High- and low-latitude forcing of Plio-Pleistocene East African climate and human evolution

Journal of Human Evolution 53 (2007) 475e486

High- and low-latitude forcing of Plio-Pleistocene East African climate and human evolution Martin H. Trauth a,*, Mark A. Maslin b, Alan L. Deino c, Manfred R. Strecker a, Andreas G.N. Bergner a, Miriam Du¨hnforth d a

Institut fu¨r Geowissenschaften, Universita¨t Potsdam, POB 601553, D-14415 Potsdam, Germany Environmental Change Research Center, Department of Geography, University College London, UK c Berkeley Geochronology Center, 2445 Ridge Road, Berkeley, CA 94709, USA d Institute of Geology, ETH Zu¨rich 8092 Zu¨rich, Switzerland

b

Received 4 November 2005; accepted 15 December 2006

Abstract The late Cenozoic climate of East Africa is punctuated by episodes of short, alternating periods of extreme wetness and aridity, superimposed on a regime of subdued moisture availability exhibiting a long-term drying trend. These periods of extreme climate variability appear to correlate with maxima in the 400-thousand-year (kyr) component of the Earth’s eccentricity cycle. Prior to 2.7 Ma the wet phases appear every 400 kyrs, whereas after 2.7 Ma, the wet phases appear every 800 kyrs, with periods of precessional-forced extreme climate variability at 2.7e2.5 Ma, 1.9e 1.7 Ma, and 1.1e0.9 Ma before present. The last three major lake phases occur at the times of major global climatic transitions, such as the onset of Northern Hemisphere Glaciation (2.7e2.5 Ma), intensification of the Walker Circulation (1.9e1.7 Ma), and the Mid-Pleistocene Revolution (1.0e0.7 Ma). High-latitude forcing is required to compress the Intertropical Convergence Zone so that East Africa becomes locally sensitive to precessional forcing, resulting in rapid shifts from wet to dry conditions. These periods of extreme climate variability may have provided a catalyst for evolutionary change and driven key speciation and dispersal events amongst mammals and hominins in East Africa. Ó 2007 Elsevier Ltd. All rights reserved. Keywords: Lake sediments;

40

Ar/39Ar dating; East African Rift System; Hominins

Introduction Long-term climate change seems to be modulated primarily by tectonic changes both at the global and local scale. Late Cenozoic global cooling has been ascribed to both the uplift of Tibet (Ruddiman and Raymo, 1988) and the closure of the Panama Isthmus (Haug and Tiedemann, 1998), although the exact role of atmospheric carbon dioxide is still unclear (Sundquist and Visser, 2004). In East Africa, long-term climate change is also controlled by local tectonics, especially the dynamic development of the branching East African Rift System (EARS). Thus, early hominin evolution in East Africa * Corresponding author. E-mail address: [email protected] (M.H. Trauth). 0047-2484/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.jhevol.2006.12.009

occurs in synchrony with both long-term global cooling and extensive local tectonic changes (deMenocal, 1995, 2004; Trauth et al., 2005). There is a compelling need to understand how these two environmental factors interact at the local scale and affect flora and fauna living in the East African Rift. The Rift Valley lakes are excellent recorders of environmental changes in East Africa. Whereas the western branch of the EARS contains several large and deep lakes that have formed during the last 10 Ma, a series of small, partly alkaline lakes have developed in the eastern branch during Plio-Pleistocene times (Tiercelin, 1986). The lake history in the Ethiopian and Kenyan Rifts is complex and closely tied to the volcano-tectonic activity leading to formation of internally drained basins and changes in catchment areas (e.g., Tiercelin and Lezzar, 2002; Fig. 1). Although much smaller and often

M.H. Trauth et al. / Journal of Human Evolution 53 (2007) 475e486

476

Fig. 1. Maps of East African showing (A) topography, faults, and lake basins (modified from Trauth et al., 2005), as well as average precipitation and air flow during the (B) long rainy season (April-May) and (C) short rainy season (OctobereNovember). Climate pattern modified from Leroux (2001).

subaerially exposed, these basins host a rich sedimentary record with intercalated volcanic deposits that allow constraining of different lake-level phases (for an overview see Yuretich, 1982; Tiercelin and Lezzar, 2002; Owen, 2002; Barker et al., 2004; Fig. 2).

A

4.5–3.4 Ma

Kinangop Tuffs (ca. 130 m)

(Clarke et al., 1990)

In this article, we define late Cenozoic changes in East African climate using several detailed paleoclimate records from fluviolacustrine sediments exposed in the Ethiopian, Kenyan, and Tanzanian Rifts. We use the sedimentary record of ten lake basins on a north-south transect in the EARS to separate

C

B

0.724±0.003 Ma

Phonolithe

(sample MUN-3)

(no exposure)

Gravell, welded on top

Pure white diatomite with plant remains

(eroded on top) Volcanic ash Pure white diatomite Pumice-lapilli tuff Pure white diatomite (Erosional unconformity)

Silts with rootlets Gravel

Cross-bedded sand Diatomite, in layers diagenetically altered

1.65±0.01 Ma

Pumice lapilli tuff

(sample MUN-1)

Cross-bedded sand Silty diatomite

Green silts with intercalated clays and sands

(Erosional unconformity)

Pure white diatomite

Pumice-lapilli tuff Bioturbated silts Pumice-lapilli tuff

Tuff Diatomite with rootlets Tuff Silty diatomite Hyaloclastite

1.75–1.65 Ma (Strecker, 1991;Boven, 1992)

Laminated silts and sands with intercalated diatomites

1.71±0.02 Ma (sample MUN-2)

Diatomite with intercalated tuffs

0.946 Ma (Evernden and Curtis, 1965)

Gray ash Pure white diatomite Gray ash Pure white diatomite Gray ash

0.977±0.01 Ma

Gray pumice lapilli tuff

(sample KAR02-2) (Erosional unconformity) Pure white diatomite Buff-colored tuff

Deep cracks filled with pyroclastics Diatomite with intercalated tuffs

Light-gray silt Pure white diatomite, greenish in lower parts, plant remains

(10 more meters) Pure white diatomite

Silty diatomite

not datable

Silts with intercalated clays, soft sediment deformation

(sample KAR02-1) Pure white diatomite

Laminated silts

(10 more meters) > 4 m pure white diatomite Layered diatomite

4.324±0.005 Ma

(no exposure) Tuffs

(sample TRSH94-2)

Pure white diatomite Green tuff

4.702±0.006 Ma

Diatomite, with intercalated silt layers, plant remains

(sample TRSH94-1)

ca. 4.6 Ma

(Erosional unconformity) Turasha Trachyte

(Strecker, 1991; Boven, 1992)

Turasha

Pure white diatomite (Erosional unconformity) Limuru Trachyte

1.96 Ma (Baker et al., 1988)

Gicheru

0.987±0.003 Ma (sample KAR94-3)

(Erosional unconformity) Gilgil Trachytes

1 Meter

Kariandusi

Fig. 2. Stratigraphic sequence of the Turasha, Gicheru, and Kariandusi sections described in the text. See Fig. 1 for locations and see Trauth et al. (2005) for 40 Ar/39Ar details.

M.H. Trauth et al. / Journal of Human Evolution 53 (2007) 475e486

477

volcano-tectonic and climatic influences on rift sedimentation. Synchronous changes in moisture balance inferred from lacustrine sediments and diatom assemblages contrast with the volcano-tectonic history and are attributed to climate change. These climatic records are used to examine possible links between climate change and faunal changes during the late Cenozoic.

at 1.2 Ma and produced the present-day rift escarpments (Foster et al., 1997). Subsequent late Quaternary structural en echelo´n segmentation during west-northwest to east-southeast oriented extension created numerous subbasins in the individual rift sectors that repeatedly hosted smaller lakes during the Pleistocene (Baker et al., 1988; Strecker et al., 1990; Bosworth and Strecker, 1997).

Volcano-tectonic evolution of the East African Rift

Reconstructing the history of the East African lakes

The East African Rift System is exemplary for the diversity of sedimentary environments (Fig. 1). Structurally and magmatically controlled processes have created complex relief and drainage conditions that are highly variable through time, with early manifestations recorded at about 45 Ma and continuing into the present (e.g., Baker et al., 1972; Strecker et al., 1990; KRISP Working Party, 1991; Morley, 1994; Prodehl et al., 1997; George et al., 1998; Ebinger et al., 2000; Macdonald et al., 2001). Volcanism and faulting were asynchronous, suggesting that rifting is related to anomalously hot mantle beneath the present rift zone (Morley, 1994; Smith, 1994; Mechie et al., 1997; Ebinger and Sleep, 1998). In the Ethiopian Rift, volcanism started between 45 and 33 Ma. In northern Kenya rifting began at around 33 Ma and continued to about 25 Ma, whereas the magmatic activity of the central and southern segments of the rift in Kenya and Tanzania started between 15 and 8 Ma (Bagdasaryan et al., 1973; Crossley, 1979; Davidson and Rex, 1980; Crossley and Knight, 1981; McDougall and Watkins, 1988; Dawson, 1992; Morley et al., 1992; Ebinger et al., 1993; George et al., 1998). Although these early stages of rifting were characterized by doming and down-warping, in the later rifting fault propagation progressed from north to south. Major faulting in Ethiopia between 20 to 14 Ma was followed by the generation of eastdipping faults in northern Kenya between 12 and 7 Ma, and superseded by normal faulting on the western side of the central and southern Kenya Rift between 9 and 6 Ma (Williams et al., 1983; Baker et al., 1988; Blisniuk and Strecker, 1990; Ebinger et al., 2000). These early half-grabens were subsequently faulted antithetically between about 5.5 and 3.7 Ma, which generated a full-graben morphology (Baker et al., 1988; Blisniuk and Strecker, 1990). Prior to the full-graben stage, the large Aberdare volcanic complex, with elevation in excess of 4,000 m and now an important orographic barrier in Kenya, was established (Baker et al., 1972; Williams et al., 1983; Fig. 1). By 2.6 Ma, the graben was further segmented in the central Kenya Rift by west-dipping faults, creating the 30km-wide intrarift Kinangop Plateau and the tectonically active 40-km-wide inner rift (Baker et al., 1988; Strecker et al., 1990). The inner rift was subsequently covered by trachytic, basaltic, and rhyolithic lavas, and continues to be affected by normal faulting, leading to further segmentation of the rift floor (Strecker et al., 1990; Bosworth and Strecker, 1997). There is no unambiguous evidence from sedimentation patterns in the Magadi-Natron and Olduvai Basins, however, that major fault-bounded basins existed at that time (Foster et al., 1997). Major normal faulting in these regions occurred

The southward propagation of rifting, including the formation of faults and the magmatic activity, is also reflected in the earliest formation of lake basins in the northern parts of the rift. The fluviolacustrine history of the Afar, Omo-Turkana, and Baringo-Bogoria Basins in the north started in the middle and upper Miocene, whereas the oldest lacustrine sequences in the central and southern segments of the rift in Kenya and Tanzania occur during the early Pliocene (Tiercelin and Lezzar, 2002). In the following, we provide a compilation of the fluviolacustrine history of these basins from the north (Ethiopian Rift, Afar Basins, Omo-Turkana Basin) to the south (Suguta Valley, Baringo-Bogoria Basin, Nakuru-Elmenteita Basin, Naivasha Basin, Munyu wa Gicheru, Olorgesailie, MagadiNatron Basin, and Olduvai Gorge; Fig. 1). For the reconstruction of the lake histories in the Central Kenya Rift, we sampled several tens-of-meters-thick lacustrine sequences at meter to half-meter intervals (Figs. 2 and 3). The sediments consist mainly of diatomite, and were investigated for sediment characteristics and diatom flora (Fig. 4). We used the following indicators for large and deep freshwater paleolakes: 1) the presence of pure white and frequently laminated diatomite, 2) typical freshwater diatom assemblages, and 3) a diatom flora clearly dominated by planktonic species, whereas benthic or epiphytic taxa are less frequent or absent. In contrast, a shallow and more alkaline lake is characterized by: 1) a significant clastic component in the diatomites, 2) diatom indicators for higher alkaline conditions, and 3) a significant benthic-epiphytic diatom community and the presence of abundant phytoliths and sponge spicules (Fig. 4).

Fig. 3. Outcrop photograph of the Gicheru section. The 29-m-thick sequence at Munyu wa Gicheru located in a w2-Ma-year-old graben consists of mainly pure white diatomite, intruded by hyaloclastic rhyolites and pumice lapilli tuffs.

478

M.H. Trauth et al. / Journal of Human Evolution 53 (2007) 475e486

Fig. 4. Microscopic images of diatomaceous sediments from the Central Kenya Rift. (A) Sediments dominated by glass shards, broken diatom frustules, sponge spicules, and phytoliths indicating an extremely shallow lake, (B) freshwater diatom assemblage containing a mixed planktonic and benthic/epiphytic flora indicating shallow freshwater conditions of a larger lake, (C) predominantly planktonic diatoms such as large Stephanodiscus spp. (i.e., S. niagarae, S. transylvanicus and S. carconensis), and Aulacoseira spp., such as A. granulata, indicating a large deep lake with transparent, very dilute water, a low silica level, and possibly cool temperatures.

Based on these characteristics, we used the following classification of paleoenvironments in the lake basins. A deep freshwater lake characterized by a size of several 100 km2, water depths of significantly higher than 150 m, and a neutral pH is documented by pure white and frequently laminated diatomite, and planktonic/litoral diatom ratios reaching 10 and 100. In contrast, shallow and more alkaline lakes are typically less than 150 km2 in size, have water depths of less than 100 m (often only a few meters), and dry out episodially. The pH of these lakes is often at around 8, but may reach significantly higher values. The corresponding sediments are clayey diatomites and silts, containing a diatom flora with planktonic/litoral diatom ratios of less than 1 (typically between 0.1e0.3 and, in some cases, up to 0.8). In extreme cases, the sediments contain authigenic silicates, such as zeolites, that document chemical weathering of silicic volcanic glass in an extremely alkaline lake environment. No lake, however, indicates the complete absence of lake sediments. The sedimentary record of a basin contains fluvial or eolian deposits and primarily deposited or reworked pyroclastic material. The age control for the lake periods was obtained by radiometric age determinations of anorthoclase and sanidine phenocryst concentrates from several tuff beds and lava flows. Analyses were performed at the Berkeley Geochronology Center using the single-crystal, total-fusion 40Ar/39Ar method. The analytical details, as well as all results from 40Ar/39Ar dating, were previously described in Trauth et al. (2005), whereas this paper now provides section graphs of the sites described in the text (Fig. 2). The lacustrine history of the central Kenya Rift based on these sections is correlated with records from published work on lake basins in Ethiopia, Kenya, and Tanzania (Fig. 5). Where available, the data contained in the cited references are interpreted using similar criteria as listed above. In many cases, the authors were contacted to discuss their data to extract a consistent interpretation of East African paleoenvironments for the last three million years. This compilation suggests that there were significant late Cenozoic lake periods between w4.7e4.3 Ma, 4.0e3.9 Ma, w3.4e3.3 Ma, w3.20e2.95 Ma, 2.7e2.5 Ma, 1.9e1.7 Ma, and 1.1e0.9 Ma before present in East Africa.

Lake-level fluctuations and climate change During the last five million years, several large lakes were present in the Ethiopian Rift and Afar Basin (Fig. 5). The Konso-Gardula (between 1.9 and 1.3 Ma: Asfaw et al., 1992) and Gadeb sedimentary sequences (between 2.7 to 2.3 Ma: Williams et al., 1979; Gasse, 1990; WoldeGabriel et al., 2000) suggest that lakes existed in the southern sector of the Ethiopian Rift at least temporarily during that time. Contemporaneous lacustrine deposits are exposed on the floors of the Dikhil and Abhe´-Gobaad Basins or are interbedded with the latest basaltic lava flows of the present-day plateaus (Gasse, 1990). The lacustrine strata in the central Afar Basin contain diatom assemblages clearly dominated by large planktonic Stephanodiscus spp., such as S. niagarae and S. carconensis, indicating large deep lakes with transparent, very dilute water, a low silica level, and possibly cool temperatures (Gasse, 1990). A second important period of generally humid climate in this region occurs between 1.0 and 0.8 Ma, again registered by freshwater diatomites containing species which still live in large temperate lakes (Stephanodiscus spp.: Gasse, 1990). After this humid phase, gradually drier conditions led to a regression of these large lakes and the disappearance of the large Stephanodiscus spp.; no lacustrine information can be attributed with certainty to the arid middle Pleistocene, whereas there is evidence for late Pleistocene and early Holocene lakes in the area (Gasse, 1977). Dominated by fluvial conditions for most of its history, the Omo-Turkana Basin offers an enormous dataset for environmental change during the last 5 Ma (e.g., McDougall, 1985; McDougall and Watkins, 1988; Brown and Feibel, 1991; Leakey et al., 1995, 1998; Potts, 1998). Where some evidence for lacustrine conditions exist at 4.2e4.0 Ma, w3.5 Ma, and w3.0 Ma, between 2.3 and 1.9 Ma the sedimentary environment was clearly dominated by fluvial conditions with a large axial meandering river fed by braided streams that descended from the eastern and western rift shoulders (Brown and Feibel, 1991). At about 1.9 Ma, the environmental conditions changed dramatically and the basin was occupied by a large lake fed from the north by the Omo River (Brown

M.H. Trauth et al. / Journal of Human Evolution 53 (2007) 475e486

479

Fig. 5. Compilation of tectonic features and prominent lake periods of the eastern branch of the East African Rift System. Tectonic features and event complied from Williams et al. (1983), Baker et al. (1988), Strecker et al. (1990), Foster et al. (1997) and Ebinger et al. (2000). Paleo-environmental and radiometric age data in million years for Olduvai Basin from Walter et al. (1991) and Ashley and Hay (2002); Magadi-Natron and Olorgesailie Basin from Potts (1998, 1999) and Behrensmeyer et al. (2002); Natron Basin has one persistent lacustrine interval (a member of the Monink Formation called the Moinik Clays) dated to 1.1e1.0 Ma (Deino, pers. communication); Gicheru Basin from Baker et al. (1988), Strecker (1991), Boven (1992) and this work; Naivasha Basin from Richardson and Richardson (1972), Richardson and Dussinger (1986), Strecker et al. (1990) and Trauth et al. (2003, 2005); Nakuru-Elmenteita Basin from Evernden and Curtis (1965), Washbourn-Kamau (1970, 1975, 1977), Richardson and Richardson (1972), Richardson and Dussinger (1988), Strecker (1991), Boven (1992) and Trauth et al. (2005); Baringo-Bogoria Basin from Owen (2002) and Deino et al. (2006); Suguta Basin from Butzer et al. (1969), Hillarie-Marcel et al. (1986) and Sturchio et al. (1993); Omo-Turkana Basin from McDougall and Watkins (1988), Brown and Feibel (1991) and Leakey et al. (1995, 1998); Ethiopian Rift from Williams et al. (1979), Gasse (1990) and WoldeGabriel et al. (2000); Afar Basin from Gasse (1990).

480

M.H. Trauth et al. / Journal of Human Evolution 53 (2007) 475e486

and Feibel, 1991). By 1.7 Ma, the central lake disappeared and the environmental conditions were again dominated by an unstable fluvial system (Brown and Feibel, 1991). Also in the Omo-Turkana Basin and the adjacent Suguta Valley, late Pleistocene and early Holocene lakes are reported (e.g., Butzer et al., 1969; Casanova, 1986; Hillaire-Marcel et al., 1986; Sturchio et al., 1993). Recent work on the upper Chemeron Formation (3.5 to 1.7 Ma) exposed in the Tugen Hills (western Baringo-Bogoria Basin) revealed a sequence of five major diatomites (Deino et al., 2006). Single-crystal 40Ar/39Ar age estimates show the diatomites occur between 2.66 and 2.56 Ma at w20-kyr intervals (Deino et al., 2006). Detailed mapping and stratigraphic analysis suggest that these cycles that correlate with precessional periodicity are not due to local tectonic movements but reflect climate changes (Deino et al., 2006). First investigations of the diatom flora contained in these diatomites suggest that they are dominated by planktonic Aulacoseira spp. and large Stephanodiscus spp., and indicate the existence of large deep lakes between 2.66 and 2.56 Ma (Owen, 2002). The Kariandusi site in the Nakuru-Elmenteita Basin contains a 31-m-thick sequence of well-stratified diatomites and tuffs, resting on 0.987  0.003 Ma trachytes (sample KAR94-3: McCall et al., 1967; Figs. 1 and 2). The diatomite assemblage at Kariandusi documents the full history of a large deep lake. The lowermost units of the diatomite contain a rich benthic-epiphytic flora replaced by a community dominated by only two genera, Stephanodiscus spp. and Aulacoseira spp., indicating a large deep lake. The flora contained in the higher parts of the section document a lake regression coupled with higher alkalinity. The higher diatomites are overlain by a 0.977  0.010 Ma gray ash (sample KAR02-2) and a tuff layer dated at 0.946 Ma (Evernden and Curtis, 1965). In contrast to the several hundreds-of-meters-deep paleo-Lake Kariandusi, the late Pleistocene (w135,000 year) and early Holocene (w9,000 to 7,000 year) lakes in the Naivasha and Elmenteita-Nakuru Basin only reached 100 to 150 m water depth (Richardson and Dussinger, 1986; Trauth et al., 2001, 2003; Fig. 4). These lakes were characterized by the deposition of diatomites up to 4 m thick, containing a mixed planktonic and benthic-epiphytic diatom community dominated by species such as Aulacoseira granulata and A. goetzeana, but also characterized by the occurrence of benthic-epiphytic species such as Epithemia adnata, E. sorex, E. zebra, Fragilaria construens, F. brevistriata, shallow-water indicators such as Gomphonema angustatum and G. gracile, and indicators of slightly alkaline conditions such as Thalassiosira faurii, Mastogloia smithi, M. elliptica, and Anomoenoneis sphaerophora (Bergner and Trauth, 2004). Additional evidence for the paleowater depths of these young lakes comes from the altitude of paleoshorelines and sediment outcrops (Washbourn-Kamau, 1970; Richardson and Richardson, 1972; Washbourn-Kamau, 1975, 1977). Further to the south in the Naivasha Basin and adjacent areas, the oldest lacustrine sequences occur on the Kinangop Plateau in the Valley of Turasha River (Figs. 1 and 2). A w90-m-thick diatomite sequence overlies w4.6 Ma old trachytic lava flows (Strecker, 1991; Boven, 1992). The

sediments are intercalated by tuff layers, 4.702  0.006 Ma and 4.324  0.005 Ma (samples TRSH94-1 and TRSH94-2) and overlain by the 130-m-thick Kinangop tuffs, between 4.5 and 3.4 Ma (Clarke et al., 1990). The diatom assemblages dominated by Fragilaria spp. (predominantly F. construens and F. brevistriata), Surirella spp., Synedra ulna, Epithemia spp. (such as E. zebra), Cocconeis placentula, Cymbella spp., Rhapolodia spp., Aulacoseira spp., and few large Nitzschia spp. reflect shallow freshwater conditions. Except for late Pleistocene and early Holocene lake sediments, no older lacustrine sequences are exposed in the Naivasha Basin (Trauth et al., 2001, 2003). However, it remains possible that the old lakes south and north also expanded into this basin; if this is the case, the deposits are buried by younger volcaniclastic deposits before the magmatic complexes separated the Naivasha from the Gicheru and Nakuru-Elmenteita Basin during the last 0.5 Ma (Strecker et al., 1990). A 29-m-thick lake sediment sequence occurs at Munyu wa Gicheru, which is located in a small 1.96 Ma graben along the eastern downfaulted flank of the southern Kenya Rift (Ngecu and Njue, 1999; Figs. 1e3). The sediment section of diatomite and intercalated tuffs overlying 1.96 Ma trachytes was intruded by 1.65 to 1.75 Ma hyaloclastic rhyolites (Baker et al., 1988; sample MUN-2). The diatom flora in this basin documents a rapid transgression of a large deep lake, as indicated by a benthicepiphytic community in the lowermost parts of the sequence replaced by an assemblage dominated by large planktonic species, such as Stephanodiscus niagarae and Aulacoseira granulata. Above the hyaloclastic rhyolite, the diatomite contains increasing amounts of clastic material and benthic diatoms, phytoliths, and sponge spicules. The upper half of the diatomite is intercalated by a prominent pumiceous horizon, 1.65  0.01 Ma (sample MUN-1). The units are eroded on top and covered by phonolites, 0.724  0.003 Ma (sample MUN-3). An 80-m-thick sequence of diatomaceous lacustrine, volcaniclastic, and alluvial sedimentary deposits exposed in the southern Kenya Rift record the past million years of the history in the Olorgesailie Basin (Behrensmeyer et al., 2002). The Olorgesailie Formation records the formation of a closed-basin environment shortly before 0.992 Ma, alternating between lacustrine and subaerial conditions for a period of w500 kyr with no episodes of major erosion (Behrensmeyer et al., 2002). The most important lake period occurred between 0.992 and 0.974 Ma, as documented by the deposition of the main diatomite bed (e.g., Potts, 1998, 1999; Behrensmeyer et al., 2002). Further to the south, Lake Natron has one persistent lacustrine interval (a member of the Moinik Formation called the Moinik Clays) dated to ca.1.1e1.0 Ma that could correspond to the wet period recorded in the Olorgesailie Basin (Deino, pers. obs.). After 0.5 Ma, the Olorgesaillie Basin experienced phases of valley cutting associated with base-level fluctuations of around 50 m over periods of 10 to 100 kyrs (Behrensmeyer et al., 2002). The southern extension of the Kenya Rift, Olduvai Gorge, Tanzania, exposes a two-million-year sedimentary record in an incised river valley draining eastward from the Serengeti Plains (Hay, 1976; Ashley and Hay, 2002). The w100 m-thick

M.H. Trauth et al. / Journal of Human Evolution 53 (2007) 475e486

sequence of dominantly volcaniclastic sediments was deposited in a 50-km-wide closed basin containing a playa lake (Hay, 1976). A detailed 40Ar/39Ar chronology (Walter et al., 1991) and magnetostratigraphic data (e.g., Tamrat et al., 1995) combined with detailed environmental reconstructions (e.g., Blumenschine et al., 2003) suggest that the lake existed for a relatively short period of timedbetween 1.92 to 1.7 Ma, according to Hay and Kyser (2001), or between 1.84 to 1.74 Ma, according to Ashley and Hay (2002). In conclusion, these synchronous changes in moisture balance inferred from lake sediments contrast with the predominantly north-south volcano-tectonic history and can, therefore, be attributed to regional climate change. It therefore appears that in East Africa there were three major late Cenozoic lake periods, which occurred at 2.7e2.5 Ma, 1.9e1.7 Ma, and 1.1e0.9 Ma before present, suggesting consistency in the moisture history of the Kenyan and Ethiopian Rifts (Fig. 5). With the exception of the ‘precessional-like’ Baringo lacustrine sequence at 2.7e2.5 Ma (Kingston et al., 2007), we cannot conclude at present whether the large lakes in the eastern rifts were characterized by relatively stable lacustrine conditions for a longer period of time (i.e., up to 100 kyrs in total) or if these lakes fluctuated on shorter orbital or suborbital timescales. In contrast, late Pleistocene and Holocene lakes were not as deep as the earlier freshwater bodies in the rift. Moreover, these lake periods were generally interrupted by periods of lower lake levels and higher alkalinity (Trauth et al., 2003), which may suggest a different relationship between their occurrence and the climatic forcing. Controls on modern and past climates of East Africa Late Cenozoic tectonic activity in the EARS led to the production of isolated basins within which lakes could form (Fig. 5). Southward propagation of rifting and magmatic activity resulted in formation of lake basins first in the northern parts of the EARS. For example, fluviolacustrine history of the Afar, Omo-Turkana, and Baringo-Bogoria Basins in the north began in the middle and upper Miocene; whereas the oldest lacustrine sequences in the central and southern segments of the rift in Kenya and Tanzania are found in the early Pliocene (Fig. 6; Tiercelin and Lezzar, 2002). In general, paleolakes first appear in the EARS earlier in the north than in the south, due to the progressive formation of separate basins. If tectonics were the sole control over lake formation, then either a north to south or northwest to southeast temporal trend would be expected. However, what is observed is the appearance of large, deep lakes synchronously across large geographical areas at specific points in time (Trauth et al., 2005), suggesting that regional climatic control is operative. Carbon isotope records from both soil carbonates (Levin et al., 2004; Wynn, 2004) and biomarkers (n-alkanes) extracted from deep-sea sediments (Feakins et al., 2005) provide clear evidence a progressive vegetation shift from C3 (wtrees and shrubs) to C4 (wtropical grasses) plants during the PlioPleistocene. This shift has been ascribed to increased aridity that arose from the progressive rifting of East Africa

481

(deMenocal, 2004). Superimposed on this regime of subdued moisture availability, three periods characterized by the occurrence of large and deep lakes have been identified broadly in East Africa at 2.7e2.5 Ma, 1.9e1.7 Ma, and 1.1e0.9 Ma (Trauth et al., 2005), indicating consistency in the moisture history of the Kenyan and Ethiopian Rifts. Though preservation of East African lake records prior to 2.7 Ma is patchy, there is limited evidence for lake phases at w4.7e4.3 Ma, 4.0e3.9 Ma, w3.4e3.3 Ma, and w3.20e2.95 Ma (Fig. 1). The lake phases correspond to drops in the East Mediterranean marine dust abundance (Larrasoa~na et al., 2003), which are thought to reflect the aridity of the eastern Algerian, Libyan, and western Egyptian lowlands located north of the central Saharan watershed (Fig. 6). The lake phases also correspond to an increased occurrence of sapropels in the Mediterranean Sea, which are thought to be caused by increased Nile River discharge (Lourens et al., 2004). The correspondence of the Mediterranean marine records with lake records of East Africa suggest a consistent moisture record for a region encompassing much of central and northern Africa over the last three to five million years. In contrast, these East African wet phases correlate with significant intermediate-term increases in the dust records from ocean sediment cores adjacent to West Africa and Arabia (deMenocal, 1995, 2004). While this seems contradictory at first, examination of these data in greater chronologic detail demonstrates that both the lake and dust records are responding to precessional forcing and that these records are in-phase. Deino et al. (2006) and Kingston et al. (2007) found that the major lacustrine episode of the Baringo Basin between 2.7e 2.55 Ma actually consisted of five paleolake phases separated by a precessional cyclicity of 23 kyrs. These occurrences are in-phase with increased freshwater discharge and thus sapropel formation in the Mediterranean Sea (Lourens et al., 2004) and out of phase with the dust records from the Indian Ocean (deMenocal, 1995, 2004). Hence, the lake records from East Africa and the Indian Ocean dust records document extreme climate variability with precession-forced wet and dry phases. Precessional forcing of vegetation change also occurred at this time in Southwest Africa, independent of glacial-interglacial cycles (Denison et al., 2005). There is also emerging evidence for precessional forcing of the 1.9e1.7 Ma lake phase in the KBS Member of the Koobi Fora Formation in the northeast Turkana Basin in Kenya (Lepre et al., 2007). During the same period an oxygen isotope record from the Buffalo Cave flowstone (Makapansgat Valley, Limpopo Province, South Africa) shows clear evidence of precessionally forced changes in rainfall (Hopley et al., 2007). There is a growing body of evidence for precessional forcing of moisture availability in the tropics, both in East Africa during the Pliocene (deMenocal, 1995, 2004; Deino et al., 2006; Kingston et al., 2007; Lepre et al., 2007; Hopley et al., 2007 and elsewhere in the tropics during the Pleistocene (Bush et al., 2002; Trauth et al., 2003; Wang et al., 2004; Cruz et al., 2005). The precessional control of tropical moisture has also been clearly illustrated by the climate modelling of Clement et al. (2004), which showed that a 180 shift in

482

M.H. Trauth et al. / Journal of Human Evolution 53 (2007) 475e486

Fig. 6. Comparison of eccentricity variations (Berger and Loutre, 1991) with high-latitude climate transitions (Cowan, 2001; St. John and Krissek, 2002), Mediterranean dust flux (Larrasoa~ na et al., 2003), soil carbonate carbon isotopes (Levin et al., 2004; Wynn, 2004), East African lake occurrence (Trauth et al., 2005, this study), and hominin evolution (Bromage et al., 1995; Kimbel et al., 1996; Reed, 1997; Dunsworth and Walker, 2002; McHenry, 2002; White, 2002; Anto´n and Swisher, 2004; White et al., 2006).

precession could change annual precipitation in the tropics by at least 180 mm/year and cause a significant shift in seasonality. Support for increased seasonality during these extreme periods of climate variability also come from mammalian community structures (Reed, 1997; Bobe and Eck, 2001; Reed and Fish, in press) and hominin paleodiet reconstructions (Teaford and Ungar, 2000). The late Cenozoic periods of extreme climate variability appear to correlate with maxima in the 400-kyr component of the Earth’s eccentricity cycle. Prior to 2.7 Ma the wet phases appear every 400 kyrs (see Figs. 5 and 6). After 2.7 Ma, however, the wet phases appear every 800 kyrs, with periods of precessionally forced extreme climate variability at 2.7e2.5 Ma, 1.9e1.7 Ma, and 1.1e0.9 Ma before present; whereas other periods of eccentricity maxima at w2.2 Ma, w1.4 Ma, and w0.6 Ma are not associated with the alternating formation of large lakes or increased dust. The three late Cenozoic lake phases do, however, correlate with significant

global climatic transitions as well as peaks in eccentricity. Hence, after 2.7 Ma, global climate changes seem to be required to cause an increased regional climate sensitivity to precessionally forced insolation and increased seasonality, which allows either large deep lakes to develop or causes extreme aridity and large dust loads in the adjacent oceans. In contrast, prior to the 2.7 Ma eccentricity maxima alone were sufficient to produce regional sensitivity. It remains to be evaluated whether the long-term drying trend in East Africa or the global cooling trend is responsible for this shift from a simple linear response to long-term eccentricity forcing. The last three major Plio-Pleistocene lake phases all correspond to global climate transitions (Fig. 6). The lake phase at 2.7e2.5 Ma corresponds to the onset of the Northern Hemisphere Glaciation (Haug and Tiedemann, 1998), 1.9e1.7 Ma to a significant intensification and shift in the Walker Circulation (Ravelo et al., 2004), and 1.1e0.9 Ma to initiation of the Mid-Pleistocene Revolution (Berger and Jansen, 1994). Each

M.H. Trauth et al. / Journal of Human Evolution 53 (2007) 475e486

of these global climate transitions was accompanied by reduced North Atlantic Deep Water (NADW) formation (Haug and Tiedemann, 1998) and increased ice rafting from both Greenland and Antarctica (Cowan, 2001; St. John and Krissek, 2002). Ice expansion in both hemispheres would have significantly increased the Pole-Equator thermal gradient, leading to north-south compression of the Intertropical Convergence Zone (ITCZ). A similar effect occurred during the Last Glacial Maximum during which a strong compression of the ITCZ is observed both in paleoreconstructions of tropical hydrology (e.g., Peterson et al., 2000; Chiang et al., 2003; Wang et al., 2004) and via climate modelling (Lautenschlager and Herterich, 1990; Bush and Philander, 1999; Bush, 2001). Compression of the ITCZ is an essential component to increasing the sensitivity of East Africa to precessional forcing of moisture availability; without it moisture is transported north and south away from the Rift Valley. Along the whole length of the rift, without this high-latitude climate control, East Africa cannot receive enough rainfall to fill large deep freshwater lakes during positive precessional periods. Hence, after 3 Ma both global climate forcing and eccentricity maxima are required to generate episodes of extreme precessionally forced climate. On timescales of more than 100 kyr, rift-related volcanotectonic processes shaped the landscape of East Africa and profoundly influenced local climate and surface hydrology through the development of relief. Through uplift of the Kenyan and Ethiopian plateaus, changes in orography and associated rain shadow are believed to be the major driving force for increased variability of moisture availability throughout Eastern Africa. This increased sensitivity has resulted in a modern Rift Valley that hydrological modelling suggests could support lakes as deep as 150 m with an annual precipitation increase of only 15e30% (Bergner et al., 2003). Prior to the onset of the Northern Hemisphere Glaciation there is a linear relationship between long-term eccentricity variations and the development of deep freshwater lakes in the East African Rift. From the onset of the Northern Hemisphere Glaciation onwards, global climate transitions, which resulted in an increased Pole-Equator gradient and compression of both northern and southern boundaries of the ITCZ, were required to make East African moisture availability sensitive to maxima in eccentricity and thus changes in precession. The alternating extreme wet and dry periods would have had a profound influence on the climate and vegetation of East Africa. The sinusoidal precessional forcing at the equator consists of periods of less than 2,000 years, during which 60% of total variation in daily insolation and seasonality occurs. These are followed by w8,000-year intervals when relatively little change in daily insolation occurred (Maslin et al., 2005). Hence, instead of precession being a smooth forcing, it combines rapid strong forcing with long periods of relatively weak forcing. Rapid stratigraphic transitions from deep lacustrine to fluvial deposition associated with the diatomite lake deposits of the Pliocene lake deposits in the Baringo Basin suggest that this sinusoidal precessional forcing caused lakes to appear rapidly, remain part of the landscape for thousands of years, and then disappear rapidly (Deino et al., 2006;

483

Kingston et al., 2007). In fact, the absence of shallow-water diatom species at key Plio-Pleistocene lake deposits (Deino et al., 2006; Kingston et al., 2007) suggests that lakes could have dried up in less than 500 years. This rapid drying has important implications for the speciation and dispersal of mammals and hominins in East Africa. Fig. 6 shows that between 0.5 Ma and 5.0 Ma the periods of highly variable climate, (oscillating from very dry to very wet) are less than a third of the total time. In contrast, 12 of the 15 hominin species (w80%) first appeared in one of these extremely variable ‘wet-dry’ periods. Even taking into the account the great difficulty in dating the first appearance of African hominins and the problem of pseudospeciation events, this is compelling evidence for the preferential evolution of hominins during extreme climate periods. We suggest that ephemeral lakes, expanding and contracting on precessional timescales, would have evoked the widespread, regional-scale, rapid, and extreme environmental variability, required by the Variability Selection Hypothesis of human evolution (Potts, 1998). However, we also note that these extreme ‘wet-dry’ periods also would have provided prolonged periods of either extremely abundant or scarce water and food resources, which may have driven evolution. The interplay of local tectonic development, global climatic transitions, and local orbital forcing were required to produce the highly variable East African environment in which our ancestors evolved. Acknowledgements This project was funded by two grants to M.T. and M.S. by the German Research Foundation (DFG) and a UCL Graduate School grant to M.M. We thank the Government of Kenya and the Kenya Wildlife Service for research permits and support. The authors thank L. Aiello, P. Blisniuk, M. Collard, P. deMenocal, F. Gasse, F. Grine, D. Kuhn, A. Lister, and G. Muchemi for inspiring discussions. We also thank S. Higgins, S. Kabingu, T. Schlu¨ter, and M. Ibs-von Seht for logistical support. References Anto´n, S.C., Swisher, C.C., 2004. Early dispersals of Homo from Africa. Ann. Rev. Anthropol. 33, 271e296. Asfaw, B., Beyene, Y., Suwa, G., Walter, R.C., White, T.D., WoldeGabriel, G., Yemane, T., 1992. The earliest Acheulean from Konso-Gardula. Nature 360, 732e734. Ashley, G.M., Hay, R.L., 2002. Sedimentation patterns in a Plio-Pleistocene volcaniclastic rift-platform basin, Olduvai Gorge, Tanzania. In: Renaut, R.W., Ashley, G.M. (Eds.), Sedimentation in Continental Rifts. SEPM Spec. Publ. 73, 107e122. Bagdasaryan, G.P., Gerasimovskiy, V.I., Polyakov, A.I., Gukasyan, R.K., 1973. Age of volcanic rocks in the rift zones of East Africa. Geochem. Int. 10, 66e71. Baker, B.H., Mitchell, J.G., Williams, L.A.J., 1988. Stratigraphy, geochronology and volcano-tectonic evolution of the Kedong-Naivasha-Kinangop region, Gregory Rift Valley, Kenya. J. Geol. Soc. (Lond.) 145, 107e116. Baker, B.H., Mohr, P.A., Williams, L.A.J., 1972. Geology of the eastern rift system of Africa. Geol. Soc. Am. Spec. Pap. 136. Barker, P.A., Talbot, M.R., Street-Perrott, F.A., Marret, F., Scourse, J.D., Odada, E., 2004. Late Quaternary climatic variability in intertropical Africa. In: Battarbee, R.W., Gasse, F., Stickley, C.E. (Eds.), Past Climate

484

M.H. Trauth et al. / Journal of Human Evolution 53 (2007) 475e486

Variability through Europe and Africa. Developments in Paleoenvironmental Research. Kluwer Academic Publishers, Dordrecht, pp. 117e138. Behrensmeyer, A.K., Potts, R., Deino, A., Ditchfield, P., 2002. Olorgesailie, Kenya: a million years in the lift of a rift basin. In: Sedimentation in Continental Rifts. SEPM Spec. Pub. 73, pp. 97e106. Berger, A., Loutre, M.F., 1991. Insolation values for the climate of the last 10 million years. Quaternary Science Reviews 10, 297e317. Berger, W.H., Jansen, E., 1994. Mid-Pleistocene climate shift: the Nansen connection. In: Johannessen, O.M., Muench, R.D., Overland, J.E. (Eds.), The Polar Oceans and their Role in Shaping the Global Environment. Geophys. Monogr. 85, 295e311. Bergner, A.G.N., Trauth, M.H., 2004. Comparison of the hydrologic and hydrochemical evolution of Lake Naivasha (Kenya) during three highstands between 175 and 60 kyr BP. Palaeogeogr. Palaeoclimatol. Palaeoecol. 125, 17e36. Bergner, A.G.N., Trauth, M.H., Bookhagen, B., 2003. Magnitude of precipitation/evaporation changes in the Naivasha Basin (Kenya) during the last 150 kyrs. Global Planet. Change 36, 117e136. Blisniuk, P., Strecker, M.R., 1990. Asymmetric rift-basin development in the central Kenya Rift. Terra Abstr. 2, 51. Blumenschine, R.J., Peters, C.R., Masao, F.T., Clarke, R.J., Deino, A.L., Hay, R.L., Swisher, C.C., Stanistreet, I.G., Ashley, G.M., McHenry, L.J., Sikes, N.E., van der Merwe, N.J., Tactikos, J.C., Cushing, A.E., Deocampo, D.M., Njau, J.K., Ebert, J.I., 2003. Late Pliocene Homo and hominid land use from western Olduvai Gorge, Tanzania. Science 299, 1217e1221. Bobe, R., Eck, G.G., 2001. Response of African bovids to Pliocene climatic change. Paleobiology 27, 1e47. Bosworth, W., Strecker, M.R., 1997. Stress field changes in the Afro-Arabian Rift System during the Miocene to Recent Period. Tectonophysics 278, 47e62. Boven, A., 1992. The applicability of the K-Ar and 40Ar/39Ar techniques to Quaternary volcanics: a fundamental and experimental investigation. PhD Thesis, Vrije Universiteit Brussel. Bromage, T.G., Schrenk, F., Zonnefeld, F.W., 1995. Paleoanthropology of the Malawi Rift: an early hominid mandible from the Chiwondo Beds, northern Malawi. J. Hum. Evol. 28, 71e108. Brown, F.H., Feibel, C.S., 1991. Stratigraphy, depositional environments and palaeogeography of the Koobi Fora Formation. In: Harris, J.M. (Ed.), Koobi Fora Research Project, vol. 3. Clarendon, Oxford, pp. 1e30. Bush, A.B.G., 2001. Simulating climates of the Last Glacial Maximum and of the mid-Holocene: wind changes, atmosphere-ocean interactions, and the tropical thermocline. In: Seidov, D., Haupt, B.J., Maslin, M. (Eds.), The Oceans and Rapid Climate Change: Past, Present, and Future. AGU, Washington, D.C., pp. 135e144. Bush, A.B.G., Philander, S.G.H., 1999. The climate of the Last Glacial Maximum: results from a coupled atmosphere-ocean general circulation model. Geophys. Res. 104, 24509e24525. Bush, M.B., Moreno, E., de Oliveira, P.E., Colinvaux, P.A., 2002. Orbitalforcing signal in sediments of two Amazonian lakes. J. Paleolim. 27, 341e352. Butzer, K.W., Brown, F.W., Thruber, D.L., 1969. Horizontal sediments of the lower Omo valley; the Kibish formation. Quaternaria 11, 15e29. Casanova, J., 1986. Les stromatolites continentaux: paleoecologie, paleohydrologie, paleoclimatologie. Application au Rift Gregory. The´se, Universite´ d’Aix-Marseille II. Chiang, J.C.H., Biasutti, M., Battisti, D.S., 2003. Sensitivity of the Atlantic Intertropical Convergence Zone to Last Glacial Maximum boundary conditions. Paleoceanography 18 (4). Clarke, M.C.G., Woodhall, D.G., Allen, D., Darling, G., 1990. Geological, Volcanological and Hydrogeological Controls on the Occurrence of Geothermal Activity in the Area Surrounding Lake Naivasha, Kenya. Ministry of Energy, Nairobi, Kenya. Clement, A.C., Hall, A., Broccoli, A.J., 2004. The importance of precessional signals in the tropical climate. Clim. Dyn. 22, 327e341. Cowan, E.A., 2001. Identification of the glacial signal from the Antarctic Peninsula since 3.0 Ma at Site 1011 in a continental rise sediment drift. In: Barker, P.F., Camerlenghi, A., Acton, G.D., Ramsay, A.T.S. (Eds.),

Proc. ODP Sci. Res. 178. Available online at: http://www-odp.tamu.edu/ publications/178_SR/chap_10/chap_10.htm Crossley, R., 1979. The Cenozoic stratigraphy and structure of the western part of the Rift Valley in southern Kenya. J. Geol. Soc. (Lond.) 136, 393e405. Crossley, R., Knight, R.M., 1981. Volcanism in the western part of the rift valley in southern. Kenya. Bull. Volcanol. 44, 117e128. Cruz, F.W., Burns, S.J., Karmann, I., Sharp, W.D., Vuille, M., Cardoso, A.O., Ferrari, J.A., Silva Dias, P.L., Viana, O., 2005. Insolation-driven changes in atmospheric circulation over the past 116,000 years in subtropical Brazil. Nature 434, 63e66. Davidson, A., Rex, D., 1980. Age of volcanism and rifting in southwestern Ethiopia. Nature 283, 657e659. Dawson, J.B., 1992. Neogene tectonics and volcanicity in the North Tanzania sector of the Gregory Rift Valley: contrasts with the Kenyan sector. Tectonophysics 204, 81e92. Deino, A.L., Kingston, J.D., Glen, J.M., Edgar, R.K., Hill, A., 2006. Precessional forcing of lacustrine sedimentation in the late Cenozoic Chemeron Basin, Central Kenya Rift, and calibration of the Gauss/Matuyama boundary. Earth Planet. Sci. Lett. 247, 41e60. deMenocal, P., 1995. Plio-Pleistocene African climate. Science 270, 53e59. deMenocal, P., 2004. African climate change and faunal evolution during the Pliocene-Pleistocene. Earth Planet. Sci. Lett. 220, 3e24. Denison, S.M., Maslin, M.A., Boot, C., Pancost, R.D., Ettwein, V.J., 2005. Precession-forced changes in south west African vegetation during marine isotope stages 101e100 (similar to 2.56e2.51 Ma). Palaeogeogr. Palaeoclimatol. Palaeoecol. 220, 275e386. Dunsworth, H., Walker, A., 2002. Early genus Homo. In: Hartwig, W.C. (Ed.), The primate fossil record. Cambridge University Press, New York, pp. 401e406. Ebinger, C.J., Rex, D.C., Yemane, T., Kelley, S., 1993. Volcanism and extension between the Main Ethiopian and Gregory rifts. In: Thorweihe, U., Schdelmeier, H. (Eds.), Geoscientific Research in Northeast Africa. Balkema, Rotterdam, pp. 301e304. Ebinger, C.J., Sleep, N.H., 1998. Cenozoic magmatism throughout east Africa resulting from impact of a single plume. Nature 385, 788e791. Ebinger, C.J., Yemane, T., Harding, D.J., Tesfaye, S., Kelley, S., Rex, D.C., 2000. Rift deflection, migration, and propagation: linkage of the Ethiopian and Eastern rifts, Africa. GSA Bull. 112, 163e176. Evernden, J.G., Curtis, G.H., 1965. The potassium-argon dating of late Cenozoic rocks in East Africa and Italy. Curr. Anthropol. 6, 177e189. Feakins, S.J., deMenocal, P.B., Eglinton, T.I., 2005. Biomarker records of late Neogene changes in northeast African vegetation. Geology 33, 977e980. Foster, A., Ebinger, C., Mbede, E., Rex, D., 1997. Tectonic development of the northern Tanzanian sector of the east African rift system. J. Geol. Soc. 154, 689e700. Gasse, F., 1977. Evolution of Lake Abhe´ (Ethiopia and TFAI), from 70,000 b.p. Nature 265, 42e45. Gasse, F., 1990. Tectonic and climatic controls on lake distribution and environments in Afar from Miocene to present. In: Katz, B.J. (Ed.), Lacustrine Basin Exploration, Case Studies, and Modern Analogs. AAPG Mem. 50, 1e34. George, R.M.M., Rogers, N.W., Kelley, S., 1998. Earliest magmatism in Ethiopia: evidence for two mantle plumes in one flood basalt province. Geology 26, 923e926. Haug, G.H., Tiedemann, R., 1998. Effect of the formation of the Isthmus of Panama on Atlantic Ocean thermohaline circulation. Nature 393, 673e675. Hay, R.L., 1976. Geology of the Olduvai Gorge. a Study of Sedimentation in a Semiarid Basin. University of California Press, Berkeley, Los Angeles, London. Hay, R.L., Kyser, T.K., 2001. Chemical sedimentology and paleoenvironmental history of Lake Olduvai, a Pliocene lake in northern Tanzania. GSA Bull. 113, 1505e1521. Hillaire-Marcel, C., Carro, O., Casanova, J., 1986. 14C and Th/U dating of Pleistocene and Holocene stromatolites from East African paleolakes. Quatern. Res. 25, 312e329. Hopley, P.J., Marshall, J.D., Weedon, G.P., Latham, A.G., Herries, A.I.R., Kuykendall, K.L., 2007. Orbital forcing and the spread of C4 grasses in

M.H. Trauth et al. / Journal of Human Evolution 53 (2007) 475e486 the late Neogene: stable isotope evidence from South African speleothems. J. Hum. Evol. 53, 620e634. Kimbel, W.H., Walter, R.C., Johanson, D.C., Reed, K.E., Aronson, J.L., Assefa, Z., Marean, C.W., Eck, G.G., Bobe, R., Hovers, E., Rak, Y., Vondra, C., Yemane, T., York, D., Chen, Y., Evensen, N.M., Smith, P.E., 1996. Late Pliocene Homo and Oldowan tools from the Hadar Formation (Kada Hadar Member), Ethiopia. J. Hum. Evol. 31, 549e561. Kingston, J., Deino, A., Hill, A., 2007. Astronomically forced climate change in the Kenyan Rift Valley 2.7e2.55 Ma: implications for the evolution of early hominin ecosystems. J. Hum. Evol. 53, 487e503. KRISP Working Party, 1991. Large scale variations in lithospheric structure along and across the Kenya Rift. Nature 354, 223. Larrasoa~ na, J.C., Roberts, A.P., Rohling, Winklhofer, M., Wehausen, R., 2003. Three million years of monsoon variability over the northern Sahara. Clim. Dyn. 21, 689e698. Lautenschlager, M., Herterich, K., 1990. Atmospheric response to ice age conditionsdclimatology near the earth’s surface. Geophys. Res. 95, 22547e22557. Leakey, M.G., Feibel, C.S., McDougall, I., Walker, A., 1995. New four-million-year-old hominid species from Kanapoi and Allia Bay, Kenya. Nature 376, 565e571. Leakey, M.G., Feibel, C.S., McDougall, I., Ward, C., Walker, A., 1998. New specimens and confirmation of an early age for Australopithecus anamensis. Nature 393, 62e66. Lepre, C.J., Quinn, R.L., Joordens, J.C.A., Swisher III, C.C., Feibel, C.S., 2007. Plio-Pleistocene facies environments from the KBS Member, Koobi Fora Formation: implications for climate controls on the development of lake-margin hominin habitats in the northeast Turkana Basin (northwest Kenya). J. Human Evol. 53, 504e514. Leroux, M., 2001. The Meteorology and Climate of Tropical Africa. SpringerPraxis Books in Environmental Sciences. Levin, N.E., Quade, J., Simpson, S.W., Semaw, S., Rogers, M., 2004. Isotopic evidence for Plio-Pleistocene environmental change at Gona, Ethiopia. Earth Planet. Sci. Lett. 219, 93e110. Lourens, L., Hilgen, F., Shackleton, N.J., Laskar, J., Wilson, D., 2004. The Neogene Period. In: Gradstein, F., Ogg, J.G., Smith, G. (Eds.), A Geologic Time Scale 2004. Cambridge University Press, pp. 409e440. Macdonald, R., Rogers, N.W., Fitton, J.G., Black, S., Smith, M., 2001. Plumelithosphere interactions in the generation of the basalts of the Kenya Rift, East Africa. J. Petrol. 42 (5), 877e900. Maslin, M.A., Mahli, Y., Phillips, O., Cowling, S., 2005. New views on an old forest: assessing the longevity, resilience and future of the Amazon Rainforest. Trans. Inst. Br. Geogr. 30 (4), 390e401. McCall, G.J.H., Baker, B.H., Walsh, J., 1967. Late Tertiary and Quaternary sediments of the Kenya Rift Valley. In: Bishop, W.W., Clark, J.D. (Eds.), Background to Evolution in Africa. University of Chicago Press, pp. 191e220. McDougall, I., 1985. K-Ar and 40Ar/39Ar dating of the hominid-bearing Pliocene-Pleistocene sequence at Koobi Fora, Lake Turkana, northern Kenya. GSA Bull. 96, 159e175. McDougall, I., Watkins, R.T., 1988. Potassium-argon ages of volcanic rocks from northeast of Lake Turkana, northern Kenya. Geol. Mag. 125, 15e23. McHenry, H.M., 2002. Introduction to the fossil record of human ancestry. In: Hartwig, W.C. (Ed.), The Primate Fossil Record. Cambridge University Press, New York, pp. 401e406. Mechie, J., Keller, G.R., Prodehl, C., Khan, M.A., Gaciri, S.J., 1997. A model for the structure, composition and evolution of the Kenya rift. Tectonophysics 278, 95e119. Morley, C.K., 1994. Interaction of deep and shallow precesses in the evolution of the Kenya rift. Tectonophysics 236, 81e91. Morley, C.K., Wescott, W.A., Stone, D.M., Harper, R.M., Wigger, S.T., Karanja, F.M., 1992. Tectonic evolution of the northern Kenya rift. J. Geol. Soc. (Lond.) 149, 333e348. Ngecu, W.M., Njue, J.N., 1999. Lithostratigraphy and distribution of Pleistocene sediments of the Munyu wa Gicheru Formation in South Kenya Rift Valley. J. Afr. Earth Sci. 29, 411e421. Owen, R.B., 2002. Sedimentological characteristics and origins of diatomaceous deposits in the East African Rift System. SEPM Spec. Publ. 73, 233e246.

485

Peterson, L.C., Haug, G.H., Hughen, K.A., Ro¨hl, U., 2000. Rapid changes in the hydrologic cycle of the tropical Atlantic during the Last Glacial. Science 290, 1947e1951. Potts, R., 1998. Environmental hypothesis of hominin evolution. Yearb. Phys. Anthropol. 41, 93e136. Potts, R., 1999. Paleolandscape variation and Early Pleistocene hominid activities: Members 1 and 7, Olorgesailie Formation, Kenya. J. Hum. Evol. 37, 747e788. Prodehl, C., Ritter, J.R.R., Mechie, J., Keller, G.R., Khan, M.A., Jacob, B., Fuchs, K., Nyambok, I.O., Obel, J.D., Riaroh, D., 1997. The KRISP 94 lithospheric investigation of southern Kenyadthe experiments and their main results. Tectonophysics 278, 121e147. Ravelo, C., Andreasen, D., Olivarez, A., Lyle, M., Wara, M., 2004. The role of tropical and extratropical processes on Northern Hemisphere Glaciation. Nature 429, 263e267. Reed, K.E., Fish, J.L. Tropical and temperate seasonal influences on human evolution, in Primate Seasonality and Human Evolution. In: Brockman, D., van Schaik, C. (Eds.), Primate Seasonality and Human Evolution. Cambridge University Press, Cambridge, in press. Reed, K.E., 1997. Early hominid evolution and ecological change through the African Plio-Pleistocene. J. Hum. Evol. 32, 289e322. Richardson, J.L., Dussinger, R.A., 1986. Paleolimnology of mid-elevation lakes in the Kenya Rift Valley. Hydrobiologia 143, 167e174. Richardson, J.L., Richardson, A.E., 1972. History of an African Rift lake and its climatic implications. Ecol. Monogr. 42, 499e534. Ruddiman, W.F., Raymo, M.E., 1988. Northern Hemisphere climate regimes during the past 3 Ma: possible tectonic connections. Phil. Trans. R. Soc. B 318, 411e430. Smith, 1994. Stratigraphic and structural constraints on mechanisms of active rifting in the Gregory Rift, Kenya. Tectonophysics 236, 3e22. St. John, K., Krissek, L., 2002. The late Miocene to Pleistocene icerafting history of southeast Greenland. Boreas 41, 28e35. Strecker, M.R., 1991. Das zentrale und su¨dliche Kenia-Rift unter besonderer Beru¨cksichtigung der neotektonischen Entwicklung. Habilitation Thesis, University of Karlsruhe. Strecker, M.R., Blisniuk, P.M., Eisbacher, G.H., 1990. Rotation of extension direction in the central Kenya Rift. Geology 18, 299e302. Sturchio, N.C., Dunkley, P.N., Smith, M., 1993. Climate-driven variations in geothermal activity in the northern Kenya rift valley. Nature 362, 233e234. Sundquist, E.T., Visser, K., 2004. The geologic history of the carbon cycle. In: Schlesinger, W.H. (Ed.), Treatise on Geochemistry . Biogeochemistry, vol. 8. Elsevier, Amsterdam, pp. 425e472. Tamrat, E., Thouveny, N., Taieb, M., Opdyke, N.D., 1995. Revised magnetostratigraphy of the Plio-Pleistocene sedimentary sequence of the Olduvai Formation (Tanzania). Palaeogeogr. Palaeoclimatol. Palaeoecol. 114, 273e283. Teaford, M.F., Ungar, P.S., 2000. Diet and the evolution of the earliest human ancestors. Proc. Natl. Acad. Sci. 97, 13506e13511. Tiercelin, J.J., 1986. The Pliocene Hadar Formation, Afar Depression, Ethiopia. In: Frostick, L.E., Renaut, R.W., Reid, I., Tiercelin, J.J. (Eds.), Sedimentation in the African Rifts. Geol. Soc. Lond. Spec. Publ. 25, 253e265. Tiercelin, J.J., Lezzar, K.E., 2002. A 300 million years history of rift lakes in Central and East Africa: an updated broad review. In: Odada, E.O., Olago, D.O. (Eds.), The East African Great Lakes: Limnology, Paleolimnology and Biodiversity. Kluwer, Dordrecht, pp. 3e60. Trauth, M.H., Deino, A., Bergner, A.G.N., Strecker, M.R., 2003. East African climate change and orbital forcing during the last 175 kyr BP. Earth Planet. Sci. Lett. 206, 297e313. Trauth, M.H., Deino, A., Strecker, M.R., 2001. Response of the East African climate to orbital forcing during the Last Interglacial (130e117 kyr BP) and the early Last Glacial (117e60 kyr BP). Geology 29 (6), 499e502. Trauth, M.H., Maslin, M.A., Deino, A., Strecker, M.R., 2005. Late Cenozoic moisture history of East Africa. Science 309, 2051e2053. Walter, R.C., Manega, P.C., Hay, R.L., Drake, R.E., Curtis, G.H., 1991. Laser-fusion 40 Ar/39Ar dating of Bed I, Olduvai Gorge, Tanzania. Nature 354, 145e194. Wang, X., Auler, A.S., Edwards, R.L., Cheng, H., Ito, E., Solheid, M., 2004. Wet periods in northeastern Brazil over the past 210 kyr linked to distant climate anomalies. Nature 432, 740e743. Washbourn-Kamau, C., 1970. Late Quaternary chronology of the NakuruElmenteita Basin, Kenya. Nature 226, 253e254.

486

M.H. Trauth et al. / Journal of Human Evolution 53 (2007) 475e486

Washbourn-Kamau, C.K., 1975. Late Quaternary shorelines of Lake Naivasha, Kenya. Azania X, 77e92. Washbourn-Kamau, C.K., 1977. The Ol Njorowa Gorge, Lake Naivasha Basin, Kenya. In: Greer, D.C. (Ed.), Desertic Terminal Lakes. Utah Water Research Laboratory, pp. 297e307. White, T., 2002. Earliest hominids. In: Hartwig, W.C. (Ed.), The Primate Fossil Record. Cambridge University Press, New York, pp. 407e417. White, T.D., WoldeGabriel, G., Asfaw, B., Ambrose, S., Beyene, Y., Bernor, R.L., Boisserie, J.-R., Currie, B., Gilbert, H., Haile-Selassie, Y., Hart, W.H., Hlusko, L.J., Howell, F.C., Kono, R.T., Lehmann, T., Louchart, A., Lovejoy, C.O., Renne, P.R., Saegusa, H., Vrba, E.S., Wesselman, H., Suwa, G., 2006. Asa Issie, Aramis and the origin of Australopithecus. Nature 440, 883e889. Williams, L.A.J., Macdonald, R., Leat, P.T., 1983. Magmatic and structural evolution of the central part of the Kenya Rift. In: Proceedings of Regional

Seminar on Geothermal Energy in Eastern and Southern Africa, UNESCO/ USAID, Nairobi, pp. 61e67. Williams, M.A.J., Williams, F.M., Gasse, F., Curtis, G.H., Adamson, D.A., 1979. Pliocene-Pleistocene environments at Gadeb prehistoric site, Ethiopia. Nature 282, 29e33. WoldeGabriel, G., Heiken, G., White, T.D., Asfaw, B., Hart, W.K., Renne, P.R., 2000. Volcanism, tectonism, sedimentation, and the paleoanthropological record in the Ethiopian Rift System. GSA Spec. Pap. 345, 83e99. Wynn, J.G., 2004. Influence of Plio-Pleistocene Aridification on Human Evolution: Evidence From Paleosols of the Turkana Basin, Kenya. Am. J. Phys. Anthropol. 123, 106e118. Yuretich, R.F., 1982. Possible influences upon lake development in the East African Rift Valleys. J. Geol. 90, 329e337.