Early anthropogenic impact on Western Central SEE COMMENTARY African rainforests 2,600 y ago Yannick Garcina,1, Pierre Deschampsb, Guillemette Menot´ c, Geoffroy de Saulieud, Enno Schefuße, David Sebagf,g,h, Lydie M. Duponte, Richard Oslislyd,i, Brian Brademannj, Kevin G. Mbusnumk, Jean-Michel Onanal,m, Andrew A. Akon, Laura S. Eppo, Rik Tjallingiij, Manfred R. Streckera, Achim Brauerj, and Dirk Sachsep aInstitute of Earth and Environmental Science, University of Potsdam, 14476 Potsdam, Germany; bAix-Marseille Universite,´ CNRS, IRD, College` de France, Centre Europeen´ de Recherche et d’Enseignement des Geosciences´ de l’Environnement UM34, 13545 Aix-en-Provence, France; cUniv Lyon, Ens de Lyon, Universite´ Lyon 1, CNRS, UMR 5276 LGL-TPE, 69342 Lyon, France; dPatrimoines Locaux et Gouvernance UMR 208, IRD, MNHN, 75005 Paris, France; eMARUM–Center for Marine Environmental Sciences, University of Bremen, 28359 Bremen, Germany; fNormandie Universite,´ UNIROUEN, UNICAEN, CNRS, M2C, 76000 Rouen, France; gHSM, LMI Picass’Eau, IRD, Universite´ de Montpellier, 34095 Montpellier, France; hInstitute of Earth Surface Dynamics, Geopolis, University of Lausanne, 1015 Lausanne, Switzerland; iAgence Nationale des Parcs Nationaux, 20379 Libreville, Gabon; jSection 5.2, Climate Dynamics and Landscape Evolution, GFZ–German Research Centre for Geosciences, 14473 Potsdam, Germany; kLaboratoire de Chimie de l’Environnement FRE 3416, Aix-Marseille Universite,´ CNRS, 13545 Aix-en-Provence, France; lDepartment of Plant Biology, Faculty of Sciences, University of Yaounde´ I, Yaounde,´ Cameroon; mHerbier National du Cameroun, Institut de Recherche Agricole pour le Developpement,´ Yaounde,´ Cameroon; nInstitute of Geological and Mining Research, Yaounde,´ Cameroon; oAlfred Wegener Institute, Helmholtz Centre for Polar and Marine Research, 14473 Potsdam, Germany; and pSection 5.1, Geomorphology, GFZ–German Research Centre for Geosciences, 14473 Potsdam, Germany Edited by Sarah J. Feakins, University of Southern California, Los Angeles, CA, and accepted by Editorial Board Member Ruth S. DeFries January 24, 2018 (received for review August 30, 2017) A potential human footprint on Western Central African rain- turbed/secondary forest, with a significant proportion of grasses forests before the Common Era has become the focus of an ongo- and pioneer trees (3). This event was attributed to aridification ing controversy. Between 3,000 y ago and 2,000 y ago, regional associated with an increase in the duration of the dry season (3– pollen sequences indicate a replacement of mature rainforests 5). An abrupt warming of sea surface temperatures (SSTs) in by a forest–savannah mosaic including pioneer trees. Although the Gulf of Guinea was inferred to have changed the monsoon some studies suggested an anthropogenic influence on this for- precipitation (3, 5). Quantitative pollen-based reconstructions est fragmentation, current interpretations based on pollen data of mean annual precipitation of the Lake Barombi record were attribute the “rainforest crisis” to climate change toward a drier, interpreted to reflect a 50% decrease in precipitation during the more seasonal climate. A rigorous test of this hypothesis, how- LHRC (9). Contemporaneous dry conditions were also inferred ever, requires climate proxies independent of vegetation changes. from nearby Lake Ossa using diatoms (10); however, a reanaly- Here we resolve this controversy through a continuous 10,500-y sis of the sediments from this lake demonstrated a ∼400-y age record of both vegetation and hydrological changes from Lake Barombi in Southwest Cameroon based on changes in carbon and Significance hydrogen isotope compositions of plant waxes. δ13C-inferred veg- etation changes confirm a prominent and abrupt appearance of Modern human societies live in strongly altered ecosys- C4 plants in the Lake Barombi catchment, at 2,600 calendar years before AD 1950 (cal y BP), followed by an equally sudden return tems. However, anthropogenic environmental disturbances to rainforest vegetation at 2,020 cal y BP. δD values from the same occurred long before the industrial revolution. About 2,600 y plant wax compounds, however, show no simultaneous hydrolog- ago, a forest–savannah mosaic replaced dense rainforests in ical change. Based on the combination of these data with a com- Western Central Africa. This rainforest crisis was previously prehensive regional archaeological database we provide evidence attributed either to the impact of climate change or, to a that humans triggered the rainforest fragmentation 2,600 y ago. lesser extent, to the expansion of Bantu peoples through Cen- Our findings suggest that technological developments, including tral Africa. A 10,500-y sedimentary record from Lake Barombi, agricultural practices and iron metallurgy, possibly related to the Southwest Cameroon, demonstrates that the rainforest crisis large-scale Bantu expansion, significantly impacted the ecosys- was not associated with any significant hydrological change. tems before the Common Era. Based on a detailed investigation of a regional archaeological database, we present evidence that humans altered the rain- Western Central Africa j late Holocene j rainforest crisis j paleohydrology j forest ecosystem and left detectable traces in the sediments human activity deposited in Lake Barombi. lthough the vast rainforests of Western Central Africa Author contributions: Y.G. designed research; Y.G. coordinated research; Y.G., P.D., G.M., EARTH, ATMOSPHERIC, A(WCA) are not considered pristine ecosystems (1), the B.B., K.G.M., and A.A.A. performed fieldwork; Y.G. conducted sample preparation, chem- AND PLANETARY SCIENCES influence of humans and climate change on their areal extent ical sample treatment, and stable hydrogen isotope analysis; E.S. carried out stable car- bon isotope analysis; Y.G., G.d.S., and R.O. compiled and analyzed archaeological data; and composition during the past remains largely unknown. Over L.M.D. carried out pollen analysis; Y.G., P.D., G.M., G.d.S., E.S., D. Sebag, L.M.D., R.O., B.B., the last three decades, regional pollen sequences recovered from K.G.M., J.-M.O., A.A.A., L.S.E., R.T., M.R.S., A.B., and D. Sachse contributed to the inter- lakes and swamps have provided an unprecedented view of pretation and the discussion of the results; and Y.G. wrote the paper with contributions Holocene vegetation changes in the region. The most striking from all authors. feature is a major vegetation disturbance that occurred between The authors declare no conflict of interest. 3,000 calendar years before present (cal y BP) and 2,000 cal y This article is a PNAS Direct Submission. S.J.F.is a guest editor invited by the Editorial Board. ANTHROPOLOGY BP (2–7). Although this vegetation disturbance appeared to Published under the PNAS license. be widespread across WCA (4), insufficient chronological con- Data deposition: The data reported in this paper have been deposited in the PANGAEA trol and sedimentary hiatuses have limited regional correla- online database (https://doi.pangaea.de/10.1594/PANGAEA.884676). tions or attribution to a single event (8). There is particularly See Commentary on page 3202. strong evidence for the late Holocene rainforest crisis (LHRC) 1 To whom correspondence should be addressed. Email: [email protected]. recorded in the sediments of Lake Barombi (or Barombi Mbo; This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. ◦ 0 ◦ 0 4 39.6 N, 9 24.3 E; Fig. 1). This sedimentary archive has 1073/pnas.1715336115/-/DCSupplemental. recorded a reversible switch from a mature rainforest to a dis- Published online February 26, 2018. www.pnas.org/cgi/doi/10.1073/pnas.1715336115 PNAS j March 27, 2018 j vol. 115 j no. 13 j 3261–3266 Downloaded by guest on October 1, 2021 445 99°23'E°23'E 99°24'E°24'E A Lake Barombi 440 Cameroon -185 5°N JJA insolation (W m LHRC 435 -180 B 430 -175 k 425 o 20 m40 m To b 60 80m m h M 4°40'N4°40'N -170 -2 420 ) (‰) h t Lake -C31 r -60 n o 100 m D 415 f Barombi -165 δ o C S BB1414 -55 -160 -50 -155 44°39'N°39'N -45 (‰) -150 Wetter -40 -C31-corr n 0 500 1,000 m (m asl) D δ Forest reserve boundary 530 -35 Drier Road Farmland/highly 270 -30 Building Elevation degraded forest (m) -29 Overflow sill 0 Catchment boundary -105 -25 D Coring site Depth -30 Fig. 1. Map of the Lake Barombi basin with topography, bathymetry (10-m Degraded/ -31 δ contour intervals), and the coring site. Also highlighted are infrastructures Secondary forest 13 C n and farmland/highly degraded forest based on SPOT 5 images and ArcGIS -C31 -32 Online Basemap–World Imagery (source: Esri, DigitalGlobe, GeoEye, i-cubed, (‰) Undisturbed USDA, USGS, AEX, Getmapping, Aerogrid, IGN, IGP, swisstopo, and the GIS mature rainforest User Community). (Inset) Location map of Lake Barombi (yellow star) in Africa. -33 -34 offset, due to the influence of aged soil carbon, and a dominant 32 control of sedimentation by local forcing factors rather than -35 regional climate, which have cast doubt on this assessment (11). E Although the LHRC is contemporaneous with a marked 31 increase in human settlements and activity in the region (7, 12– 16), it is generally assumed that humans at that time were not 30 yet major agents of vegetation disturbances (4, 13, 15, 17–20). A recent study (21) suggesting that human land-use intensifica- 29 tion associated with the southward migration of Bantu-speaking farmers across Central Africa contributed to forest destruction 28 has, however, fueled a fierce debate in the paleoecology and SST MD03-2707 (°C) archaeology communities (22, 23). To support the notion of a climatic, i.e., hydrological driver 27 of the LHRC, it is necessary to demonstrate independently that vegetation and hydrological changes were coeval, which has not 26 been confirmed to date. Alternatively, to support anthropogenic 0 2,000 4,000 6,000 8,000 10,000 Age (cal y BP) impacts, changes in hydrological and vegetation proxies should be disconnected and unambiguous evidence of coeval human Fig.