The environmental context for the origins of modern human diversity: A synthesis of regional variability in African climate 150,000–30,000 years ago

https://doi.org/10.1016/j.jhevol.2012.01.011Get rights and content

Abstract

We synthesize African paleoclimate from 150 to 30 ka (thousand years ago) using 85 diverse datasets at a regional scale, testing for coherence with North Atlantic glacial/interglacial phases and northern and southern hemisphere insolation cycles. Two major determinants of circum-African climate variability over this time period are supported by principal components analysis: North Atlantic sea surface temperature (SST) variations and local insolation maxima. North Atlantic SSTs correlated with the variability found in most circum-African SST records, whereas the variability of the majority of terrestrial temperature and precipitation records is explained by local insolation maxima, particularly at times when solar radiation was intense and highly variable (e.g., 150–75 ka). We demonstrate that climates varied with latitude, such that periods of relatively increased aridity or humidity were asynchronous across the northern, eastern, tropical and southern portions of Africa. Comparisons of the archaeological, fossil, or genetic records with generalized patterns of environmental change based solely on northern hemisphere glacial/interglacial cycles are therefore imprecise.

We compare our refined climatic framework to a database of 64 radiometrically-dated paleoanthropological sites to test hypotheses of demographic response to climatic change among African hominin populations during the 150–30 ka interval. We argue that at a continental scale, population and climate changes were asynchronous and likely occurred under different regimes of climate forcing, creating alternating opportunities for migration into adjacent regions. Our results suggest little relation between large scale demographic and climate change in southern Africa during this time span, but strongly support the hypothesis of hominin occupation of the Sahara during discrete humid intervals ∼135–115 ka and 105–75 ka. Hominin populations in equatorial and eastern Africa may have been buffered from the extremes of climate change by locally steep altitudinal and rainfall gradients and the complex and variable effects of increased aridity on human habitat suitability in the tropics. Our data are consistent with hominin migrations out of Africa through varying exit points from ∼140–80 ka.

Introduction

Fossil and genetic data are consistent with an African origin of Homo sapiens by 195,000 years ago (ka) (Ingman et al., 2000; Clark et al., 2003; White et al., 2003; McDougall et al., 2005; Gonder et al., 2007). Numerous studies have emphasized the diversity in morphology, life history, and genetic signatures likely present among populations of Middle and Late Pleistocene hominins, some of which dispersed from Africa in the Late Pleistocene, replacing existing hominin populations in parts of Eurasia and eventually colonizing Australia and North and South America. The variability among the source populations within Africa is compounded by local demographic changes and range expansions or contractions, as well as the persistence of ancestral or sister taxa (Lahr and Foley, 1998; Howell, 1999; Excoffier, 2002; Forster, 2004; Harding and McVean, 2004; Eswaran et al., 2005; Prugnolle et al., 2005; Trinkaus, 2005; Garrigan and Hammer, 2006; Garrigan et al., 2007; Smith et al., 2007; Bräuer, 2008; Pearson, 2008; Crevecoeur et al., 2009; Gunz et al., 2009; Hammer et al., 2011; Harvati et al., 2011). Middle Stone Age (MSA) archaeological sites provide the strongest record of the behavior of early African populations of H. sapiens, and these sites record pronounced spatial and temporal variation not seen in earlier periods (McBrearty and Brooks, 2000; Henshilwood and Marean, 2003; Marean and Assefa, 2005; Jacobs et al., 2008a; Shea, 2011).

Paleoenvironmental history and variability during the Middle and Late Pleistocene may have played a crucial role in shaping the biological diversity, distribution and behavior of H. sapiens populations during this period. Environmental change is a causal mechanism for the dispersal and isolation of animal populations, and the accumulation of variation through drift resulting from geographic isolation (allopatry) is a central cause of biological divergence and speciation (e.g., Barraclough and Nee, 2001; Mayr, 2001). Human behavioral change occurs as a result of a range of causal factors and we do not advocate strict environmental determinism. However, recent and ancient human forager subsistence, technology, land use, and some social behaviors show clear relationships with environment (Kelly, 1995; Binford, 2001; Kuhn and Stiner, 2001; Marlowe, 2005), and archaeological changes at some Middle Stone Age (MSA) sites have been interpreted as behavioral responses to variation in resource structure and availability as a result of environmental shifts (Ambrose and Lorenz, 1990; Marean et al., 2007; McCall, 2007). Our objective here is to synthesize available African Middle and Late Pleistocene paleoclimatic and paleoanthropological records relevant to early populations of H. sapiens. Although a number of studies have focused on providing the environmental context for subsequent hominin dispersals out of Africa (cf. Carto et al., 2009; Osbourne et al., 2008; Vaks et al., 2007; but see; Basell, 2008; Cowling et al., 2008; Drake et al., 2008), recent genetic evidence has also indicated the importance of Pleistocene population dispersal within Africa and its importance in shaping modern human genetic diversity (e.g., Reed and Tishkoff, 2006; Behar et al., 2008; Tishkoff et al., 2009; Verdu et al., 2009; see also Hammer et al., 2011). Our goal therefore is to explore climatic variability within Africa during this time period to better understand both the context of intra-African hominin population dispersal and conditions relevant to out-of-Africa scenarios.

We emphasize the importance of considering different temporal and spatial frameworks for interpreting the role of climate change in human evolution, and these can be conceived of as occupying a continuum from macro-scale to micro-scale approaches. Macro-scale approaches include those that examine continental scale changes across long time intervals (e.g., Potts, 1998; deMenocal, 2004) or use coarse temporal frameworks such as the Marine Isotope Stage boundaries and glacial/interglacial variation (e.g., Lahr and Foley, 1998; Marean and Assefa, 2005; Basell, 2008; Crowley and Hyde, 2008). At the other end of the continuum are micro-scale approaches that seek to understand the response of small populations to environmental change over relatively short periods recorded at a single archaeological site or depositional basin (e.g., Potts et al., 1999). We are not suggesting that either macro-scale or micro-scale approaches are ‘better’, but rather that the scale of the available data should match the scale of the questions being asked.

Following other recent efforts (e.g., Fisher et al., 2010; Marean et al., 2010; Drake et al., 2011), we employ what might be termed a ‘meso-scale’ approach in the examination of sub-continental variation in paleoclimate and the human fossil and archaeological records. To do so, we synthesize paleoclimate data from diverse datasets of varying degrees of temporal and spatial resolution, divided into three broad categories: (1) cores from offshore marine sites, whose age models are largely constrained by the global oxygen isotope time scale. These provide relatively continuous data sampled over a large area, and provide key data on sea surface temperatures and ocean circulation patterns, changes in terrestrial vegetation patterns and relative aridity; (2) cores from a number of African lakes, particularly Lake Malawi and Lake Tanganyika. As well as the impact crater lakes Tswaing and Bosumtwi, which provide relatively continuous records of watershed and regional environmental change; (3) a number of other terrestrial sedimentary archives, including caves, rockshelters, and open-air archaeological sites that provide temporally discontinuous, often highly local records of climatic and environmental change. We compare these climatic and environmental data to a database of published, radiometrically dated hominin fossil- or artifact-bearing deposits to test whether these changes are coincident with demographic changes in Pleistocene African hominin populations.

We focus on the 150–30 ka time interval for theoretical and practical reasons. This time period is important as it includes much of the early history of H. sapiens, and from an archaeological perspective, it includes the shift from MSA to Later Stone Age (LSA) technologies, an important change that may signal changes in human cognition or demography (cf. Klein, 2009; Powell et al., 2009). Pragmatically, 150 ka marks the beginning of the time interval for which statistically significant numbers of detailed, well-constrained records are available from both the oceanic sites surrounding Africa and for a number of the key lake records such as Lake Malawi. In addition, the paleoclimate records for the past 30 kyr (thousand years) in Africa have long been a focus of earlier reviews (e.g., Street and Grove, 1979; Nicholson and Flohn, 1980). Because we are aggregating datasets of varying degrees of temporal resolution, often particularly poor for the archaeological record, we examine paleoclimate change using 5 kyr intervals and paleoanthropological change using estimated 10 kyr intervals, as detailed below.

Section snippets

Research questions and hypotheses

We synthesize African paleoclimate and paleoanthropological data from 150 to 30 ka, and use these data to test a number of hypotheses.

Methods

To test these hypotheses, we constructed a database from the published literature, synthesizing all relevant dated climatic and paleoanthropological data. The geographic locations of these datasets are shown in Figure 1, Figure 2, with data synthesized in Tables 1 and 2. Note that throughout the paper, paleoclimate data archives are referenced by the numbers presented in Table 1.

Interpretive frameworks for African paleoclimates

Modern African climates provide the basic interpretive framework for interpreting Pleistocene climate and environment. The climate of a given location describes the average annual as well as seasonal variation in temperature and precipitation over a long period of time. Climate dynamics refers to the physical mechanisms that create a particular climate in one location, and a different climate in another.

Paleoclimate results: sea surface temperatures and paleoceanographic conditions

Sea surface temperatures (SSTs) typically reflect changes in mean global temperature because they are global in extent, and water has a high heat capacity. The various measures of SST are summarized in Table 3. Fig. 4 summarizes various measures of sea surface temperatures for the northern Atlantic, Mediterranean, southern Atlantic, Indian Ocean, and Southern Ocean regions.

Paleoclimate results: terrestrial and marine records of continental temperature and precipitation variability

SSTs are an important measure of global temperature and have an important impact on continental moisture availability, but the relationship between oceanic and continental records is complex. The continental records, although often providing less continuous data than the oceanic records, are more directly relevant for understanding the climatic and environmental context of Pleistocene human evolution. Areal coverage and the types of paleoclimate proxies and indicators available are more varied

Regional coherence of temporal trends in African climate change

The evidence presented above indicates that SSTs often co-vary with climate changes predicted by δ18O-derived MIS boundaries, with terrestrial data often out of phase although varying by sub-region. Principal components analysis (PCA) provides a more formal means of comparing aggregate data from both marine and terrestrial sources. On a PCA plot, datasets with similar patterns of variability will plot together. The PCA results are presented in Figure 9, Figure 10, Figure 11. For the purposes of

An evaluation of causal mechanisms

Throughout the discussion of the SST records, there was a recurring theme: all records seem to follow global MIS boundaries, except where they do not. In other words, circum-Africa SSTs record both the global temperature variability, as well as perturbations in that overall signal by local or regional effects such as changing wind zonality and/or upwelling activity. As illustrated in the previous section, terrestrial temperature and precipitation across Africa do not seem to be responding to

Synthesis: African climate from 150 to 30 ka

African paleoclimates from 150 to 30 ka are spatially and temporally complex, with variation in the outcome of multiple related processes. We summarize these here before considering their implications for human evolution, determining the degree to which our hypotheses are supported or rejected.

Hypothesis 1

Pleistocene African climate change is coincident with North Atlantic glacial/interglacial periods.

Fig. 14 demonstrates that although SSTs do tend to vary with glacial/interglacial cycles and MIS

Paleoanthropological implications

The results of our regional comparisons of paleoanthropological site density and inferred climate for each of the four sub-regions are summarized in Fig. 15. The results are used to test each of our three hypotheses linking climate change to Pleistocene African hominin demography, followed by suggestions for further testing at similar and finer analytical scales. Overall, there is a general increase in the relative frequency of sites over time in most regions (Fig. 15). This pattern is

Conclusions

In this paper. we synthesized African paleoclimate from 150 ka to 30 ka using diverse datasets at a regional scale, testing for coherence with North Atlantic glacial/interglacial phases, Northern and Southern Hemisphere insolation cycles, and the atmospheric dynamics responsible for the observed regional climates. We have examined the temporal and spatial components of paleoenvironmental change across the continent, in order to further investigate the relationship between the varying regional

Acknowledgments

Many thanks to the NSF Integrative Graduate Education and Research Traineeship (IGERT) program and the Foreign Language and Area Studies (FLAS) fellowship program for providing financial support during the research and writing of this manuscript. In addition, many thanks to colleagues who provided comments, advice, and support throughout the process, especially Michael McGlue, Christine Gans, Jessica Conroy, Sarah Ivory, Mark Warren, Daniel Peppe, and Tyler Faith. Sarah Pilliard and Ryan

References (255)

  • C. Batini et al.

    Phylogeography of the human mitochondrial L1c haplogroup: genetic signatures of the prehistory of Central Africa

    Mol. Phylogenet. Evol.

    (2007)
  • D.M. Behar et al.

    The dawn of human matrilineal diversity

    Am. J. Hum. Genet.

    (2008)
  • M.I. Bird et al.

    Radiocarbon dating from 40 to 60 ka BP at Border Cave, South Africa

    Quatern. Sci. Rev.

    (2003)
  • S.L. Burrough et al.

    Multiphase Quaternary highstands at Lake Ngami, Kalahari, northern Botswana

    Palaeogeogr. Palaeoclimatol. Palaeoecol.

    (2007)
  • S.L. Burrough et al.

    Late Quaternary hydrological dynamics in the Middle Kalahari: forcing and feedbacks

    Earth-Sci. Rev.

    (2009)
  • K.W. Butzer et al.

    Lithostratigraphy of Border Cave, Kwazulu, South-Africa: a Middle Stone-Age sequence beginning c. 195,000 b.p

    J. Archaeol. Sci.

    (1978)
  • A.S. Carr et al.

    Developing a 150 ka luminescence chronology for the barrier dunes of the southern Cape, South Africa

    Quatern. Geochronol.

    (2007)
  • A.S. Carr et al.

    Climatic and sea level controls on Late Quaternary eolian activity on the Agulhas Plain, South Africa

    Quatern. Res.

    (2006)
  • S.L. Carto et al.

    Out of Africa and into an ice age: on the role of global climate change in the late Pleistocene migration of early modern humans out of Africa

    J. Hum. Evol.

    (2009)
  • C. Causse et al.

    Humidity changes in southern Tunisia during the late Pleistocene inferred from U-Th dating of mollusc shells

    Appl. Geochem.

    (2003)
  • B.M. Chase

    South African palaeoenvironments during marine oxygen isotope stage 4: a context for the Howiesons Poort and Still Bay industries

    J. Archaeol. Sci.

    (2010)
  • M.K. Crombie et al.

    Age and isotopic constraints on Pleistocene pluvial episodes in the Western Desert

    Egypt. Palaeogeogr. Palaeoclimatol. Palaeoecol.

    (1997)
  • P.B. deMenocal

    African climate change and faunal evolution during the Pliocene-Pleistocene

    Earth Planet. Sci. Lett.

    (2004)
  • M. Domínguez-Rodrigo et al.

    A new archaic Homo sapiens fossil from Lake Eyasi, Tanzania

    J. Hum. Evol.

    (2008)
  • N.A. Drake et al.

    Palaeohydrology of the Fazzan Basin and surrounding regions: the last 7 million years

    Paleogeogr. Paleoclimatol. Paleoecol.

    (2008)
  • L.M. Dupont et al.

    Vegetation change in equatorial West Africa: time slices for the last 150 ka

    Palaeogeogr. Palaeoclimatol. Palaeoecol.

    (2000)
  • V. Eswaran et al.

    Genomics refutes an exclusively African origin of humans

    J. Hum. Evol.

    (2005)
  • L. Excoffier

    Human demographic history: refining the recent African origin model

    Curr. Opin. Genet. Dev.

    (2002)
  • J.T. Faith et al.

    Taxonomic status and paleoecology of Rusingoryx atopocranion (Mammalia, Artiodactyla), an extinct Pleistocene bovid from Rusinga Island, Kenya

    Quatern. Res.

    (2011)
  • J.K. Feathers

    Luminescence dating in less than ideal conditions: case studies from Klasies River main site and Duinefontein, South Africa

    J. Archaeol. Sci.

    (2002)
  • J.K. Feathers et al.

    Luminescence dating of Middle Stone Age deposits at Die Kelders

    J. Hum. Evol.

    (2000)
  • J.K. Feathers et al.

    Luminescence dating at Katanda: a reassessment

    Quatern. Sci. Rev.

    (2001)
  • E.C. Fisher et al.

    Middle and Late Pleistocene paleoscape modeling along the southern coast of South Africa

    Quatern. Sci. Rev.

    (2010)
  • J.C. Fontes et al.

    Palhydaf (paleohydrology in Africa) program: objectives, methods, major results

    Palaeogeogr. Palaeoclimatol. Palaeoecol.

    (1991)
  • R. Grün et al.

    On the age of Border Cave 5 human mandible

    J. Hum. Evol.

    (2003)
  • R.M. Harding et al.

    A structured ancestral population for the evolution of modern humans

    Curr. Opin. Genet. Dev.

    (2004)
  • C. Hillaire-Marcel et al.

    14C and Th/U dating of Pleistocene and Holocene stromatolites from East Africa paleolakes

    Quatern. Res.

    (1986)
  • K. Holmgren et al.

    Paleoclimatic significance of the stable isotope composition and petrology of a Late Pleistocene stalagmite from Botswana

    Quatern. Res.

    (1995)
  • S. Holzkamper et al.

    Late Pleistocene stalagmite growth in Wolkberg Cave, South Africa

    Earth Planet. Sci. Lett.

    (2009)
  • S.H. Ambrose

    Middle and Later Stone Age settlement patterns in the Central Rift Valley, Kenya: comparisons and contrasts

  • S.H. Ambrose

    Population bottleneck

  • S.H. Ambrose et al.

    Social and ecological models for the Middle Stone Age in southern Africa

  • S.J. Armitage et al.

    The southern route “Out of Africa”: evidence for an early expansion of modern humans into Arabia

    Science

    (2011)
  • Z. Assefa et al.

    Symbolic use of terrestrial gastropod opercula during the Middle Stone Age at Porc-Epic Cave, Ethiopia

    Curr. Anthropol.

    (2008)
  • E. Bard et al.

    Interhemispheric synchrony of the last deglaciation inferred from alkenone paleothermometry

    Nature

    (1997)
  • L.H. Barham

    The Middle Stone Age of Zambia, South Central Africa

    (2000)
  • L.H. Barham et al.

    The First Africans: African Archaeology from the Earliest Tool Makers to Most Recent Foragers

    (2008)
  • B.E. Barich et al.

    Between the Mediterranean and the Sahara: geoarchaeological reconnaissance in the Jebel Gharbi, Libya

    Antiquity

    (2005)
  • C. Batini et al.

    Insights into the demographic history of African Pygmies from complete mitochondrial genomes

    Mol. Biol. Evol.

    (2011)
  • P.B. Beaumont et al.

    On a timescale for the past million years of human history in central South Africa

    S. Afr. J. Sci.

    (2006)
  • Cited by (229)

    View all citing articles on Scopus
    View full text