Climate and Human Evolution

See allHide authors and affiliations

Science  04 Feb 2011:
Vol. 331, Issue 6017, pp. 540-542
DOI: 10.1126/science.1190683

Did climate change shape human evolution? This question has old, deep roots (1, 2), but in recent decades, the fossil record of hominin evolution and behavior has improved, although it remains incomplete, and great progress has been made in the quality and number of African paleoclimate records from land and ocean sediments (3). A recent National Research Council (NRC) report (4) examines emerging faunal and paleoclimate evidence underlying the hypothesis that past climate changes may have influenced our evolution.

The basic premise is that large-scale shifts in climate alter the ecological structure and resource availability of a given setting, which leads to selection pressures (3, 5). Indeed, some of the larger climate shifts in Earth history were accompanied by unusually high rates of faunal turnover—bursts of biotic extinction, speciation, and innovation (68). For example, a large turnover event occurred near 34 million years ago (Ma) when Earth cooled abruptly and large glaciers first expanded upon Antarctica (8, 9). Many of the taxa that appeared after 34 Ma were better adapted to the new environments that emerged, which included cooler polar regions, greater seasonality, and arid grasslands.

Notable hominin extinction, speciation, and behavioral events appear to be associated with changes in African climate in the past 5 million years. First appearance and extinction events, as well as key behavioral milestones, cluster between 2.9 and 2.6 Ma and again between 1.9 and 1.6 Ma (see the figure, panel A). In the earlier group, these events include the extinction of Australopithecus afarensis (“Lucy”) near 2.9 Ma; the emergence of the robust australopiths (Paranthropus spp.), with large jaws and grinding teeth, near 2.7 Ma; and the emergence of the larger-brained Homo lineage sometime after 2.6 Ma, near the time when the first evidence for Oldowan stone tool manufacture, use, and transport appears (10).

Important evolutionary developments between 1.9 and 1.6 Ma (4) included the first appearance of Homo erectus—the first hominin species to resemble modern humans, with large brains, similar dentition, and a lithe frame—near 1.9 Ma. By 1.6 Ma, the more developed and refined Acheulean stone tool industry appears, including bifacial hand axes. This period also includes the first hominin exodus out of Africa and into Europe and South Asia.

For any given time period, the number and diversity of hominin fossils is low relative to most other taxa. Paleontologists have turned to another mammalian group that shared the African landscape with our ancestors, the bovids, for evidence of climate influences. These even-toed ungulates, such as antelopes, represent roughly one-third of African fossils [hominin fossils constitute <1% (11)]. An all-Africa analysis of bovid evolution spanning the past 6 million years revealed several turnover events where rates of speciation and extinction were well above background levels (5). The two largest of these events occurred near 2.8 Ma and 1.8 Ma (5), and the nature of the faunal changes implicates aridification and grassland expansion as a likely cause; for example, many new grazing bovid species appeared with specialized dentition (hypsodont molars) for processing the abrasive, grassy diet (12).

African climate changes during the past 5 million years bear the signatures of two separate processes (13). Orbital precession forcing (with a period of ∼20,000 years) acted as a “monsoonal pacemaker” that switched between wet and dry conditions. A long-term trend toward increasing drier and more variable conditions is superimposed on these wet-dry cycles, commencing after ∼3 Ma and peaking near 1.8 to 1.6 Ma.

A snapshot of African evolutionary and paleoclimate changes.

(A) Summary diagram of human evolution spanning the past 4.6 million years [no phylogenetic relations are indicated; compiled from (4, 10)]. First appearances and approximate durations of Mode 1 (Oldowan) and Mode 2 (Acheulean) stone tools are indicated (11). (B) Occurrences of Mediterranean sapropel deposits compiled from marine and land sediment sequences (15). (C) Compilation of sedimentary evidence indicating deep lake conditions recorded in several East African paleolake basins (16, 17). (D) Carbon isotopic analyses of plant-wax biomarker compounds measured at Site 231 in the Gulf of Aden, currently the most proximal ocean drilling site to hominin fossil localities (18). The shift to higher values after 3 Ma indicates a greater proportions of C4 vegetation, or savannah grasslands. (E) Carbon isotopic values of soil carbonate nodules compiled from several studies (19, 20), also indicating grassland expansion after ∼3 Ma, peaking between 1.8 and 1.6 Ma. (F) Relative abundance of African mammals indicative of seasonally arid grasslands in the lower Omo Valley (Ethiopia), showing an initial increase in grassland-adapted mammals after 2.5 Ma with peak values after 1.8 Ma (12).


The African wet-dry cycles are impressive: From 15,000 to 5000 years ago, the modern Saharan Desert was nearly completely vegetated, with large, permanent lakes and abundant fauna (14). Precessional increases in summer radiation invigorated the monsoon, delivering more rainfall deeper into Africa, and enhanced Nile river runoff flooded into the eastern Mediterranean Sea. The resulting freshwater stratification created anoxic conditions and led to deposition of organic-rich sediments (sapropels) on the seafloor. Stratigraphic sections representing many millions of years contain hundreds of these sapropel layers (15), and these layers are commonly bundled into 100,000- and 412,000-year packages associated with the modulation of orbital precession monsoon cycles with the eccentricity of Earth's orbit (see the figure, panel B). Many East African rift valley lakes in Ethiopia, Kenya, and Tanzania were high during a few, but not all, of these high-eccentricity intervals (see the figure, panel C) (16, 17).

The long-term drying trend is documented by increases in African wind-borne dust after 2.8 Ma, with peak values near 1.8 to 1.6 Ma off East Africa (13). Carbon isotopic analyses of plant wax biomarker compounds from a drill site in the Gulf of Aden (18) (see the figure, panel D) and analyses of soil carbonate nodules near hominin fossil localities (19, 20) (see the figure, panel E) indicate that East African savannah grasslands expanded at these times. Increased numbers of grazing bovid species parallel the grassland expansion (see the figure, panel F) (12).

Hypotheses linking African climate and faunal change are constrained by these new observations. Faunal lineages typically persisted throughout dozens of wet-dry climate cycles, so it is unlikely that the orbital-scale variability alone was a selection agent. Similarly, early hypotheses emphasizing only the unidirectional development of open vegetation do not capture the now-evident complexity of African climate variability. An emerging view is that African fauna, including our forebears, may have been shaped by changes in climate variability itself. These views posit that increasing climate variability led to climate and ecological shifts that were progressively larger in amplitude (3, 21, 22).

The evolutionary signatures of these climate changes are recorded by the appearance of new traits, such as the increase in larger, more hypsodont grassland bovid species, after 3 Ma and especially after 1.8 Ma (5, 11, 12). For hominins, these environmental transition periods were coincident with fundamental speciation, extinction, and morphological and behavioral milestones that ultimately produced those traits that define us as human and distinguish us from other primates (4, 5, 21). On the basis of an analysis of coexisting East African hominin and bovid fossil assemblages, Reed (23) concluded that “Homo species appear the first to be adapted to open, arid environments.”

One strategy articulated in the NRC report (4) is to investigate the key evolutionary milestone events as natural history experiments. The grand challenge will be to develop coordinated sets of observations to test proposed links between African climate and faunal change. The foremost task will be to improve the fossil and paleoclimate records, especially for those intervals where available evidence is most suggestive of climatic forcing of adaptive evolutinary change.


Navigate This Article