 18 (a) Chapter 1 550 LGM LIG 500 W/m2 450 400 350 300 250 0 (b) 20 40 60 80 kyr BP 100 120 140 160 140 Fig. 1.7 Strength of insolation to the upper atmosphere (a) at 20 °N and (b) at 20 °S over the last 160 160 kyr, showing the 19–23 kyr influence of the precession cycle. Data from Analyseries (Berger, 1992; Paillard et al., 1996). 550 500 W/m2 450 400 350 300 250 0 20 40 60 80 kyr BP 100 Summer Winter been questioned by Fritz et al. (2004) based on results from the Salar de Uyuni (see Chapter 8, section 8.4.3). Precession is also recorded by the methane (CH4) in ice cores (e.g. Brook et al., 2000), mainly because of the importance of tropical wetlands as a source. Changes in methane emissions are seen as part of an early (quick) response to 120 insolation change. The role of the tropics in greenhouse gas emissions is discussed further below. In addition to the link to the monsoon, changes in precession are also associated with changes in ENSO. The possibility that strong precession-driven insolation during the early Holocene suppressed El Niño has been mentioned above (Nederbragt Introduction and Thurow, 2005; Peterson and Haug, 2006). A stronger monsoon was associated with a more La Niña-like state. Modelling (Clement et al., 1999) using the Zebiak–Cane model and precession forcing showed an ENSO-like response to precession (but not obliquity). These authors also suggest that modelled changes in ENSO would be a mechanism for generating a globally synchronous response to Milankovitch forcing (via changes in Hadley cell strength and heat and moisture transport to high latitudes). Although precession is an obvious focus for driving millennial scale change in the tropics, it is also linked to eccentricity and tilt. Precession is modulated by eccentricity; low eccentricity reduces the impact of precession on insolation. Overall, orbital cycles produce a variability in tropical insolation of around ±8% to the upper atmosphere, but at times of higher eccentricity (e.g. at the LGM), the precession signal would have been stronger than it is under low eccentricity conditions (modern). Tilt, unlike precession, does change annual insolation receipt at a given latitude and is important for monsoons due to its impact of warming over continental interiors. The importance of this is evident in the results of modelling experiments (e.g. Prell and Kutzbach, 1992; see Chapter 9, section 9.2.4 for further discussion). Some tropical (especially marine) records are actually dominated by the 41 kyr (tilt) and 100 kyr (eccentricity) signals (e.g. Lawrence et al., 2006; Medina-Elizalde and Lea, 2010 from either side of the Pacific), perhaps because the precession cycle itself is too short to affect the oceanic heat budget (Philander and Fedorov, 2003). The influence of orbital variability as a pacemaker of tropical climate change was subject to modification by other factors. Differences in cloudiness, the presence of large ice masses or changes in ocean circulation could easily have exerted strong local climate influences that disrupted the orbital signature. Alternatively these influences could have magnified the ecological effects of the change in insolation. A last point to make about these gradual changes in insolation is that they can cause abrupt changes in ecosystems. Modelled and 19 empirical data suggest that even with a gradual change in climate, a tipping point can be reached that results in very fast and strong ecological change (Scheffer et al., 2001, 2009; Bush et al., 2010). Another form of solar variability that is expressed in higher resolution tropical records is sunspot cycles. Although the exact mechanisms through which these small changes in solar output are propagated to drive global climate change are not clear (low solar activity seems to be associated with a more equatorward position of the ITCZ, weaker Hadley cell and monsoon circulations, i.e. drier in the tropics; van Loon et al., 2004; Versteegh, 2005), these solar cycles are being reported with increasing frequency. A cycle of around 200 years (the Suess Cycle) seems to be particularly common. Examples of solar variability records in tropical latitudes include the work of Hodell et al. (2001) and Stager et al. (2005). Even as the magnitude of Quaternary change in tropical latitudes became increasingly apparent, there was still a view that the tropics were essentially ‘passive’ in the global climate system, simply responding to changes driven by high latitudes and perhaps particularly in and around the North Atlantic. This view has only started to change fundamentally in the last 10–15 years as the importance of the tropical oceans and tropical controls on greenhouse gas concentrations have become increasingly evident (see Kerr, 2001). One of reasons behind the persistence of this ‘passive’ view may have been the highly influential CLIMAP SST reconstructions (CLIMAP, 1976, 1981). These suggested little change in tropical ocean SSTs (about 1 °C) and as these values were used to prescribe SSTs in early GCM experiments for the LGM they, unsurprisingly, showed little effect. In contrast tropical continental reconstructions (largely pollen, but also noble gases) showed major temperature depression at the LGM (5–8 °C). Guilderson et al. (1994) reconstructed SSTs based on Barbados corals and showed a temperature reduction of 5 °C at 19 kyr BP. Their data showed major and rapid changes in tropical SSTs and it was suggested that these, with changes in CO2 concentrations and in the thermohaline circulation, could explain the 20 Chapter 1 apparent synchroneity of global change in the period after the LGM. Reconstructing LGM tropical SSTs at the global scale has been problematic. One of the difficulties seems to be that there are no modern analogues for some LGM foraminiferal assemblages that form the basis for the temperature reconstructions. In a reanalysis of the CLIMAP data, Mix et al. (1999) addressed this problem and determined that the original analysis had underestimated SST change, by significant amounts in some places. Their new analysis suggested that the East Pacific cold tongue had actually cooled by 6 °C at the LGM. Concern about sources of error in faunal based tropical and subtropical SSTs leading to bias led to the development of bias-adjusted SST estimates and exploration of these on modelled climate (e.g. Hostetler et al., 2006). Other more recent studies of SST change include that of Kucera et al. (2005) based on a multiproxy approach and showing significant tropical ocean cooling, and that of Ballantyne et al. (2005) who undertook a Bayesian meta-analysis and identified tropical SST cooling of 2.7 ± 0.5 °C. See Chapter 3, sections 3.2.1 and 3.4.1 for further discussion of SST reconstructions. Given the known significance of ENSO, driven by tropical Pacific SSTs, in causing climate change at the global scale, it is perhaps surprising that recognition of the tropical oceans as drivers of change, over a range of timescales, has been slow to develop. Cane (1998) suggested that the tropical Pacific could drive both orbital and millennial scale climate variability, using ENSO teleconnections as an example. Long term ENSO type variability could favour interglacial/interstadials (El Niño/ warm phase) or glacial/stadials (cold phase). Even recent changes in the North Atlantic/European climate have been attributed to changes in tropical SSTs (specifically in the Indo-Pacific region) (Hoerling et al., 2001). Broecker (2003) considered the relative roles of the Atlantic thermohaline circulation and changes in the tropical atmosphere–ocean system (basically ENSO) in driving abrupt climate change, focusing on D–O events and the timing of warming, and Heinrich (H) Events. The Younger Dryas (sometimes called H0) was seen as a distinct event, almost certainly driven by meltwater dis- charge into the North Atlantic due to ice sheet dynamics, but Broecker recognised that warming prior to D–O events might suggest a tropical trigger. These early warming signals prior to H Events have been detected in sites far removed from the North Atlantic such as the Santa Barbara Basin off California (Hendy et al., 2002). 1.5 The tropics as drivers of change Although the focus above has been largely on factors that might drive climate change in the tropics, there are a number of climate forcings that originate in, or are significantly controlled by, the tropics. In simple energy terms, the tropical atmosphere and oceans are a major energy source, as they help to redistribute incoming solar radiation (see Chapters 2 and 3). In addition, the tropics are important in controlling a number of long lived radiatively active gases and aerosols, including water vapour. Here the role of the tropics in regulating the greenhouse gases carbon dioxide (CO2), methane (CH4) and nitrous oxide (N2O) (positive forcings), and the impact of tropical volcanoes in sulphate aerosol production and deserts as mineral dust sources (negative forcings), are summarised. 1.5.1 The tropics and greenhouse gas concentrations The tropics play a major role in terms of both sources and sinks of CO2. The major tropical CO2 sources are land use change and deforestation; over the last 20 years the IPCC (2007) estimate that the CO2 flux due to land use change (1.6 GtC yr-1) has been dominated by tropical deforestation. Although this may be balanced by uptake by tropical live biomass, it is not well quantified. Annual variability in CO2 fluxes, such as that evident in the Mauna Loa (Hawaii) record, is strongly influenced by tropical land areas. The importance of tropical areas (specifically the forests) as CO2 sinks and the potentially dramatic impacts of future warming (and drying) have been explored using GCMs (e.g. Cox et al., 2000). Friedlingstein et al. (2006) provide a more recent estimate of the possible feedbacks. The Introduction tropical oceans also play a part and are thought to outgas CO2 to atmosphere (estimated mean flux ∼0.7 GtC yr-1). Upwelling waters are particularly CO2 rich and the highest CO2 emissions seem to occur in the equatorial central and eastern Pacific. Both terrestrial and oceanic CO2 fluxes are affected by El Niño. On land, sources increase (drier, more forest fires), but oceanic sources decrease as upwelling in the eastern Pacific is reduced (see above). Biogenic sources dominate CH4 emissions (> 70%) and natural CH4 comes mainly from wetlands, both tropical and boreal, where emissions are associated with the occurrence of anaerobic conditions. According to Loulergue et al. (2008) the contribution is about one third boreal and two thirds tropical, and they suggest that in the pre-industrial period wetlands may have provided 85% of the total source. Other major biogenic sources include rice agriculture (paddies), biomass burning and ruminants; again these have significant contributions from tropical areas. δ13C can be used to fingerprint sources of CH4 and this has been used extensively in relation to ice core records (see below). Keppler et al. (2006) have suggested that it may also be possible for CH4 to be emitted by living vegetation (primarily from tropical forests and grasslands) under aerobic conditions. CH4 emissions are particularly sensitive to climate change via temperature and precipitation (moisture). Methane sinks are dominated by OH and the more water vapour there is, the more OH is produced. The abundance of OH is also affected by large volcanic eruptions (see below), which subsequently affects CH4 concentrations. Uncertainties in estimates of N2O sources are even larger than for CO2 and CH4, but tropical soils are important sources, probably contributing more than 50% of N2O emissions (Prinn et al., 1990; Hirsch et al., 2006). Other sources include agriculture (especially with fertiliser use) and the oceans (through denitrification). Hirsch et al. (2006) estimate that about 26% of N2O comes from the oceans. The records of greenhouse gas concentrations in ice cores have played a large part in our developing understanding of the association between atmos- 21 pheric composition and climate. Particularly in the case of CH4, it is clear that the tropics are a major methane source. The high resolution Greenland (GRIP) and Antarctic (Byrd) ice core CH4 records (Chappellaz et al., 1997) have been interpreted in terms of changes in source strengths in tropical and boreal wetlands. Up to about 5 kyr BP, the dominant source seems to have been the tropical wetlands, with particularly high values between 11.5 and 9.5 kyr BP. Drier conditions in the tropics between about 7 and 5 kyr BP were apparently reflected in lower overall CH4 concentrations. Dallenbach et al. (2000) extended the study into the last glacial (to 46 kyr BP) and demonstrated that mean CH4 levels in the GRIP core at the LGM were 362 ppbv compared with an average of 654 ppbv over the available Holocene record. These studies also explore interpolar differences, which were at a minimum at the LGM and a maximum during warm periods, especially 5–2.5 kyr BP. Dallenbach et al. (2000) attribute the larger portion of total CH4 during much of the last glacial to northern high latitudes, except at the LGM, but tropical sources predominate during cold periods (although totals are low). Wolff and Spahni (2007) revisited the ice core CH4 and N2O records and place more emphasis on changes in sinks which they see as necessary to explain the magnitude of observed changes in concentration. The authors also used isotopic measurements (δ13C) to help discriminate between sources; biogenic sources are isotopically light, whilst fossil fuels and biomass burning produces an isotopically heavier signal. Current CH4 levels exceed any recorded over the last 650 000 years. The Holocene story of CH4 is complex. Ruddiman (2003) suggested that human impacts on global CH4 could have begun as early as 5 kyr BP, largely due to activity in the tropics (rice paddies etc.), although his interpretation of the records has been disputed (Sowers, 2009; Singarayer et al., 2011). Interestingly, Singarayer et al. (2011) conclude that the main driver of the observed late Holocene increase in CH4 is emissions from the Southern Hemisphere tropics, driven by precession. N2O records are not as complete as CH4; they show a similar overall trend, but recent change is not as 22 Chapter 1 extreme. Some source discrimination is possible using δ15N (agricultural sources, especially fertilisers, are isotopically light). The link between tropical climate and methane and N2O emissions is illustrated by the correlation of CH4 with precession cycles (and monsoons); this correlation persists on millennial timescales. Ivanochko et al. (2005) suggest that changes in the tropical hydrological cycle during stadials/interstadials (D–O warm interstadials and cold stadials) could drive temperature changes in northern wetlands and hence changes in CH4 emissions. Changes in monsoon strength would also drive changes in N2O emissions from the oceans (via denitrification) and from terrestrial wetlands. δ15N values in marine sediments can serve as a proxy for denitrification and loss of N2 and N2O to the atmosphere. They see impacts of changes in these greenhouse gas emissions as a tropical mechanism for amplifying and perpetuating millennial scale climatic changes initiated in the North Atlantic (D–O cycles). Loulergue et al. (2008) describe an 820 kyr record from EPICA Dome C. They propose that tropical sources and sinks are the dominant control of the CH4 budget (linked to changes in monsoon strength), and that boreal wetlands are only significant during major terminations. Their CH4 record shows a dominant 100 kyr (eccentricity) cycle until 420 kyr BP, then a stronger precession signal. 1.5.2 Impacts of low latitude volcanic eruptions While the tropics clearly help to control warming greenhouse gas concentrations, eruptions of tropical volcanoes that emit large quantities of SO2 and SO4 (leading to the formation of sulphate aerosol) are most likely to have global and persistent cooling impacts on climate. Aerosols reduce radiative forcing (dimming) and are associated with general cooling, although warming may occur over NH continents in the first winter after an eruption. The lifetime of aerosols is quite short, so the effects based on observations only last about 2 years. Other impacts of these large eruptions (including those arising from emissions of water vapour, vegetation/albedo effects and other alterations to atmospheric chemistry), are more complicated. Robock (2000) provides a general overview of the effects of eruptions on atmospheric inputs and radiative forcing. It is clear that the eruption of large tropical volcanoes provides a possible mechanism to explain abrupt climatic change. The most recent example is Mt Pinatubo (Philippines) which erupted in 1991. This had a VEI (Volcanic Explosivity Index) of 6, and emitted about 20 Mt SO2 into the stratosphere (Robock, 2002). The duration of any climatic impacts arising from really large eruptions is still unclear. The super-eruption of Toba (northern Sumatra) about 73 kyr BP, with a VEI of 8, may have been the largest volcanic event of the last 2 Myr (certainly of the last 100 kyr), and its effects were seen globally (e.g. in GRIP). The duration of its impact is disputed, but Williams et al. (2009), based on pollen and isotopic data, suggest that it caused significant cooling and drying over 2 kyr and had a severe global impact upon the biosphere, hydrosphere and humans. The record of this massive eruption is discussed further in Chapter 6, section 6.3.3. D’Arrigo et al. (2009) have explored the impact of volcanic forcing on temperatures within the tropics over the last 400 years. They developed a zonally averaged annual tropical temperature record for the tropics between 30 °N and °S, using a range of proxies, and compared this to an index of volcanic forcing. They found that tropical temperatures were affected more by tropical eruptions than by those occurring at high latitudes (even large ones), although the magnitude of the response was smaller than at high latitudes. They noted that the most sustained cool period in their record occurred in the early nineteenth century (especially 1815–1818), which included the eruption of Tambora in 1815 (although they acknowledge that this was also a period of low solar irradiance). In contrast, Oman et al. (2006) found evidence of the impact of high latitude eruptions in the tropics, specifically on Nile River flow via changes in the monsoon following the eruption of Laki (Iceland) in 1783–1784. It is, however, interesting to note Introduction that although 1782–1783 was a strong El Niño period (see Ortlieb, 2000) this is not considered by Oman et al. In contrast, in their study, D’Arrigo et al. (2009) tried to filter out the impact of ENSO, but showed little effect on composite temperature (which masks spatial impacts). Another area of interest has been the possible ENSO type response to volcanic eruptions, originally suggested by Handler (1984) (see Adams et al., 2003). This was followed up by Emile-Geay et al. (2008) who modelled the impact of tropical eruptions on ENSO using the Zebiak–Cane model (see above), with a particular focus on a very large eruption of about 1258. They suggested that eruptions larger than that of Mt Pinatubo could increase the likelihood of an El Niño occurring (but not its intensity) whilst recognising that El Niños can be forecast using oceanic/atmospheric data alone. They proposed that the 1258 eruption may have caused a moderate/strong El Niño in the midst of a period dominated by La Niña-like conditions (the Mediaeval Climatic Anomaly). It is worth noting, however, that Robock (2000) does not support a causal link between eruptions and ENSO. One further aspect of volcanic eruptions is their effect on stratospheric ozone depletion. It is evident that recent large eruptions (such as Pinatubo) have impacted upon stratospheric O3 levels, since sulphate aerosol inputs to the stratosphere react with chlorine (largely anthropogenic in origin) to provide sites for catalytic reactions that lead to ozone depletion. Robock (2002) suggests that, as this process relies on the presence of chlorine, it would not have occurred prior to the mid twentieth century. 1.5.3 Dust emissions from the tropics and subtropics In recent years, the correlation between glacial periods and the flux of mineral dust to the atmosphere has attracted considerable interest. Model and data compilations (e.g. from marine sediments, polar and tropical ice caps, and continental loess deposits) indicate that much of the world experienced increased dust deposition 23 around the LGM, with global average mineral dust loading in the atmosphere around 50% higher than in preindustrial times (Kohfeld and Harrison, 2001). However, some areas, most notably the tropics and the poles, experienced higher loadings. For example, dust fluxes from Africa to the tropical and subtropical Atlantic during the LGM were 3–5 times higher than modern values (see Chapter 4, section 4.4), whilst fluxes into the North Pacific from the Americas and east Asia were 1–2 times higher (see Harrison et al., 2001). Modelling studies suggest that such elevated dust levels may have induced an average cooling of up to 0.72 °C in surface air temperature over the tropical oceans (Yue et al., 2011). Records from the Antarctic Vostok and Dome C ice cores show even more dramatic increases in dust deposition during glacials, with 10–12 times larger fluxes of dust at glacial maxima relative to the mean flux and 27–30 times larger fluxes of dust during the LGM compared to the present day (Petit et al., 1999; Delmonte et al., 2004). Analyses of mineralogical and isotopic tracers have been used to suggest that glacial dust in these cores is likely to have been transported from Patagonia (Grousset et al., 1992; Basile et al., 1997) with an Australian contribution during interglacials (e.g. Delmonte et al., 2004, 2008). The degree to which atmospheric dust loading is a response to, or a contributory cause of, climate changes on glacial–interglacial timescales is still uncertain (Harrison et al., 2001; Bar-Or et al., 2008). Glaciation, for example, has been suggested to increase dust flux into the atmosphere by (i) enhancing the effects of continentality (due to lower sea level), (ii) increasing the potential for soil erosion (through a reduction of vegetation cover caused by lower moisture availability) and (iii) increasing wind speeds (due to steeper pole– Equator pressure gradients) (e.g. Harrison et al., 2001; Harrison and Prentice, 2003). However, increased dust flux can also have both positive and negative feedbacks on glaciations through the aerosol direct radiative effect (see Yoon et al., 2005), the effect of dust deposition on snow and ice albedo, and the impact of aerosol particles on the reflection and absorption properties of clouds 24 Chapter 1 (Rosenfeld et al., 2006; Bar-Or et al., 2011). Dust transported to the oceans can also affect climate indirectly by modulating the supply of elements such as bioavailable iron, a micronutrient essential to photosynthesis in phytoplankton (Martin et al., 1991). Through this mechanism, variations in dust flux can influence the uptake of carbon in marine ecosystems and, in turn, the atmospheric concentration of CO2 (Maher et al., 2010). The relationship between changes in atmospheric dust loading and other palaeoenvironmental indicators from ice cores is not straightforward. For example, CO2 concentrations had already reached near-glacial levels by the time dust concentrations in the Vostok ice core began to increase around 65 kyr BP (Petit et al., 1999). In contrast, the decrease in dust loading evident in the Vostok core appears to be synchronous with, or even to precede, the increases in atmospheric CO2 concentrations during deglaciations (Harrison et al., 2001). 1.6 Extra-tropical forcing As described in a number of chapters in this volume, environmental changes affecting the North Atlantic (e.g. D–O cycles and H Events, the YD, changes in thermohaline circulation) remain the primary focus of research into the origins of millennial scale climate variability. Such changes are recognised well beyond the North Atlantic and are present in many tropical records of sufficient resolution (see examples in Chapters 4, 6, 7 and 8). Initially, interest centred upon the Younger Dryas (c. 13 kyr or 12.6–11.5 kyr BP). Evidence for the YD and earlier H Events (marking the end of D–O cycles) is widespread in the tropics and subtropics (e.g. Baker et al., 2001; Lea et al., 2003; Peterson and Haug, 2006; Wang X et al., 2006, 2007; Wang, Y et al., 2008). The near synchroneity of these events with the chronology of the Greenland ice cores has led to the suggestion that the tropics played an active role in propagating signals from the North Atlantic, probably via changes in the location of the ITCZ, monsoon strength and methane emissions (see above). The expression of these events (e.g. wetter/ drier) is, however, spatially variable. For example, at Botuvera in Brazil, the YD was wetter than present, but at Hulu in China it was drier (Wang, Y et al., 2001). This serves to highlight the complexity of response to climate forcings and the need to consider specific locations within the climate system. It is a timely reminder of the pitfalls of assuming that patterns are simply replicated, as was the case when the early evidence of lake level rise in the southwest USA during glacials was assumed to apply to all lower latitude locations (see section 1.2.1 above). Changes in the high latitudes of the Southern Hemisphere (primarily the Southern Ocean) have also been considered drivers of millennial scale climate variability, particularly in relation to warming during the early phases of deglaciation (terminations) caused by insolation forcing from the Southern Hemisphere and its effects on the global CO2 budget (e.g. Broecker and Henderson, 1998; Shulmeister et al., 2006). Many of the regional chapters (e.g. Chapters 4, 7 and 8) include examples of tropical records which appear to show changes more consistent with those over Antarctica than the North Atlantic. Although most of these are from the southern tropics, there are a number of NH sites which also seem to show a SH influence (e.g. Williams et al., 2010). 1.7 Organisation of the volume Quaternary Environmental Change in the Tropics is organised into three sections. Section A (‘Global contexts’) includes this introduction plus an overview of the contemporary climatology of the tropics (Chapter 2: Stefan Hastenrath). The latter chapter is designed to provide a background to the major features of tropical climate zones, with specific features of regional climate developed more fully in each of Chapters 3 to 8. Section B (‘Regional environmental change’) contains six substantive chapters. These review the evidence for environmental changes in the tropical oceans (Chapter 3: Jan-Berend W. Stuut, Matthias Prange, Ute Merkel and Silke Steph), Africa (Chapter 4: David Nash and Mike Meadows), India, Arabia and adjacent areas (Chapter 5: Ashok Introduction Singhvi, Nilesh Bhatt, Ken Glennie and Pradeep Srivastava), China and Southeast Asia (Chapter 6: Dan Penny), Australia and the southwest Pacific (Chapter 7: Peter Kershaw and Sander van der Kaars) and Latin America and the Caribbean (Chapter 8: Mark Bush and Sarah Metcalfe). The authors of each of the regional chapters were requested to address a series of specific issues within their reviews. First, they were asked to summarise the evidence for environmental change in their specific region, making reference to available sedimentological, geochemical (including isotopic), biological, geomorphological and archaeological evidence spanning the entire Quaternary. Second, they were asked to highlight any issues of spatial (e.g. longitudinal, Northern vs. Southern Hemisphere) and temporal (e.g. glacial vs. interglacial) variability within and between records for their region. Third, they were requested to consider the drivers of environmental changes, drawing attention to climatic versus human-induced forcing mechanisms where appropriate. We consider that all have more than adequately met this brief. The volume concludes with Section C (‘Global syntheses’) which contains three chapters designed to span the tropics and give a global perspective on key issues. Chapter 9 (Zhengyu Liu and Pascale Braconnot) reviews the contributions made by modelling studies to our understanding of tropical environments during the Quaternary. Chapter 10 (Georgina Endfield and Robert Marks) considers the evidence for environmental change in the tropics over the last 1000 years. Finally, Chapter 11 (David Nash and Sarah Metcalfe) draws together the evidence for Quaternary environmental changes in the tropics in a global synthesis, considers the impact of future climate change upon tropical regions and identifies a number of recommendations for areas of future research. Acknowledgements The authors would like to thank Elaine Watts (School of Geography, University of Nottingham) for producing the figures for this chapter. 25 References Adams, J., Mann, M. and Ammann, C. (2003) Proxy evidence for an El Niño like response to volcanic forcing. Nature 426, 274–278. Agassiz, L. (1840) Etudes sur les glaciers. Jent et Gassmann, Neuchâtel, Switzerland. Agassiz, L. (1868) A Journey in Brazil. Ticknor and Fields, Boston, USA. Aharon, P. and Chappell, J. (1986) Oxygen isotopes, sea level changes and temperature history of a coral reef environment in New Guinea over the last 105 years. Palaeogeography, Palaeoclimatology, Palaeoecology 56, 337–379. An, Z. (2000) The history and variability of the East Asian paleomonsoon climate. Quaternary Science Reviews 19, 171–187. Baillie, J.E.M., Hilton-Taylor, C. and Stuart, S.N. (2004) 2004 IUCN Red List of Threatened Species. A Global Species Assessment. IUCN, Gland, Switzerland. Baker, P.A., Rigsby, C.A., Seltzer, G.O. et al. (2001) Tropical climate changes at millennial and orbital timescales on the Bolivian Altiplano. Nature 409, 698–701. Ballantyne, A.P., Lavine, M., Crowley, T.J., et al. (2005) Meta-analysis of tropical surface temperatures during the last glacial maximum. Geophysical Research Letters 32, L05712. Bar-Or, R., Erlick, C. and Gildor, H. (2008) The role of dust in glacial-interglacial cycles. Quaternary Science Reviews 27, 201–208. Bar-Or, R., Gildor, H. and Erlick, C. (2011) The aerosolBenard cell effect on marine stratocumulus clouds and its contribution to glacial-interglacial cycles. Journal of Geophysical Research-Atmospheres 116, article number: D10119, doi: 10.1029/2010JD014470. Bard, E., Hamelin, B. and Fairbanks, R.G. (1990) U-Th ages obtained by mass spectrometry in corals from Barbados: sea level during the past 130,000 years. Nature 346, 456–458. Basile, I., Grousset, F.E., Revel, M., et al. (1997) Patagonian origin of glacial dust deposited in East Antarctica (Vostok and Dome C) during glacial stages 2, 4 and 6. Earth and Planetary Science Letters 146, 573–589. Beerling, D.J. and Woodward, F.I. (1993) Ecophysiological responses of plants to global environmental change since the Last Glacial Maximum. New Phytologist 125, 641–648. Berger, A. (1992) Astronomical theory of paleoclimates and the last glacial-interglacial cycle. Quaternary Science Reviews 11, 571–581. 26 Chapter 1 Betancourt, J.L., Van Devender, T.R. and Martin, P.S. (1990) Packrat Middens. The Last 40,000 Years of Biotic Change. The University of Arizona Press, Tucson. Birks, H.J. (2005) Quantitative palaeoenvironmental reconstructions from Holocene biological data. In: Mackay, A., Battarbee, R., Birks, J. and Oldfield, F. (eds) Global Change in the Holocene. Hodder Arnold, London, pp. 107–123. Birks, H.J.B. and Birks, H.H. (1980) Quaternary Palaeoecology. Edward Arnold, London. Braconnot, P., Otto-Bliesner, B., Harrison, S., et al. (2007) PMIP2 coupled simulations of the mid-Holocene and Last Glacial Maximum – part 1: experiments and large scale features. Climate of the Past 3, 261–277. Bridgman, H.A. and Oliver, J.E. (2006) The Global Climate System, Patterns, Processes and Teleconnections. Cambridge University Press, Cambridge. Broecker, W.S. (2003) Does the trigger for abrupt climate change reside in the ocean or in the atmosphere? Science 300, 1519–1522. Broecker, W.S. and Henderson, G.M. (1998) The sequence of events surrounding Termination II and their implications for the cause of glacial-interglacial CO2 changes. Paleoceanography 13, 352–364. Brook, E.J., Harder, S., Severinghaus, J., et al. (2000) On the origin and timing of rapid changes in atmospheric methane during the last glacial period. Global Biogeochemical Cycles 14, 559–572. Bush, M.B., Hanselman, J.A. and Gosling, W.D. (2010) Nonlinear climate change and Andean feedbacks: an imminent turning point? Global Change Biology 16, 3223–3232. Candy I., Black S. and Sellwood B.W. (2004) Quantifying timescales of pedogenic calcrete formation using Useries disequilibria. Sedimentary Geology 170, 177–187. Cane, M.A. (1998) A role for the tropical Pacific. Science 282, 59–62. Cane, M.A., Braconnot, P., Clement, A., et al. (2006) Progress in paleoclimate modeling. Journal of Climate 19, 5031–5057. Chappellaz, J., Barnola, J.-M., Raynaud, D., et al. (1990) Atmospheric CH4 record over the last climatic cycle revealed by the Vostok ice core. Nature 345, 127–131. Chappellaz, J., Blunier, T., Kints, S., et al. (1997) Changes in the atmospheric CH4 gradient between Greenland and Antarctica during the Holocene. Journal of Geophysical Research 102, 15987–15997. Chase, B.M., Meadows, M.E., Scott, L., et al. (2009) A record of rapid Holocene climate change preserved in hyrax middens from southwestern Africa. Geology 37, 703–706. Clement, A.C., Seager, R. and Cane, M.A. (1999) Orbital controls on the El Nino/Southern Oscillation and the tropical climate. Paleoceanography 14, 441–456. Clement, A.C., Cane, M.A. and Seager, R. (2001) An orbitally driven tropical source for abrupt climate change. Journal of Climate 14, 2369–2375. Clement, A.C., Hall, A. and Broccoli, A.J. (2004) The importance of precessional signals in the tropical climate. Climate Dynamics 22, 327–341. Clift, P.D. and Plumb, R.A. (2008) The Asian Monsoon. Causes, History and Effects. Cambridge University Press, Cambridge. CLIMAP Project members (1976) The surface of ice-age earth. Science 191, 1131–1137. CLIMAP Project members (1981) Seasonal reconstructions of the earth’s surface at the last glacial maximum. Geological Society of America Map and Chart Series MC-36. COHMAP Members (1988) Climatic changes of the last 18,000 years. Observations and model simulations. Science 241, 1043–1052. Cox, P.M., Betts, R.A., Jones, C.D., et al. (2000) Acceleration of global warming due to carbon cycle feedbacks in a coupled climate model. Nature 408, doi:10.1038/ 35041539. Crowley, T.J., Kim, K-Y., Mengel, J.G. and Short, D.A. (1992) Modeling 100,000 year climate fluctuations in pre-Pleistocene time series. Science 225, 705–707. D’Arrigo, R., Wilson, R. and Tudhope, A. (2009) The impact of volcanic forcing on tropical temperatures during the past four centuries. Nature Geoscience 2, 51–56. Dallenbach, A., Blunier, T., Fluckiger, J., et al. (2000) Changes in the atmospheric CH4 gradient between Greenland and Antarctica during the Last Glacial and transition to the Holocene. Geophysical Research Letters 27, 1005–1008. Dallmeyer, A., Claussen, M. and Otto, J. (2010) Contribution of oceanic and vegetation feedbacks to Holocene climate change in monsoonal Asia. Climate of the Past 6, 195–218. Damuth, J.E. and Fairbridge, R.W. (1970) Arkosic sands of the last glacial stage in the tropical Atlantic off Brazil. Geological Society of America, Bulletin 81, 189–206. de Beer, G. R. (1953) Bernard Friedrich Kuhn’s investigations on glaciers. Annals of Science 9, 323–341. de Menocal, P.B. (1995) Plio-Pleistocene African climate. Science 270, 53–59. de Menocal, P.B. (2004) African climate change and faunal evolution during the Pliocene-Pleistocene. Earth and Planetary Science Letters 220, 3–24. Introduction de Saussure, H.-B. (1779) Voyages dans les Alpes, précèdes d’un essai sur l’histoire naturelle des environs de Genève, Vol. 1. Barde, Manget and Compagnie, Geneva, Switzerland. Delmonte, B., Basile-Doelsch, I., Petit, J.-P., et al. (2004) Comparing the EPICA and Vostok dust records during the last 220,000 years: stratigraphical correlation and provenance in glacial periods. Earth-Science Reviews 66, 63–87. Delmonte, B., Andersson, P.S., Hansson, M., et al. (2008) Aeolian dust in East Antarctica (EPICA-Dome C and Vostok): provenance during glacial ages over the last 800 kyr. Geophysical Research Letters 35, L07703. doi:10.1029/2008GL033382. Diaz, H. and Markgraf, V. (eds) (1992) El Niño, Historical and Paleoclimatic Aspects of the Southern Oscillation. Cambridge University Press, Cambridge. Diaz, H. and Markgraf, V. (eds) (2000) El Niño and the Southern Oscillation, Multiscale Variability and Global and Regional Impacts. Cambridge University Press, Cambridge. Diester-Haas, L. (1976) Late Quaternary variations in northwest Africa deduced from east Atlantic sediment cores. Quaternary Research 6, 299–314. Drake, N.A., Blench, R.M., Armitage, S.J., et al. (2011) Ancient watercourses and biogeography of the Sahara explain the peopling of the desert. Proceedings of the National Academy of Science 108, 458–462. Emile-Geay, J., Seager, R., Cane, M.A., et al. (2008) Volcanoes and ENSO over the past millennium. Journal of Climate 21, 3134–3148. Emiliani, C. (1955) Pleistocene temperatures. Journal of Geology 63, 538–575. Fairbanks, R.G. (1989) A 17,000-year glacio-eustatic sea level record: influence of glacial melting rates on the Younger Dryas event and deep-ocean circulation. Nature 342, 637–642. Fedorov, A.V., Dekens, P.S., McCarthy, M., et al. (2006) The Pliocene paradox (mechanisms for a permanent El Niño). Science 312, 1485–1489. Flenley, J.R. (1979) The Equatorial Rain Forest: A Geological History. Butterworths, London. Friedlingstein, P., Cox, P., Betts, R., et al. (2006) Climatecarbon cycle feedback analysis: results from the C4MIP model intercomparison. Journal of Climate 19, 3337–3353. Fritz, S.C., Cumming, B.F., Gasse, F. and Laird, K.R. (1999) Diatoms as indicators of hydrologic and climatic change in saline lakes. In: Stoermer, E.F. and Smol, J.P. (eds) The Diatoms: Applications for the Environmental and Earth Sciences. Cambridge University Press, Cambridge, pp. 41–72. 27 Fritz, S.C., Baker, P.A., Lowenstein, T.K., et al. (2004) Hydrologic variation during the last 170,000 years in the southern hemisphere tropics of South America. Quaternary Research 61, 95–104. Fujioka, T., Chappell, J., Honda, M., et al. (2005) Global cooling initiated stony deserts in central Australia 2–4 Ma, dated by cosmogenic 21Ne-10Be. Geology 33, 993–996. Garnier, B.J. (1958) Some comments on defining the humid tropics. Research Note 11, Department of Geography, University of Ibadan. Gates, W.L. (1976a) Modelling the ice-age climate. Science 191, 1138–1144. Gates, W.L. (1976b) The numerical simulation of ice-age climate with a global general circulation model. Journal of Atmospheric Sciences 33, 1844–1873. Gergis, J.L. and Fowler, A.M. (2009) A history of ENSO events since A.D. 1525: implications for future climate change. Climatic Change 92, 343–387. Gibbard, P., Head, M.J. and Walker, M.J.C. (2010) Formal ratification of the Quaternary System/Period and the Pleistocene Series/Epoch with a base at 2.58 Ma. Journal of Quaternary Science 25, 96–102. Gilbert, G.K. (1890) Lake Bonneville. US Geological Survey Monograph 1. USGS. Goudie, A.S. (1972) The concept of post-glacial progressive desiccation. Research Papers, School of Geography, University of Oxford, pp. 4–48. Goudie, A.S. (1999) The Ice Age in the tropics and its human implications. In: Slack, P.A. (ed.) Environments and Historical Change. Oxford University Press, Oxford, pp. 10–32. Goudie, A.S., Allchin, B. and Hegde, K.T.M. (1973) The former extensions of the Great Indian Sand Desert. Geographical Journal 139, 243–257. Grab, S.W. and Nash, D.J. (2010) Documentary evidence of climate variability during cold seasons in Lesotho, southern Africa, 1833–1900. Climate Dynamics 34, 473–499. Groot, M.H.M., Bogota, R.G., Lourens, L.J., et al. (2011). Ultra-high resolution pollen record from the northern Andes reveals rapid shifts in montane climates within the last two glacial cycles. Climate of the Past 7, 299–316. Grousset, F.S., Biscaye, P.E., Revel, M., et al. (1992) Antarctic (Dome C) ice-core dust at 18 k.y. B.P.: isotopic constraints on origins. Earth and Planetary Science Letters 111, 175–182. Grove, A.T. (1958) The ancient erg of Hausaland, and similar formations on the south side of the Sahara. Geographical Journal 124, 528–533. 28 Chapter 1 Grove, A.T. (1969) Landforms and climate change in the Kalahari and Ngamiland. Geographical Journal 135, 192–212. Grove, A.T. and Goudie, A.S. (1971) Late Quaternary lake levels in the Rift Valley of southern Ethiopia and elsewhere in tropical Africa. Nature 234, 403–405. Grove, A.T. and Warren, A. (1968) Quaternary landforms and climate on the south side of the Sahara. Geographical Journal 134, 194–208. Grove, A.T., Street, F.A. and Goudie, A.S. (1975) Former lake levels and climatic change in the rift valley of southern Ethiopia. Geographical Journal 141, 177–194. Guilderson, T.P., Fairbanks, R.G. and Rubenstone, J.L. (1994) Tropical temperature variations since 20,000 years ago: modulating interhemispheric climate change. Science 263, 663–665. Hahn, D.G. and Manabe, S. (1975) The role of mountains in the south Asian monsoon circulation. Journal of Atmospheric Science 32, 151–154. Handler, P. (1984) Possible association of stratospheric aerosols and El Nino type events. Geophysical Research Letters 11, 1121–1124. Harris, N.B.W. (2006) The elevation history of the Tibetan Plateau and its implications for the Asian monsoon. Palaeogeography, Palaeoclimatology, Palaeoecology 241, 4–15. Harrison, S.P. and Prentice, I.C. (2003) Climate and CO2 controls on global vegetation distribution at the last glacial maximum: analysis based on palaeovegetation data, biome modelling and palaeoclimate simulations. Global Change Biology 9, 983–1004. Harrison, S.P., Kohfeld, K.E., Roelandt, C. and Claquin, T. (2001) The role of dust in climate changes today, at the last glacial maximum and in the future. Earth Science Reviews 54, 43–80. Hays, J.D., Imbrie, J. and Shackleton, N.J. (1976) Variations in the earth’s orbit: pacemaker of the ice ages. Science 194, 1121–1132. Hendy, I.L., Kennett, J.P., Roark, E.B. and Ingram, B.L. (2002) Apparent synchroneity of submillennial scale climate events between Greenland and Santa Barbara basin, California from 30 – 10 ka. Quaternary Science Reviews 21, 1167–1184. Hirsch, A.I., Michalak, A.M., Bruhwiler, L.M., et al. (2006) Inverse modeling estimates of the global nitrous oxide surface flux from 1998–2001. Global Biogeochemical Cycles 20, GB1008, doi:10.1029/2004GB 002443. Hodell, D.A., Brenner, M, Curtis, J.H. and Guilderson, T. (2001) Solar forcing of drought in the Maya lowlands. Science 292, 1367–1370. Hoerling, M.P., Hurrell, J.W. and Xu, T. (2001) Tropical origins for recent North Atlantic climate change. Science 292, 90–92. Holmgren, C.A., Norris, J. and Betancourt, J.L. (2007) Inferences about winter temperatures and summer rains from the late Quaternary record of C-4 perennial grasses and C-3 desert shrubs in the northern Chihuahuan Desert. Journal of Quaternary Science 22, 141–161. Holmes, J.A. and Engstrom, D.R. (2005) Non-marine ostracod records of Holocene environmental change. In: Mackay, A., Battarbee, R., Birks, J. and Oldfield, F. (eds) Global Change in the Holocene. Hodder Arnold, London, pp. 310–327. Hostetler, S., Giorgi, F., Bates, G.T. and Bartlein, P.J. (1994) Lake-atmosphere feedbacks associated with paleolakes Bonneville and Lahontan. Nature 263, 665–668. Hostetler, S., Pisias, N. and Mix, A. (2006) Sensitivity of Last Glacial Maximum climate to uncertainties in tropical and subtropical ocean temperatures. Quaternary Science Reviews 25, 1168–1185. Hubbell, S.P., He, F., Condit, R., Borda-de-Agua, L., et al. (2008) How many tree species are there in the Amazon and how many of them will go extinct? Proceedings of the National Academy of Sciences 105, 11498–11504. Hutton, J. (1795) Theory of the Earth; With Proofs and Illustrations. Creech, Edinburgh. Imbrie, J. and Imbrie, K.P. (1979) Ice Ages. Solving the Mystery. Macmillan, London. Imbrie, J., Hays, J.D., Martinson, D.G., et al. (1984) The orbital theory of Pleistocene climate: support from a revised chronology of the marine δ18O record. In: Berger A.L., Imbrie, J., Hays, J., et al. (eds) Milankovitch and Climate Part 1. D. Reidel, Dordrecht, pp. 269–305. IPCC (2001) Climate Change 2001: The Physical Science Basis. Contribution of Working Group I to the Third Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge. IPCC (2007) Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge. Ivanochko, T.S., Ganeshram, R.S., Brummer, G.-J., et al. (2005) Variations in tropical convection as an amplifier of global climate change at the millennial scale. Earth and Planetary Science Letters 235, 302–314. Jackson, I.J. (1989) Climate, Water and Agriculture in the Tropics. Longman, Harlow. Jamieson, T.F. (1863) On the parallel roads of Glen Roy, and their place in the history of the glacial period. Introduction Quarterly Journal of the Geological Society of London 19, 235–259. Keppler, F., Hamilton, J.T., Braß M. and Rockmann, T. (2006) Methane emissions from terrestrial plants under aerobic conditions. Nature 439, 187–191. Kerr, R.A. (2001) The tropics return to the climate system. Science 292, 660–661. Kohfeld, K.E. and Harrison, S.P. (2001) DIRTMAP: the geological record of dust. Earth-Science Reviews 54, 81–114. Köppen, W. (1936) Das geographische System de Klimate. In: Handbuch der Klimatologie, Vol. 1. Borntager, Berlin. Kottek, M., Grieser, J., Beck, C., et al. (2006) World Map of the Köppen-Geiger climate classification updated. Meteorologische Zeitschrift 15, 259–263. Kroon, D., Steens, T. and Troelstra, S.R. (1991) Onset of monsoonal related upwelling in the western Arabian Sea as revealed by planktonic foraminifers. Proceedings of the Ocean Drilling Program, Scientific results, vol. 117. College Station Texas, ODP, pp. 257–263. Kucera, M., Rosell-Melé, A., Schneider, R., et al. (2005) Multiproxy approach for the reconstruction of glacial ocean surface (MARGO). Quaternary Science Reviews 24, 814–819. Kuhn, B.F. (1787) Versuch über den Mechanismus der Gletscher. Magazin für die Naturkunde Helvetiens 1, 117–136. Kutzbach, J.E. and Guetter, P.J. (1986) The influence of changing orbital parameters and surface boundary conditions on climate simulations for the past 18,000 years. Journal of Atmospheric Sciences 43, 1726–1759. Larrasoaña, J.C., Roberts, A.P., Rohling, E.J., et al. (2003) Three million years of monsoon variability over the northern Sahara. Climate Dynamics 21, 689–698. Lawrence, K.T., Liu, Z. and Herbert, T.D. (2006) Evolution of the eastern tropical Pacific through Plio-Pleistocene glaciation. Science 312, 79–83. Lea, D.W., Pak, D.K., Peterson, J.C. and Hughen, K.A. (2003) Synchroneity of tropical and high-latitude Atlantic temperatures over the last glacial termination. Science 301, 1361–1364. Leng, M.J. (ed.) (2006) Isotopes in Palaeoenvironmental Research. Developments in Paleoenvironmental Research, Volume 10. Springer, Dordrecht. Livingstone, D. (1857) Missionary Travels and Researches in South Africa. Ward, Lock and Co., New York. Loulergue, L., Schilt, A., Spahni, R., et al. (2008) Orbital and millennial-scale features of atmospheric methane over the past 800,000 years. Nature 354, 383–386. 29 Lourandos, H. (1997) Continent of Hunter-Gatherers: New Perspectives in Australian Prehistory. Cambridge University Press, Cambridge. Lowe, J.J. and Walker, M.J.C. (1997) Reconstructing Quaternary environments (2nd edition). Longman, Harlow. Mace, G., Masundire, H., Baillie, J. coordinating lead authors (2005) Biodiversity. In: Hassan, R., Scholes, R. and Ash, N. (eds) Ecosystems and Human Well-being: Current State and Trends, volume 1. Island Press, Washington, pp. 77–122. Maher, B.A., Prospero, J.M., Mackie, D., et al. (2010) Global connections between aeolian dust, climate and ocean biogeochemistry at the present day and at the last glacial maximum. Earth-Science Reviews 99, 61–97. Manabe, S. and Hahn, D.G. 1977. Simulation of the tropical climate of an ice age. Journal of Geophysical Research 82, 3889–3911. Martin, J.H., Gordon, R.M. and Fitzwater, S.E (1991) The case for iron. Limnology and Oceanography 36, 1793–1802. McGregor, G.R. and Nieuwolt, S. (1998) Tropical Climatology. John Wiley & Sons, Chichester. McGuffie, K. and Henderson Sellers, A. (2001) Forty years of numerical climate modelling. International Journal of Climatology 21, 1067–1109. Medina-Elizalde, M. and Lea, D.W. (2010) Late Pliocene equatorial Pacific. Paleoceanography 25, PA2208, doi: 10.1029/2009PA001780. Meigs, P. (1953) World distribution of arid and semi-arid homoclimates. In: Arid Zone Hydrology. UNESCO Arid Zone Research Series 1, 203–209. Middleton, N.J., Thomas, D.S.G. and UNEP (eds) (1997) World Atlas of Desertification (2nd edition). UNEP. Milankovitch, M.M. (1941) Canon of Insolation and the Ice Age Problem. Koniglich Serbische Akademie, Belgrade. Mix, A., Morey, A.E., Pisias, N.G. and Hostetler, S.W. (1999) Foraminiferal faunal estimates of paleotemperature: circumventing the no-analog problem yields cool ice-age tropics Paleoceanography 14, 350–359. Molnar, P and Cane, M.A. (2002) El Niño’s tropical climate and teleconnections as a blueprint for pre-Ice Age climates. Paleoceanography 17, 1021, doi: 10. 1029/2001PA000663. Mudelsee, M. and Raymo, M.E. (2005) Slow dynamics of the Northern Hemisphere glaciations. Paleoceanography 20, PA4022, doi:10.1029/2005PA001153. Munyikwa, K. (2005) Synchrony of Southern Hemisphere Late Pleistocene arid episodes: a review of luminescence chronologies from arid aeolian landscapes 30 Chapter 1 south of the Equator. Quaternary Science Reviews 24, 2555–2583. Nash D.J. and Endfield, G.H. (2002) A nineteenth century climate chronology for the Kalahari region of central southern Africa derived from missionary correspondence. International Journal of Climatology 22, 821–841. Nash, D.J. and Endfield, G.H. (2008) ‘Splendid rains have fallen’: links between El Niño and rainfall variability in the Kalahari, 1840–1900. Climatic Change 86, 257–290. Nash, D.J. and Grab, S.W. (2010) ‘A sky of brass and burning winds’: documentary evidence of rainfall variability in Lesotho, 1824–1900. Climatic Change 101, 617–653. Nederbragt, A.J. and Thurow, J. (2005) Amplitude of ENSO cycles in the Santa Barbara Basin, off California, during the past 15,000 years. Journal of Quaternary Science 20, 447–456. Newberry, J.S. (1861) Report upon the Colorado River of the West, Part III. Geological report, Washington DC. Nicholson, S.E. (2000) The nature of rainfall variability over Africa on time scales of decades to millennia. Global and Planetary Change 26, 137–158. Nicholson, S.E. (2001) Climatic and environmental change in Africa during the last two centuries. Climate Research 17, 123–144. Nott, J., Thomas, M.F. and Price, D.M. (2001) Alluvial fans, landslides and Late Quaternary climatic change in the wet tropics of northeast Queensland. Australian Journal of Earth Sciences 48, 875–882. O’Connell, J.F. and Allen, F.J. (2005) Dating the colonization of Sahul (Pleistocene Australia-New Guinea): a review of recent research. Journal of Archaeological Science 31, 835–853. Oman, L., Robock, A., Stenchikov, G.L. and Thordarson, T. (2006) High-latitude eruptions cast shadow over the African monsoon and the flow of the Nile. Geophysical Research Letters 33, L18711, doi:10.1029/2006GL 027665. Ortlieb, L. (2000) The documentary historical records of El Niño events in Peru: an update of the Quinn record (sixteenth through nineteenth centuries). In: Diaz, H. and Markgraf, V. (eds) El Niño and the Southern Oscillation: Multiscale Variability and Global and Regional Impacts. Cambridge University Press, Cambridge, pp. 207–295. Paillard, D., Labeyrie, L. and Yiou, P. (1996) Macintosh program performs time-series analysis. Eos Transactions AGU 77, 379. Parker, A.G., Eckersley, L., Smith, M.M., et al. (2004) Holocene vegetation dynamics in the northeastern Rub’ al-Khali desert, Arabian Peninsula: a phytolith, pollen and carbon isotope study. Journal of Quaternary Science 19, 665–676. Patterson, L. J., Sturchio, N.C., Kennedy, B.M., et al. (2005) Cosmogenic, radiogenic, and stable isotopic constraints on groundwater residence time in the Nubian Aquifer, Western Desert of Egypt, Geochemistry Geophysics Geosystems 6, Q01005, doi:10.1029/ 2004GC000779. Pearson, S. and Dodson, J.R. (1993) Stick-nestrat middens as sources of palaeoecological data in Australian deserts. Quaternary Research 39, 347–354. Peltier, W.R. and Fairbanks, R.G. (2008) Barbados Coral Extended Last Glacial Maximum Sea Level Record. IGBP PAGES/World Data Center for Paleoclimatology Data Contribution Series # 2008-101. NOAA/NCDC Paleoclimatology Program, Boulder CO, USA. Pena, L.D., Cacho, I., Ferretti, P. and Hall, M.A. (2008) El Nino-Southern Oscillation-like variability during glacial terminations and interlatitudinal teleconnections. Paleoceanography 23, PA3101, doi:10.1029/2008PA001620. Peterson, L.C. and Haug, G.H. (2006) Variability in the mean latitude of the Atlantic Intertropical Convergence Zone as recorded by riverine inputs of sediments to the Cariaco Basin (Venezuela). Palaeogeography, Palaeoclimatology, Palaeoecology 234, 97–113. Petit, J.R., Jouzel, J., Raynaud, D., et al. (1999) Climate and atmospheric history of the past 420,000 years from the Vostok Ice Core, Antarctica. Nature 399, 429–436. Philander, S.G. and Fedorov, A.V. (2003) Role of tropics in changing the response to Milankovitch forcing some three million years ago. Paleoceanography 18, 1045, doi:10.1029/2002PA000837. Playfair, J. (1802) Illustrations of the Huttonian Theory of the Earth. Creech, Edinburgh. Prell, W.L. and Kutzbach, J.E. (1987) Monsoon variability over the past 150,000 years. Journal of Geophysical Research 92, D7, 8411–8425. Prell, W.L. and Kutzbach, J.E. (1992) Sensitivity of the Indian monsoon to forcing parameters and implications for its evolution. Nature 360, 647–652. Price, W.A. (1944) Greater American deserts. Texas Academy of Science Proceedings and Transactions 27, 163–170. Prinn, R.D., Cunnold, D., Rasmussen, R., et al. (1990) Atmospheric emissions and trends of nitrous oxide deduced from 10 years of Ale-gauge data. Journal of Geophysical Research-Atmospheres 95, 18369–18385. Quinn, W.H., Neal, V.T., and Antunez e Mayolo, S. (1987) El Niño occurrences over the past four and half Introduction centuries. Journal of Geophysical Research 93, 14449– 14461. Raymo, M.E. and Ruddiman, W.F. (1992) Tectonic forcing of late Cenozoic climate. Nature 359, 117–122. Raynaud, D., Chappellaz, J., Barnola, J.M., et al. (1988) Climatic and CH4 cycle implications of glacial-interglacial CH4 change in the Vostok ice core. Nature 333, 655–657. Reading, A.J., Thompson, R.D. and Millington, A.C. (1995) Humid Tropical Environments. Blackwell, Oxford. Richards, P.W. (1952) The Tropical Rainforest. An Ecological Study. Cambridge University Press, Cambridge. Richey, J.E, Melack, J.M., Aufdenkampe, A.K., et al. (2002) Outgassing from Amazonian rivers and wetlands as a large tropical source of atmospheric CO2. Nature 416, 617–620. Roberts, R.G., Jones, R. and Smith, M.A. (1990) Thermoluminescence dating of a 50,000 year old human occupation site in northern Australia. Nature 345, 153–156. Roberts, R.G., Jones, R. and Smith, M.A. (1993) Optical dating at Deaf Adder Gorge, Northern Territory, indicates human occupation between 53,000 and 60,000 years ago. Australian Archaeology 37, 58–59. Roberts, R.G., Jones, R., Spooner, N.A., et al. (1994) The human colonisation of Australia: optical dates of 53,000 and 60,000 years bracket human arrival at Deaf Adder Gorge, Northern Territory. Quaternary Science Reviews 13, 573–583. Robock, A. (2000) Volcanic eruptions and climate. Reviews of Geophysics 38, 191–219. Robock, A. (2002) The climatic aftermath. Science 295, 1242–1244. Rosenfeld, D., Kaufman, Y. and Koren, I. (2006) Switching cloud cover and dynamical regimes from open to closed Benard cells in response to aerosols suppressing precipitation. Atmospheric Chemistry and Physics 6, 2503–2511. Ruddiman, W.F. (2003) The anthropogenic greenhouse era began thousands of years ago. Climatic Change 61, 261–293. Ruddimann, W.F. (2006) Orbital changes and climate. Quaternary Science Reviews 25, 3092–3112. Ruddiman, W.F. and Kutzbach, J.E. (1989) Forcing of late Cenozoic northern hemisphere climate by plateau uplift in southern Asia and the American west. Journal of Geophysical Research 94, D15, 18409–18427. Ruddiman, W.F. and Raymo, M.E. (1988) Northern hemisphere climate regimes during the past 3 Ma: possible tectonic connections. Philosophical Transactions of the Royal Society of London B318, 411–430. 31 Russell, I.C. (1885) Geological History of Lake Lahontan: a Quaternary lake of northwestern Nevada. US Geological Survey Monograph 11. USGS. Sarachik, E.S. and Cane, M.A. (2010) The El Niño-Southern Oscillation Phenomenon. Cambridge University Press, Cambridge. Savage, J.M., Goldman, D.P., Janos, D.P., et al. (1982) Ecological Aspects of Development in the Humid Tropics. National Academy Press, Washington DC. Scheffer, M., Carpenter, S., Foley, J.A., et al. (2001) Catastrophic shifts in ecosystems. Nature 413, 591– 596. Scheffer, M., Bascompte, J., Brock, W. A., et al. (2009) Early-warning signals for critical transitions. Nature 461, 53–59. Schilt, A., Baumgartner, M., Blunier, T., et al. (2010) Glacial–interglacial and millennial-scale variations in the atmospheric nitrous oxide concentration during the last 800,000 years. Quaternary Science Reviews 29, 182–192. Schmittner, A. and Galbraith, E.D. (2008) Glacial greenhouse-gas fluctuations controlled by ocean circulation changes. Nature 456, 373–376. Schwarz, E.H.L. (1923) The Kalahari or Thirstland Redemption. Oxford University Press, Cape Town. Shackleton, N.J., Berger, A. and Peltier, W.R. (1990) An alternative astronomical calilbration of the lower Pleistocene timescale based on ODP site 677. Transactions of the Royal Society of Edinburgh, Earth Science 81, 251–261. Shulmeister, J., Rodbell, D.T., Gagan, M.K. and Seltzer, G.O. (2006) Inter-hemispheric linkages in climate change: paleo-perspectives for future climate change. Climates of the Past 2, 167–185. Singarayer, J.S., Valdes, P.J., Friedlingstein, P., et al. (2011) Late Holocene methane rise caused by orbitally controlled increase in tropical sources. Nature 470, 82– 85. doi:10.1038/nature09739. Singhvi, A.K. and Porat, N. (2008) Impact of luminescence dating on geomorphological and palaeoclimate research in drylands. Boreas 536–558. Smagorinsky, J. (1963) General circulation experiments with the primitive equations. I. The basic experiment. Monthly Weather Review 91, 99–164. Sowers, T. (2009) Atmospheric methane isotope records covering the Holocene period. Quaternary Science Reviews 29, 213–221. Stager, J.C., Ryves, D., Cumming, B.F., et al. (2005) Solar variability and the levels of Lake Victoria, East Africa, during the last millennium. Journal of Paleolimnology 33, 243–251. 32 Chapter 1 Street, F.A. and Grove, A.T. (1976) Environmental and climatic implications of Late Quaternary lake level fluctuations in Africa. Nature 261, 385–390. Street-Perrott, F.A., Mitchell, J.F.B., Marchand, D.S. and Brunner, J.S. (1979) Global maps of lake-level fluctuations since 30 000 yr. BP. Quaternary Research 12, 83–118. Street-Perrott, F.A., Ficken, K.J., Huang, Y. and Eglinton, G. (2004) Late Quaternary changes in carbon cycling on Mt. Kenya, East Africa: an overview of the δ13C record in lacustrine organic matter. Quaternary Science Reviews 23, 861–879. Telfer, M.W. and Thomas, D.S.G. (2006) Complex Holocene lunette dune development, South Africa: implications for palaeoclimate and models of pan development in arid regions. Geology 34, 853– 856. Therrell, M.D., Stahle, D.W., Ries, L.P. and Shugart, H.H. (2006) Tree-ring reconstructed rainfall variability in Zimbabwe. Climate Dynamics 26, 677–685. Thomas, M.F. (2008) Understanding the impacts of Late Quaternary climate change in tropical and sub-tropical regions. Geomorphology 101, 146–158. Thomas, M.F., Nott, J. and Price, D.M. (2001) Late Quaternary stream sedimentation in the humid tropics: a review with new data from NE Queensland, Australia. Geomorphology 39, 53–68. Thomas, D.S.G., Knight, M. and Wiggs, G.F.S. (2005) Remobilization of southern African desert dune systems by twenty-first century global warming. Nature 435, 1218–1221. Thomas, M.F., Nott, J., Murray, A.S. and Price, D.M. (2007) Fluvial response to late Quaternary climate change in NE Queensland, Australia. Palaeogeography, Palaeoclimatology, Palaeoecology 251, 119–136. Thornthwaite, C.W. (1948) An approach towards a rational classification of climate. Geographical Review 38, 55–94. Trauth, M.H., Deino, A.L., Bergner, A.G.N. and Strecker, M.R. (2003) East African climate change and orbital forcing during the last 175 kyr BP. Earth and Planetary Science Letters 206, 297–313. Tudhope, A.W., Chilcott, C.P., McCulloch, M.T., et al. (2001) Variability in the El Niño-Southern Oscillation through a glacial-interglacial cycle. Science 291, 1511–1517. Van Loon, H., Meehl, G.A. and Arblaster, J.M. (2004) A decadal solar effect in the tropics in July and August. Journal of Atmospheric and Solar-Terrestrial Physics 66, 1767–1778. Versteegh, G.J.M. (2005) Solar forcing of climate. 2: evidence from the past. Space Science Reviews 120, 243–286. Walker, G.T. (1924) Correlation in seasonal variations in weather IV. A further study of world weather. Memoir India Meteorological Department 24, 275–332. Wang, B. and Ding, Q. (2008) Global monsoon: dominant mode of annual variation in the tropics. Dynamics of Atmospheres and Oceans 44, 165–183. Wang, X., Auler, A.S., Edwards, R.L., et al. (2006) Interhemispheric anti-phasing of rainfall during the last glacial period. Quaternary Science Reviews 25, 3391–3403. Wang, X., Auler, A.S., Edwards, R.L., et al. (2007) Millennial scale precipitation changes in southern Brazil over the past 90,000 years. Geophysical Research Letters 34, L23701. Wang, Y., Cheng, H., Edwards, R.L., et al. (2001) A high resolution absolute-dated Late Pleistocene monsoon record from Hulu Cave, China. Science 294, 2345–2348. Wang, Y., Cheng, H., Edwards, R.L., et al. (2008) Millennial- and orbital-scale changes in the East Asian monsoon over the past 224,000 years. Nature 451, 1090–1093. Williams, C., Flower, B.P., Hastings, D.W., et al. (2010) Deglacial abrupt climate change in the Atlantic Warm Pool: a Gulf of Mexico perspective. Paleoceanography 25, PA4221, doi:10.1029/2010PA001928. Williams, M.A., Ambrose, S.H., van der Kars, S., et al. (2009) Environmental impact of the 73 ka Toba supereruption in South Asia. Palaeogeography, Palaeoclimatology, Palaeoecology 284, 295–314. Wilson, E.O. (ed.) (1986) Biodiversity. National Academy Press, Washington D.C. Wolff, E. and Spahni, R. (2007) Methane and nitrous oxide in the ice core record. Philosophical Transactions of the Royal Society A 365, 1775–1792. Yoon, S.C., Won, J.G., Omar, A.H., et al. (2005) Estimation of the radiative forcing by key aerosol types in worldwide locations using a column model and AERONET data. Atmospheric Environment 39, 6620– 6630. Yue, X., Wang, H.J., Liao, H. and Jiang, D.B. (2011) Simulation of the direct radiative effect of mineral dust aerosol on the climate at the Last Glacial Maximum. Journal of Climate 24, 843–858. Zachos, J., Pagani, M., Sloan, L., et al. (2001) Trends, rhythms and aberrations in global climate 65Ma to present. Science 292, 686–693. Introduction Zebiak, S.E. and Cane, M.A. (1987) A model El NiñoSouthern Oscillation. Monthly Weather Review 115, 2262–2278. Zinke, J., Dullo, W.-Chr., Heiss, G.A. and Eisenhauer, A. (2004) ENSO and Indian Ocean subtropical dipole variability is recorded in a coral record off southwest Mada- 33 gascar for the period 1659 to 1995. Earth and Planetary Science Letters 228, 177–194. Zinke, J., Pfeiffer, M., Timm, O., et al. (2005) Atmosphereocean dynamics in the Western Indian Ocean recorded in corals. Philosophical Transactions of the Royal Society A 363, 121–142.